34
Review Article QoS Strategies for Wireless Multimedia Sensor Networks in the Context of IoT at the MAC Layer, Application Layer, and Cross-Layer Algorithms Muwonge Ssajjabbi Bernard , 1,2 Tingrui Pei , 1,3 and Kimbugwe Nasser 1,2 1 College of Information Engineering, Xiangtan University, Xiangtan, Hunan 411105, China 2 Department of Networks, College of Computing & I.S, Makerere University, Kampala, Uganda 3 Key Laboratory of Hunan Province for Internet of ings & Information Security, Xiangtan University, Xiangtan, China Correspondence should be addressed to Tingrui Pei; [email protected] Received 24 May 2019; Revised 11 September 2019; Accepted 9 October 2019; Published 29 December 2019 Academic Editor: Zhiyong Xu Copyright © 2019 Muwonge Ssajjabbi Bernard et al. is is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Wireless multimedia sensor networks (WMSNs) have got capacity to collect both scalar sensor data and multidimensional sensor data. It is the basis for the Internet of things (IoT). Quality of service (QoS) pointers like energy efficiency, reliability, bit error rate, and latency can be helpful in data collection estimation over a network. In this paper, we review a number of QoS strategies for WMSNs and wireless sensor networks (WSNs) in the IoTcontext from the perspective of the MAC and application layers as well as the cross-layer paradigm. Considering the MAC layer, since it is responsible for regulating the admittance to the shared medium and transmission reliability and efficiency through error correction in wireless transmissions, and for performance of framing, addressing, and flow control, the MAC protocol design greatly affects energy efficiency. We thus review a number of protocols here including contention-free and contention-based protocols as well as the hybrid of these. is paper also surveys a number of state-of-the-art machine-to-machine, publish/subscribe, and request/response protocols at the application layer. Cross-layer QoS strategies are very vital when it comes to system optimization. Many cross-layer strategies have been reviewed. For these QoS strategies, the challenges and opportunities are reviewed at each of the layers considered. Lastly, the future research directions for QoS strategies are discussed for research and application before concluding this paper. 1. Introduction Wireless sensor networks (WSNs) have a significantly large number of interconnected sensor nodes which can sense their environmental attributes like pressure, light, temper- ature, sound, humidity, and location. ey further cooperate among themselves over wireless transmission media across which they transmit their data as they monitor their envi- ronment [1]. For this to be done more accurately, there is a need for multimedia system support for better information gathering and environmental monitoring. WSNs have thus given birth to a new paradigm shift toward WMSNs. is is inspired by the recent advances in technology that have given rise to portable, cheap multimedia capture, trans- mission, and storage devices such as digital video cameras, microphones, low-cost smart phones, imaging sensors, memory cards, and hard disks. ese technologies are easy to integrate into a node and have made information gath- ering and monitoring of their environment easier and cheaper. Being low-cost smart devices, they have motivated many scholars to undertake research on WMSNs. ese can promptly transmit, store, compare, and combine data from heterogeneous sources. WMSNs are an enhanced kind of WSNs that can sense and/or transmit both scalar and multimedia data including image, audio, and video streams in real-time or non-real- time transmission. ey are networks for wireless embedded devices that can permit the users to retrieve multimedia information from their environment [2]. ey have many applications in surveillance and environmental monitoring Hindawi Journal of Computer Networks and Communications Volume 2019, Article ID 9651915, 33 pages https://doi.org/10.1155/2019/9651915

ReviewArticle - downloads.hindawi.com · ReviewArticle QoSStrategiesforWirelessMultimediaSensorNetworksinthe ContextofIoTattheMACLayer,ApplicationLayer,and Cross-LayerAlgorithms MuwongeSsajjabbiBernard

  • Upload
    others

  • View
    5

  • Download
    0

Embed Size (px)

Citation preview

Page 1: ReviewArticle - downloads.hindawi.com · ReviewArticle QoSStrategiesforWirelessMultimediaSensorNetworksinthe ContextofIoTattheMACLayer,ApplicationLayer,and Cross-LayerAlgorithms MuwongeSsajjabbiBernard

Review ArticleQoS Strategies for Wireless Multimedia Sensor Networks in theContext of IoT at the MAC Layer Application Layer andCross-Layer Algorithms

Muwonge Ssajjabbi Bernard 12 Tingrui Pei 13 and Kimbugwe Nasser12

1College of Information Engineering Xiangtan University Xiangtan Hunan 411105 China2Department of Networks College of Computing amp IS Makerere University Kampala Uganda3Key Laboratory of Hunan Province for Internet of ings amp Information Security Xiangtan University Xiangtan China

Correspondence should be addressed to Tingrui Pei peitingruixtueducn

Received 24 May 2019 Revised 11 September 2019 Accepted 9 October 2019 Published 29 December 2019

Academic Editor Zhiyong Xu

Copyright copy 2019 Muwonge Ssajjabbi Bernard et al -is is an open access article distributed under the Creative CommonsAttribution License which permits unrestricted use distribution and reproduction in anymedium provided the original work isproperly cited

Wireless multimedia sensor networks (WMSNs) have got capacity to collect both scalar sensor data and multidimensional sensordata It is the basis for the Internet of things (IoT) Quality of service (QoS) pointers like energy efficiency reliability bit error rateand latency can be helpful in data collection estimation over a network In this paper we review a number of QoS strategies forWMSNs and wireless sensor networks (WSNs) in the IoTcontext from the perspective of the MAC and application layers as wellas the cross-layer paradigm Considering the MAC layer since it is responsible for regulating the admittance to the sharedmedium and transmission reliability and efficiency through error correction in wireless transmissions and for performance offraming addressing and flow control the MAC protocol design greatly affects energy efficiency We thus review a number ofprotocols here including contention-free and contention-based protocols as well as the hybrid of these -is paper also surveys anumber of state-of-the-art machine-to-machine publishsubscribe and requestresponse protocols at the application layerCross-layer QoS strategies are very vital when it comes to system optimization Many cross-layer strategies have been reviewedFor these QoS strategies the challenges and opportunities are reviewed at each of the layers considered Lastly the future researchdirections for QoS strategies are discussed for research and application before concluding this paper

1 Introduction

Wireless sensor networks (WSNs) have a significantly largenumber of interconnected sensor nodes which can sensetheir environmental attributes like pressure light temper-ature sound humidity and location-ey further cooperateamong themselves over wireless transmission media acrosswhich they transmit their data as they monitor their envi-ronment [1] For this to be done more accurately there is aneed for multimedia system support for better informationgathering and environmental monitoring WSNs have thusgiven birth to a new paradigm shift toward WMSNs -is isinspired by the recent advances in technology that havegiven rise to portable cheap multimedia capture trans-mission and storage devices such as digital video cameras

microphones low-cost smart phones imaging sensorsmemory cards and hard disks -ese technologies are easyto integrate into a node and have made information gath-ering and monitoring of their environment easier andcheaper Being low-cost smart devices they have motivatedmany scholars to undertake research on WMSNs -ese canpromptly transmit store compare and combine data fromheterogeneous sources

WMSNs are an enhanced kind of WSNs that can senseandor transmit both scalar and multimedia data includingimage audio and video streams in real-time or non-real-time transmission-ey are networks for wireless embeddeddevices that can permit the users to retrieve multimediainformation from their environment [2] -ey have manyapplications in surveillance and environmental monitoring

HindawiJournal of Computer Networks and CommunicationsVolume 2019 Article ID 9651915 33 pageshttpsdoiorg10115520199651915

systems traffic monitoring target tracking intrusion de-tection systems telemedicine for advanced health care etcIqbal et al for instance developed an efficient power al-location and personal wireless hub (PWH) placementstrategy for maximizing the data rate under the cognitiveradio interference constraint that has an efficacy with lowcomplexity to facilitate paramedic staff in next-generationhealth care facilities using multimedia in smart hospitals [3]

11 Classification of WSNs (Wireless Sensor Networks)Different types of WSNs include terrestrial WSNs un-derground WSNs underwater WSNs multimedia WSNsand mobile WSNs

111 Terrestrial WSNs -ese can communicate with basestations efficiently and comprise 100 s to 1000 s of wirelesssensor nodes deployed in the unstructured (ad hoc) orstructured (preplanned) style In the ad hoc mode nodes arerandomly distributed in the target region that is droppedfrom a fixed plane -e structured mode considers optimalplacement grid placement and 2D and 3D placementmodels Energy is conserved by use of low duty cycling delayminimization and optimal routing

112 Underground WSNs -ese are more expensive thanterrestrial WSNs in terms of deployment maintenanceequipment cost and careful planning -e WSNs comprisemany sensor nodes hidden in the ground to monitor un-derground conditions To relay information from the sensornodes to the base station additional sink nodes are locatedabove the ground -ese are highly affected by attenuationand signal loss and are very difficult to charge

113 Underwater WSNs -ese have multiple sensor nodesand vehicles deployed under water Autonomous un-derwater vehicles are used to gather data from the sensorLong propagation delay and bandwidth and sensor failuresare a major challenge here

114 Multimedia WSNs (WMSNs) -ese have been pro-posed to enable tracking and monitoring of events in theform of multimedia such as imaging video and audio-esenetworks comprise low-cost sensor nodes equipped withmicrophones and cameras -e nodes are interconnectedwith each other over a wireless connection for data com-pression retrieval and correlation -e challenges withWMSNs include high energy consumption data processingand compressing techniques and high bandwidth re-quirements for proper and easy content delivery

115 Mobile WSNs -ey comprise a collection of sensornodes which can move on their own and can interact with thephysical environment Mobile nodes have the ability to com-pute sense and communicate -e mobile WSNs are moreversatile than static sensor networks -e advantages ofWMSNs over the static WSNs include better and improved

coverage better energy efficiency and superior channelcapacity

12 e Architecture of a WMSN -e WMSN architecturehas the following 3 subdivisions

(i) Single-tier flat architecture this consists of homo-geneousmultimedia nodes which are able to executeany function to the sink via multihop routes

(ii) Single-tier clustered architecture this consists ofheterogeneous nodes passing sensed information tothe cluster head for processing

(iii) Multitier architecture this too consists of hetero-geneous nodes and does object sensing and targetcapturing and tracking

Figure 1 shows a typical example of a WMSN archi-tecture Figure 1(a) shows a single-tier flat and homogeneousarchitecture in which sensors having the same physicalabilities are utilized In Figure 1(b) we have a single-tierclustered and heterogeneous architecture having nodes withdifferent physical capabilities eg multimedia nodes andscalar nodes -e multitier clustered and heterogeneousarchitecture (Figure 1(c)) has several layers of nodes havingdiverse types and processing tasks per layer [2] WMSNs arewidely deployed to offer infrastructural support and sensoraccessibility rendering them suitable for Internet of multi-media things (IoMT) transmission [4] Most applications(apps) in the IoMT eg wearable devices utilize theWMSNtechnology

13e Internet of ings (IoT) In the IoTtoday we visualizea situation whereby smart devices connect to a single net-workmdashthe Internet It is becoming more popular than it hasbeen because of the multitude of connected devices availablerecently A number of IoT- and IoMT-based applications arecoming lately while attracting enormous attention

-ese applications include smart cities smart vehicleshomes factories (Figure 2) and GPS tracking devices [5 6]and have attracted many inventions in the Americas AsiaEurope etc A number of commercial and military appli-cations come up because of introduction of multimediaobjects in transmission of data such as remote patientmonitoring in telehealth and telemedicine and trafficmanagement systems enhanced by smart video camerasamong others [4] -is calls for an upgrade in functionalityof IoTsystems to the IoMT Figure 2 shows an illustration ofthe IoT architecture which could be aided by RFID opticaltags QR codes Bluetooth low energy Wi-Fi direct andLTE-Advanced among others Most of the scholars em-phasize improvement of efficiency for handling a lot of real-time information (info) but ignore multimedia transmissionaspects [7 8] -e direction of research is shifting from theordinary IoT to the multimedia-based IoT because of theneed to enable smart devices to efficiently observe sense andunderstand their environment through multimedia data[9 10] hence resulting in the emergence of the newer field ofInternet of multimedia things (IoMT)

2 Journal of Computer Networks and Communications

IoMT is ldquothe IoT-based paradigm that enables objects toconnect and exchange structured and unstructured data withone another to enable multimedia-based services and appli-cationsrdquo [3] ere is a need for vast processing powermemory and bandwidth for high QoS when transmittingmultimedia data in the IoMT in comparison with scalar datain a traditional IoT IoT system functionality shouldtherefore be upgraded to the IoMT We compare the two asdiscussed in [7]

(i) e IoT has standardized communication pro-tocols whereas the IoMTrsquos protocols arenonstandardized

(ii) In terms of QoS the IoT requires low bandwidthwhereas the IoMT requires higher bandwidth

(iii) e IoMT transmits heterogeneous multimediadata whereas IoT data transmitted have limitedheterogeneity

(iv) IoT sensor nodes consume less energy than IoMTsensor nodes

(v) IoT devices are deployed in application-dependentRFID tags but the IoMT is in video and audiosensors

(vi) In terms of service composition the IoMT has noavailable specialized middleware whereas the IoThas specialized service-oriented architecture-basedand event-based middleware

For improved QoS best eort services and higher en-ergy eciency in IoMT networks and applications enor-mous multimedia-supported routing is gaining ground inthe research arena in the WMSN area in routing protocolsalgorithms and techniques based on network architecturesand application requirements [11] Nevertheless accordingto Ahmad et al [12] the enormous resource-constrainedheterogeneous environment of the IoT challenges its ex-pansion and deployment is is because most existing IoTapps comprise overlaid deployments of wireless sensor andactuator networks in which apps cannot interact with eachother or share and reuse the few available resources Inaddition to that ecient sensing and propagation of info

C

Users

WMSN

Image sensor

CM

CM

Audio sensor

Scalar sensor

CC

Internet

Laptops

Laptop

Wireless gateways

Storage

Sink

Scalar sensor

Video sensor

Figure 1 Architecture of the WMSN (a) single-tier at and homogeneous (b) single-tier clustered and heterogeneous (c) multitierheterogeneous

Journal of Computer Networks and Communications 3

and swift response to changes in the physical world arechallenging IoT requirements [12]

14 What Has Been Done In wireless communicationsystems routing is one issue that is quite a challenge yetthere are very few surveys available especially regardingWMSN protocols [1] and more so the MAC layer appli-cation layer and cross-layer protocols although some sur-veys have been published on WSNs To the best of ourknowledge no survey combines all these aspects -is makesour survey important to add to the existing literature -eauthors in [1] discuss a number of WMSN routing tech-niques and the properties and shortcomings of theseHowever they do not discuss any application layer tech-niques Muzakkari et al [13] survey some recent WSNcontention-based scheduling-based and hybrid MACprotocols whereby they focus on the underlying principlesadvantages limitations and their applications But much asthey only focus on MAC protocols they still leave out anumber of them in their survey Yigitel et al [14] carry out asurvey on QoS-aware MAC protocols for WSNs -ey re-view the QoS challenges and views for WSNs study the QoSmechanisms categorize the state-of-the-art QoS-awareMAC protocols and also talk about the advantages anddrawbacks of the same But being a 2011 release a number ofkey protocols developed since then to date are not presented

therein Abbas and Kure [15] review various methods forQoS provisioning at the levels of routing MAC layer andcross-layer including the schemes for admission control andscheduling for QoS provisioning as well as the problems andchallenges involved But they leave out the protocols from2010 to date Shatnawi [16] and Karagiannis et al [17] reviewthe application layer protocols for the Internet of things Intheir survey they address a number of application layerprotocols which are employed for IoT for affirming a re-liable tie among objects and things -ey evaluate thereviewed protocols in terms of architecture communicationmodel security and QoS as well as the weaknesses andstrengths per reviewed protocol However they only con-centrate on application layer protocols and still leave outsome critical protocols In a recent survey Bernard et al [18]discuss a number of cross-layer QoS strategies for IoT Butbecause of space constraint we leave out some criticalprotocols and besides this paper does not talk about ap-plication layer and MAC layer protocols AlAmri andAbdullah [19] carry out a survey on cross-layer QoS pro-tocols for WMSNs in which they state that the cross-layerarchitecture is a novel idea that brings together a number oflayers for enabling integration and exchange of informationin between them with higher efficiency compared to thetraditional layered model -ey discuss and compare theavailable cross-layer WMSN protocols that cross the uses ofadjacent or nonadjacent layers But they concentrate on

Database

MMS user databases

MMS relayserver

Home location register

External servers(eg e-mail and fax)

MMS value-added servicesservers (premium content

e-mails web )

Foreign MMSC(MMS relay server)

MM

S us

er ag

ents

Internet

Figure 2 Example of the multimedia service architecture in the context of IoT

4 Journal of Computer Networks and Communications

cross-layer protocols and leave out the MAC layer andapplication layer protocols and also some cross-layer pro-tocols are left out -e surveys that have been reviewed arecompared in Table 1

-emajor contributions that make our survey importantare outlined below

15 Importance of the Survey

(i) We discuss and draw a comparison between the IoTand the IoMT by discussing the differences betweenthe two

(ii) We further discuss the newer paradigm of WMSNsand compareWSNs andWMSNs in our discussion

(iii) We make a comprehensive survey of the recentstate-of-the-art routing protocols focusing on theMAC layer the application layer and the cross-layer paradigm

(iv) We discuss the WMSN protocols as well as theWMSN-enabled protocols that have been de-veloped in the recent studies by several scholars

(v) In some of the reviewed protocols we talk about therelative advantages and disadvantages of some ofthe routing approaches for readers to be in positionto comprehend different techniques and so choosethe most appropriate technique depending on theuser requirements

(vi) In all these we cover the challenges and oppor-tunities existing for the discussed categories

(vii) We consider the possible future research trendsbefore concluding this paper

16 Challenges

(i) Since WSNs and WMSNs are usually distributed andad hoc networks their nodes are powered by batteriesmaking energy optimization a challenge that needs tobe addressed when designing their routing protocols

(ii) Considering WSNs their routing protocols do notput into consideration multimedia applicationswhich need a lot of bandwidth and processingenergy and must be transmitted in real time withutmost fidelity Extra challenges in routing areimposed in WMSNs in deployment of heteroge-neous sensors since there could be audio video andstill pictures in addition to scalar data duringtransmission Besides the heterogeneous datathereby handled all have different QoS re-quirements and with many different businessneeds different services with varying requirementsgive rise to a big challenge for the routing design

(iii) WMSNs have also got nonrechargeable batteriesplacing a limitation on the energy of the nodes yetmultimedia applications consume a lot of energyQuite a number of complications and limitingfactors exist practically despite the fact that sensorscan be self-powering as enabled by energy

harvesting Predictive video encoding using MPEG-4 and other standards is also an energy-consumingprocess and so causes enormous degradation inrouting efficiency complicating the design ofWMSN protocols even more

(iv) -e heterogeneous nature of the network is anotherchallenge since there is a difference in the categoriesof nodes needed for communication so as to enablethe facilitation of effective data collection andprocessing as well as efficient transmission It is noteasy to have a uniform communication protocolplatform because of different functionalities unlikethe case of the conventional WSNs compared to theWMSNs

(v) -ere must be a trade-off between the energy effi-ciency and the multimedia QoS when deciding onthe route taken Data aggregation and compressionrouting protocols may be applicable for energy-saving though they can lead to intolerable delays inWMSNs Furthermore there might be networkcongestion due to the many-to-many and mutualinterference in wireless routes and the scarceWMSN resources In case a given node gets so manyhigh-rate streams it lowers the performance of thenetwork and raises the possibility of node failuredue to energy drainage

-is papermainly discusses QoS strategies forWSNs andWMSNs in the context of IoT from the MAC layer appli-cation layer and cross-layer perspective -is paper presentsa state-of-the-art survey on routing in WSNs and WMSNs-e rest of this paper is organized as follows In Section 2 wediscuss the QoS strategies at the MAC layer In Section 3 wediscuss the QoS strategies at the application layer Section 4discusses the cross-layer QoS strategies and in Section 5 wegive some future research directions before concluding thispaper in Section 6

2 QoS Strategies at the MAC Layer

-e medium access control (MAC) protocol is meant toregulate admittance to the shared medium as well astransmission reliability and efficiency through error cor-rection in wireless transmissions among others It is alsoresponsible for performance of framing addressing andflow control -e MAC protocol design affects energy effi-ciency too -is is majorly categorized into contention-freeprotocols contention-based protocols and a hybrid of these

21 MAC Protocols forWSNs Hybrid MAC protocols thesecombine contention-free and contention-based MAC pro-tocols on whose advantages they capitalize Some of these arediscussed below

211 S-MAC (Sensor MAC) -is is a MAC protocoldesigned for WSNs -e main components of S-MAC in-clude (i) periodic listen and sleep (ii) collision and over-hearing avoidance and (iii) message passing Energy

Journal of Computer Networks and Communications 5

consumption and self-configuration are controlled usingthese three techniques Nodes periodically go into the sleepmode to control the energy consumed in listening to idlechannels and virtual clusters are formed by neighbor nodesfor sleep schedule autosynchronization -e protocol em-ploys in-channel signaling when the radio is in the sleepmode and message passing is applied to decrease contentionlatency for store-and-forwardWSN applications during datatransmission [22] For adjacent nodes to solve their clockdrift S-MAC requires periodic synchronization To reducesynchronization errors exchanged timestamps are relativenot absolute

Advantages

(i) S-MAC has a longer listening period than the clockerror

(ii) It has more capacity to conserve energy than IEEE80211

(iii) It can do trade-off between energy and latencydepending on traffic conditions [22]

212 B-MAC (Berkeley MAC) -is is a carrier-sense MACprotocol designed for WSNs featuring a simplistic scalableimplementation that does network variations and is effectivein channel valuation [23] -e protocol has an interface withsome flexibility leading to very low-power usage and con-trols collisions through collision avoidance mechanismsleading to effective use of the frequency band It uses asampling scheme that is adjustable to minimize duty cyclingand idle listening so as to reduce power wastage Nodes arescheduled with their duty cycle each within which they sendthe preamble to the channel in case of availability of data fortransmission In different duty-cycle schedules channelusage is checked by the destination nodes that is if thepreamble found is long it will remain powered on until thedestination address is obtained and in case it is the target itwill either await senderrsquos data or go to sleep-e protocol haslower latency than synchronous MAC protocols but has aproblem of overhearing where the neighbor node getslengthy preambles though not the intended receiver hencewasting a lot of power in doing so [24] It outperforms anumber of protocols via reconfiguration feedback andbidirectional interfaces for upper layer services It can run at

low duty cycles and its apps cannot experience synchro-nization overhead [23]

Advantages

(i) -e protocol features a simplistic scalableimplementation that does network variations andis effective in channel valuation

(ii) It has an interface with flexibility leading to verylow-power usage

(iii) It controls collisions through collision avoidancemechanisms leading to effective use of the fre-quency band

(iv) It uses an adjustable sampling scheme to minimizeduty cycling and idle listening reducing powerwastage

Disadvantages

(i) It has a problem of overhearing where the neighbornode gets lengthy preambles though not theintended receiver hence wasting a lot of power indoing so

213 T-MAC -is is a contention-based MAC protocol forWSNs that dynamically with little difficulty via fine-grainedtimeouts adapts a listensleep duty cycle It uses this dutycycle by ending its active part so as to vary load in time andposition thereby decreasing the energy wastage on idlelistening [25] When starting an active period it uses a shortlistening window thereby improving the use of energy byS-MAC -rough adaptive duty cycling T-MAC savespower but at a cost of less throughput and enhanced delayBoth S-MAC and T-MAC under similar conditions performin the same way though in variable workloads S-MAC uses 5times more power than T-MAC But T-MAC is complex andnot scalable If its active window is decreased its snoopingcapability on nearby traffic and adaptation to variableconditions of the network is reduced [23] All nodes in thisprotocol are configured to wake up periodically listen toadjacent nodes and go back into the sleep mode till the nextframe as more messages come into the queue Collisionavoidance and reliable communication are ensured throughRequest-to-Send (RTS) Clear-to-Send (CTS) and Ac-knowledgment (ACK) messages among nodes [25]

Table 1 Comparison of this survey with existing surveys

Author(s) Year CharacteristicsBhandary et al [1] 2016 Routing strategies in WMSNs properties and limitationsMuzakkari et al [13] 2018 WSN contention-based scheduling-based and hybridMAC protocolsAhmad et al [12] 2011 QoS-aware MAC protocols for WSNsMuzakkari et al [13] 2010 QoS provisioning at the levels of routing MAC layer and cross-layerShatnawi [14] 2016 IoT application layer protocolsKaragiannis et al [15] 2015 Application layer protocols for the Internet of thingsBernard et al [16] 2019 Cross-layer QoS strategies for the IoT in WMSNsAlAmri and Abdullah [17] 2017 Cross-layer QoS protocols for WMSNsGuo et al [20] 2012 Cross-layer and multipath-based video transmission for WMSNs

Ksentini et al [21] 2006 Toward an improvement of H264 video transmission over IEEE80211e through a cross-layer architecture for WMSNs

6 Journal of Computer Networks and Communications

T-MAC design the major forms of energy consump-tion in T-MAC are through (i) idle listening (ii) col-lisions (iii) protocol overhead and (iv) overhearing Incontrast in the active period the node remains lis-tening and sending data If no activation event occursfor a certain time t then the active period ends -eactivation event might be data reception on a radiocommunication sensing on a radio and end oftransmission for the data packet of the node or ACKknowing that the neighbor has finished transmitting

Clustering and synchronization on waking up a nodebegins by listening to the channel If after a certain timethere is nothing heard it chooses a schedule fortransmitting Synchronization (SYNC) packetsappended with the time of the next frame startup Ifduring this time it hears another nodersquos SYNC packetit follows its schedule subsequently transmitting itsown SYNC Following both schedules enables it to havean activation incident at the beginning of both frames

RTS and t choice if a node does not get a CTS messageafter sending an RTS then either (i) there is a collisionpreventing the destination node from hearing the RTS(ii) RTS or CTS is overheard and destination node barredfrom responding or (iii) the destination node is in thesleep mode So if after time t there is no answer thesource node switches to the sleepmode It should only doso if none of its neighbors is still communicating A newinterval t can be triggered by a neighborrsquos RTS or CTSreception A node may be out of range to hear theinitiating RTS-e length t of the interval should be longenough to get the CTS packetrsquos beginning [25] and t isthus given by the comparison in the following equation

tgtC + R + T (1)

where C contention interval length RRTS packet lengthand T turn-around time

Advantages

(i) It saves power through adaptive duty cycling

Disadvantages

(i) It is unscalable and more complex(ii) It offers less throughput and enhanced delay

214 CU-MAC -is protocol supports IoT standards re-quiring request-response transmissions It employs a multi-channel mechanism for constant bidirectional transmissionof packets at low duty cycle to resolve hidden-terminal issuesAccording to Danmanee et al [24] CU-MAC is an SI-MACprotocol that is aimed at improving bandwidth of the channelby initiating connections and transmitting data throughdifferent channels using the multichannel approach

Protocol design the protocol has the following threeproperties that make it perform more efficiently withWSNs in the IoT

(i) It enables multichannel transmission using 9useable channels in IEEE 802154

(ii) It can enable linked data transmission using oneconnection which reduces additional overheadand delay

(iii) It can support continuous transmission of packets-is is enabled depending on how the bufferoverflow protection policy is set in the buffer stack-is stack has 2 parts (a) bidirectional buffer slotsfor enabling sensor nodes to support linked datatransmission and (b) normal buffer slots fornormal data queuing Buffer stack policies are setas follows

(a) A packet is only dropped in the buffer when itexpires

(b) Unless the packet is from a target node to which areceiver is sending the data the receiver deniespacket reception when buffer slots fill up

(c) -e bidirectional buffer slot is used for bi-directional buffer transfer

CU-MAC can reduce power wastage by choosing toemploy only a single channel for transmitting both controlpackets and data in case there is only one sender in the sameregion of coverage It can further enable multiple trans-missions through another usable channel if there are othersending nodes in this area [24] -e protocol has two phases(i) advertising phase in which a source node announces toneighbors before transmission and (ii) data transfer phasewhich commences when the source node receives a Ready-to-Receive Information packet from the destination node

Advantages

(i) It has a higher packet delivery ratio compared toother MAC protocols

(ii) It performs well in high traffic situations since itsduty-cycle is lower

(iii) It reduces power wastage by employing only asingle channel for transmitting both controlpackets and data

(iv) It can enable multiple transmissions through an-other usable channel if there are other sendingnodes

215 R-MAC -is is a reservation-based MAC protocoldesigned majorly with an aim to be fair and efficient in termsof energy use To eliminate collision of data packets R-MAC[26] schedules control and data packet transmissions at thesource and destination nodes rather than using RTSCTSmessage exchanges Channel utilization in R-MAC is en-hanced by an ARQ method called burst-based acknowl-edgment which together with other scheduling algorithmsaddresses the exposed terminal problem besides enhancingthe network throughput

R-MAC design all nodes in this protocol intermittentlyfunction in listen and sleep states so as to avoid wastingenergy in overhearing and idle modes -e time takento listensleep is the same for all nodes each of which

Journal of Computer Networks and Communications 7

chooses a schedule for itself rendering central sched-uling and synchronization irrelevant So in the absenceof traffic in the locality the node simply does periodiclistening and sleeping and to eliminate collisionstransmissions are distributively synchronized using areservation-based model in case there is a sender totransmit data R-MAC differs from S-MAC in that ittransmits data to a number of nodes using the sleepmode [24] -ere are 3 stages in this protocol (i) la-tency detection (ii) period announcement and (iii)periodic operation of which (i) and (ii) are for nodesynchronization and (iii) is for listensleep procedures

(i) Latency detection phase here all nodes broadcast aNeighbor Discovery control packet (ND) at an ar-bitrarily chosen time since they are powered onWhen the ND is received from the neighbor nodethe receiver notes its arrival time and chooses arandom time to send an acknowledgment (ACK-ND)mdashthe same as the receivedND in sizemdashinwhichthe duration from ND arrival to ACK-ND trans-mission I2 is specified and the time from ND packettransmission to ACK-ND arrival I1 is computed

(a) Propagation latency between the 2 nodesL (I1 minus I2)2

(b) -us propagation latency can be defined as ldquotheinterval from the time the first node sends the firstbit of a packet to the time the second node receivesthe last bit of the packetrdquo [24]

(ii) Period announcement phase here nodes choosetheir own random listensleep time and start timeand broadcast these and when packets are re-ceived the node turns the received schedule intoits own All nodes will record their neighborsrsquoschedules in relation to theirs -ese two phasesmake a number of rounds to ensure all nodes areinformed about their neighboring nodes

(iii) Periodic operation phase here there is datatransmission [26] and nodes intermittently go intothe sleep mode and wake up One period is a listensleep cycle and nodes possess similar periods Twoparts make a period T0 listen window TL and sleepwindow TS as shown in the following equation

T0 TL + TS (2)

In this protocol the nodes make use of control packetsfor communication-ese include REV (reservation packet)ACKREV (acknowledgment packet for REV) and DATAACK-DATA (acknowledgment packet for data) messageexchanges all of which are similar in size

Advantages

(i) It is fair and efficient in terms of energy use(ii) It eliminates collision of data packets

216 AR-MAC (Adaptive-Reliable Media Access Control)-is is a TDMA-based MAC protocol for wireless body areanetworks (WBANs) designed to reduce energy consumption

as proposed by Rahim et al [27] -e protocol allocates aguaranteed time slot (GTS) to all nodes for transmissiondepending on their needs It utilizes periodic sleep andwakeup to minimize overhearing and idle listening based onthe needs of the node It has a central node (CN) having bigbatteries with higher computational power and single ordouble transceivers For double transceivers the sum of thetime frame TFrame is assigned for node transmission But inAR-MAC a single transceiver CN is considered with TFramehaving 3 subdivisions contention-free period (CFP) forsensor connection contention access period (CAP) foremergencies and time TMS to communicate with themonitoring station (MS)

Channel selection the CN first scans for redundantchannels from which it chooses one for transmissionand broadcasts its address and other information to thenodes -e destination nodes search for channels byscanning the radio frequency (RF) channels If free itswitches to the next and if busy it will wait for time TCPlistening to packets and again switch to the subsequentchannel if it does not get the channel packets When thenode finally succeeds in getting a channel packettransmission commences and an ACK packet is sent tothe CNTime slot assignment a Time Slot Request (TSR) packetis sent to the CN by the sensor node after it has selectedthe RF channel -e TSR packet has data rate and timeslot info for the node Static-size time slots with staticguard-band time TGB are proposed in [28] -isprotocol employs the adaptive algorithm of time slot(TS) and GB time and the CN allocates TS whilesending Time Slot Request Reply (TSRR) based on thenode traffic pattern -e size of the time slots variesdepending on the node requirements Based on thetransmissionmodel the data packet transmission ACKpacket reception and some amount of delay can betolerated by allocated TS To prevent interference be-tween adjacent time slots resulting from the node andCN clock drift TGB is inserted between them and itsvalue is given as follows

TGB1

FxTS1100

TGBn

FxTSn

100

TGBnn+1

F

100X12

TSn + TSn+11113858 1113859

(3)

where FGB factor depending on the mean driftvalue Nodes go to sleep after slots have been suc-cessfully assigned and wake up to transmit in assignedslots thus saving energy wasted in idle listening sinceeven allocated slots are collision freeSynchronization for periodic synchronization TDMAneeds additional energy [29] It needs a lot of energy tosynchronize nodes after several cycles To reduce en-ergy wastage and control collisions AR-MAC employs

8 Journal of Computer Networks and Communications

a new synchronization scheme When a data packetarrives the current and expected packet arrival timeswith allowable delay (D) are compared by the CN and adrift value (DV) is computed based on the differenceΔT -e DV is sent to the node in the ACK packet forchanging the slot in subsequent transmissions thoughthe value is based on allowable delay and F In caseΔTgtD the CN will send the DV in the SYNC-ACKpacket for subsequent SYNC to nodes or else a mereACK packet is sent by the CN for the received packetFor any forthcoming data transmissions the node willchange its wake-up time plan based on the DV Withsuch a sync approach a sleeping node will not lose syncfor all cycles N Acceptable delay (D) is given as

D min TS1 TSn( 1113857XF

100 (4)

where F is the guard-band factor

ΔT expected arrival time minus current arrival time

DV 0 if |ΔT|ltD

ΔT if |ΔT|gtD1113896

(5)

Frame format the AR-MAC protocol employs twoforms of packets namely (i) data packets in which anode transmits in its assigned time slot and employsCAP for emergency data and (ii) control packets thatinclude channel packet Time Slot Request (TSR)packet Time Slot Request Reply (TSRR) packet Syn-chronization-Acknowledgment (SYNC-ACK) packetData Request (DR) packet and Acknowledgment(ACK) packet [27]Energy consumption if N is the number of cyclesenergy consumption for these is measured as follows

ETotal 1113944N

k1ESleepk

+ 1113944N

k1EActivek

ESleep TSleep x ISleep x V

TSleep Tframe minus Tactive

(6)

where ISleep current from the voltage source V whenthe node is asleep If Esw is the switching energy Etransthe transmission energy Erec the receiving energy andETout the time-out energy then

EActive 2Esw + Etrans + Erec + ETout (7)

Advantages

(i) It reduces energy consumption(ii) It minimizes overhearing and idle listening based

on the needs of the node via periodic sleep andwakeup

217 ER-MAC It is a combination of TDMA and CSMAmethods making it flexible and adaptable to topological andtraffic changes and hence applicable in rapid response

situations inWSNs Collision-free slots can be planned usingTDMA [30] ER-MAC is designed in such a way that thefollowing are met

(a) It has a higher ratio of packet delivery and low delayunder lower energy consumption levels out-performing Z-MAC

(b) It keeps two priority queues in order to distinguishhigh from least priority packets

(c) It permits contention in TDMA slots so as to handleenormous traffic volumes

(d) It has a harmonized slot shape and the nodes areable to locally change their plans enabling themconnect to or exit the network with easeProtocol design it discovers the topology in such away that it uses the simple flooding mechanism tobuild the tree as initiated by the base station (BS) -etopology discovery aims at neighbor discovery andchange tracking in addition to setting up a routingtree A TOPOLOGY_DISCOVERY message com-prising hop_count new_parent_id and old_-parent_id is generated by the BS A node broadcasts itto discover if it has any potential children and aresponse to the parent In case of the desire to changethe parent the former parent is notified For all nodesthe hop count to the BS parent ID children and one-hop neighbor are registered with the BS at this stageTDMA slot assignment when a BS sends a SYNCmessage it switches to the TDMAmode which whenreceived by the child uses the broadcast slot for itschildrenrsquos synchronization Neighboring nodescannot share a slot since slots are assigned andschedules are exchanged accordingly Slot assign-ment begins with a leaf node in a bottom-up modesuch that the flow of messages to the BS may besmooth Before getting the child-node schedules anonleaf node will not transmit its data via a unicastslot descendantrsquos data via many unicast slots andchild synchronization via a broadcast slot Childnodes send SCHEDULE_NOTIFICATION mes-sages to the BS to end this phase

Local time synchronization is done via parent-childbroadcast sync since they both need to operate the same clockso that when one is transmitting the other is ready to receivePacket prioritization happens in such a way that a highpriority queue is first emptied before a lower priority queue isallowed to transmit In case a queue is filled up the shortestslack packet is dropped and then fairness over source isconsidered as queue modification is done to enable the BSbalance info from different nodes Energy conservation is bysender turning off the radio in absence of any info to transmitand the receiver will switch to the sleep mode in case nopackets are received after a certain timeout [30]

Advantages

(i) It is flexible and adaptable to topological and trafficchanges

Journal of Computer Networks and Communications 9

(ii) It is applicable in rapid response situations inWSNs

(iii) It has a higher ratio of packet delivery and lowdelay under lower energy consumption levels

218 RI-MAC (Receiver-Initiated MAC) -is is an asyn-chronous duty-cycle protocol that tries to reduce timeoccupied by the communicating nodes on the medium toagree on time for data exchange-e sending node waits forthe receiving node to signify the beginning of the com-munication session through a short beacon frame that itsends -e channel is only occupied by the beacon and datatransmission reducing channel occupancy and thus en-abling data exchange for other nodes too -is implies animprovement in channel utilization that subsequently leadsto increment in throughput packet delivery ratio andpower efficiency over a wide-ranging traffic load [31] Beinga receiver-initiated model it ensures substantial reductionin overhearing and lowers the probability of colliding andthe cost of recovery compared to B-MAC [23] or X-MAC[32] RI-MAC is further capable of using a beacon withACK and Ready-to-Receive packets though it has aproblem if a sending node with Ready packets needs to havethe radio on till a ready-to-receive node is awake [13] whichthen sends it a beacon and transmission begins immediatelyand later is acknowledged by another beacon Nodes in-termittently wake up to listen to any possible frames meantfor them depending on their schedules A node announcesits readiness to receive a frame through a broadcast beaconon waking up when it switches on the radio When a dataframe has been received the receiver will spend some moretime known as ldquodwell timerdquo in the active state so as toreceive queued packets which is the time in RI-MAC thatdepends on the number of contending senders If there areseveral contending senders that might not have beenpredicted there is a challenge in trying to reduce the re-ceiverrsquos active time to optimize power efficiency and reducethe cost of collision detection and lost data recovery even inthe case of hidden senders For this protocol this can beaddressed by a beacon frame from the receiving node forcoordinating data frame transmissions of contendingsending nodes-e protocol also beats B-MAC and X-MACin minimizing the cost of collision detection and datarecovery through coordination of the receiversrsquo dataframes -is is because the receiving node is aware of themaximum delay prior to the arrival of the frame-e abilityto broadcast is supported by RI-MAC when it sends data asunicast to each senderrsquos neighboring node or continuouslytransmits the frame in a back-to-back mode for as long asthe sleep interval [31]

Advantages

(i) It improves channel utilization leading to in-crement in throughput packet delivery ratio andpower efficiency over a wide-ranging traffic load

(ii) It ensures substantial reduction in overhearing andlowers the probability of colliding and the cost ofrecovery

219 SRI-MAC (Synchronous Receiver-Initiated MAC)-is is a synchronous duty-cycle protocol that implementsthe method of receiver-initiated data transmission with anaim of enhancing network lifetime through energy wastageavoidance via collision overhearing and idle listening To dothis and also minimize duty cycle it uses a sequence ofadaptive beacons and RTSCTS packets In this protocolsending nodes do not go to sleep till they have got CTSpackets from receiving nodes to begin transmitting exceptfor nonsending nodes that return into the sleep mode rightaway as if the channel was idle [33] Since sending nodeshave separate schedules it reduces the contention proba-bility leading to more energy conservation A time frame isdivided into the information period the allocation periodand the communication period which are discussed asfollows

Information period at frame initiation a beacon is sentby the receiving node declaring its return from sleepand readiness to accept data packets in case the channelis not busy -e beacon is a small packet alerting otherlistening nodes of the transmitterrsquos readiness to get arequest for channel It consists of the receiverrsquos ID andthe duration allocation period (DAP) that relies onnumber of the receiving nodersquos neighbors [13]Allocation period here the receiving node stays awake forsome little time after broadcasting its beacon to confirm ifthere is any node having anRTS packet to transmitWhenthe beacon has been received an RTS packet (with 3fieldsmdashsender ID receiver ID and data size) is sent by thesender prompting the receiver to send a CTS for SYNC asthere is neighbor information exchangeCommunication period SRI-MAC is capable of or-ganizing and scheduling transmissions by means ofsensor node IDs -e receiver assigns order andtransmitting time for every sender such that they switchoff their radios till their time to transmit is dueHowever the receiving node remains in the wake-upmode to get all sending nodersquos data SRI-MAC canenable sensors have lengthy sleeping times duringtransmission such that energy can be saved as long as itis neither sending nor receiving data -e protocolrsquoscommunication period is premised on the TDMAtechnique which conserves energy since the radiorsquosduty cycle is lowered and collision is eliminated [33]

Advantages

(i) Since it employs the TDMA principle it eliminatescollisions while maintaining energy conservationtoo

(ii) Noncommunicating nodes switch off their radioand get into the sleep mode which eliminatesoverhearing

(iii) Since receiving nodes broadcast SYNC signals toall nodes in the form of CTS packets overemittingis avoided

(iv) It decreases duty cycling as sleep time increasesand lowers idle listening

10 Journal of Computer Networks and Communications

2110 H-MAC -is protocol is built on the power savingmechanism (PSM) of IEEE 80211 and slotted Aloha pro-tocols to enhance performance using multiple slots Timehere comprises large frames each having an active and asleeping part [34] If any node has got data to send it will askfor slots from the receiver node during active time and latertransmit the packets in the prenegotiated slots Nodes gointo the sleep mode during sleep-time slots if they do nothave data to transmit

Latency in IEEE 80211 if N is the number of hops nthe value for the current hop tcsn the back-off time andttx the transmission delay the total latency is calculatedusing the following equation

D(N) 1113944N

n1tcsn + ttx1113872 1113873 (8)

and the average latency is given by the followingequation

E[D(N)] N tcsn + ttx1113872 1113873 (9)

-roughput in H-MAC is given by the followingequation

Th npnm

tact + Cts (10)

where np is the number of transmitted packets nm is themaximum number of nodes connected in a time frameC is the number of slots with the same length tact is thelistening time and ts is the sleep delayAdvantages

(i) It enhances performance using multiple slots

2111 Z-MAC (ZebraMAC) -is rides on the strengths andoffsets weaknesses of TDMA and CSMA It yields very goodchannel usage and low delay under low contention and lowerscollision among two-hop neighbors at a reduced cost Itsperformance is strong via error synchronization failure inassigning slots and channel conditions that change with timeIn the set-up phase the protocol operates through certainsteps including neighbor discovery slot assignment localframe exchange and global time synchronization whichunless the topology of the network changes significantly donot run again When the data are being transmitted there isimprovement in network throughput and energy efficiencywhich covers up for the initial upfront operational costs [35]

Disadvantages

(i) Initial slot assignment is quite a cumbersome taskhere since it is done at the beginning yet this maynot be at the same time in reality It further as-sumes static links until setup happens again whichtoo is impractical

(ii) It also suffers from the hidden-terminal problemsince slot owners have a small contention windowand priority for slots allocated to them due to theECN messages

(iii) Synchronizing time also poses a challenge in thisprotocol due to a large clock drift between nodes

(iv) Source and destination nodes too are not wellcoordinated in Z-MAC [36]

2112 Q-MAC (QoS-Aware Media Access Control) -isprotocol offers high QoS with improved energy efficiencyand accesses the wireless channel using MACAW as theunderlying protocol [37] Traffic from different nodes isprioritized for QoS satisfaction depending on how criticalthe data are It mainly has two scheduling algorithms inWSNs (i) intranode and (ii) internode scheduling -eintranode scheduling scheme adopts a first-in first-out-(FIFO-) based queuing algorithm based on application andMAC layer abstraction for data packet classification In-ternode scheduling reduces idle listening and collision on thechannel for improved energy consumption

Advantages

(i) It gives a high QoS with improved energy efficiency

2113 WiseMAC (Wireless Sensor MAC) WiseMAC isbased on the Aloha protocol and employs preamble sam-pling in achieving low-power transmissions in infrastructuresensor networks by occasionally sampling the channel tocheck for activity [3238] -e protocol employs the samemethod as B-MAC except that the sender shortens theextended preamblersquos length by scheduling its transmissionsaccordingly after learning the receiverrsquos awake schedules-is is done by the receiver node putting its subsequentawake time on the ACK packet such that when the senderneeds to send to it again it starts the preamble shortly beforeit wakes up so as to avoid energy wastage in transmitting thepreamble [32] -e data frame is repeated instead of theextended preamble for less traffic transmissions that have alengthier preamble than the data frame in WiseMAC Afterprocessing the frame a node that is not the target recipientwill go back into the sleep mode while the recipient staysawake till the session ends thereby sending an ACK packetFor an idle channel the power consumption is rather lowwith this method Nevertheless there is a problem of longwake-up preamble limiting throughput and causing anenormous power consumption overhead in reception [38]All nodes overhearing a transmission bear the overheadtogether with the recipient node -e protocol can addressissues concerning low-power transmission but has nomechanism of nodes adapting to varying traffic patterns[32] To compensate for drift between clock at the accesspoint and on the sensor node there is a need to compute thewake-up preamble duration -e drift varies directly withtime since previous resynchronization and is given by thefollowing equation

TP min 4θl TW( 1113857 (11)

where θ is the time-base quartz frequency tolerance and l isthe interval between two transmissions Increase in trafficvolume naturally reduces overhearing

Journal of Computer Networks and Communications 11

Radiomodel for protocols with lower power transitiondelays between transceiver states and their powerconsumption are modeled Some of the states includeDOZE where the transceiver neither transmits norreceives but is prepared to power on to the receivetransmit state very fast RXwhere the transceiver listensto the channel receives data or demodulates data fromthe noisyidle channel and TX which is the trans-mission state by the transceiver Let TS be the set-uptime needed to turn the transceiver from the DOZE tothe RX or TX state TT the turn-around time needed toswitch between RX and TX states and B the trans-ceiverrsquos bit rate PZ PR and PT are the values of powerconsumed in the DOZE RX and TX statesrespectively

1113954PR PR minus PZ

1113954PT PT minus PZ(12)

where 1113954PR and 1113954PT are the changes in power as a result ofbeing in RX and TX states respectivelyTraffic model let N be the number of sensor nodesconnected to an access point (AP) and λ be the global rateat which the downlink Poisson traffic arrives at the APfrom the fixed network Traffic toward each sensor node isthe same and is given by λN Average packet interarrivaltime Lwith which data packets arrive at a node is given byL Nλ If TC and TD are the control packets and datapackets respectively assuming a small traffic volume theglobal interarrival 1λ far exceeds the summation ofdurations for the data packet and control packet and theturn-around time Mathematically it is given as

1λ≫TD + TC + TT (13)

-e average power consumed by WiseMAC andtransmission delay incurred are given by equations (14)and (15) [38]

PW PZ +1113954PR TS +(1B)( 1113857

TW+

1113954PR X + TD + TT( 1113857

L

+ 1113954PR(N minus 1)Y

L

(14)

where

X 2θL 1 minus eminus TD4θL( )1113874 1113875

Y T2D + 12TDθL

2TW1 minus e

minus TW4θL( )1113874 1113875

DW TD +TW

21 minus e

minus TW4θL( )1113874 1113875

+ 2θL 2 minus eminus TD4θL( ) minus e

minus TW4θL( )1113874 1113875

(15)

where PW and DW are the power consumed byWiseMAC and the transmission delay for WiseMACrespectivelyAdvantages

(i) It achieves low-power transmissions in in-frastructure sensor networks through preamblesampling while checking for activity

(ii) It reduces overhearing due to increased trafficvolume

Disadvantages

(i) It has a problem of long wake-up preamble limitingthroughput and causing enormous power con-sumption overhead

2114 X-MAC It is a simple minimal transmission powerMAC protocol that uses a short preamble with decouplingfor the sending and receiving node sleep plans inWSNs-isleads to a significant reduction in energy usage at sendingand receiving nodes and decrease in per-hop latency andgives flexibility in adapting to bursty and periodic sensordata sources [32]

Protocol design X-MACrsquos design is aimed at increasingenergy efficiency reducing data latency and increasingthroughput -is enables application in different kindsof packetized and bit stream digitized radios as well assimplified low-overhead distributed application -edesign is as such in order to address the low-powerlistening problems including overhearing excessivepreamble and incompatibility with packetizing radiosas in [32]Asynchronous duty cycling a node with information totransmit will send the packets after sending a preamblewhereas the remaining nodes will preserve their sleepschedules that are not synchronized -e receiver willsample the link on waking up and stays awake if it detectsa preamble and if it is the target it will return to sleep if itis not the target after getting the whole preambleEmbedding the target ID in the preamble to avoidoverhearing the lengthy preamble is divided into manyshortened packets embedded with the receiver nodersquosID so as to eliminate overhearing One lengthy pre-amble is constituted by the shortened preamble packetstream Any node on receiving the short preamble afterwaking up will examine the target ID on the packet Itgoes back into the sleepmode right away resuming dutycycling if it is not the targeted recipient or stay awakefor successive packets if it is the one -is approachchecks energy wastage due to network density It canalso be used on several types of radiosReducing excessive preamble using strobing it ispossible to enable low-power transmission and alsosave energy by reducing time for preamble trans-mission through the use of enhanced preambles inaddition to preamble sampling X-MAC does not sendan endless preamble packet stream but puts slight

12 Journal of Computer Networks and Communications

pauses in between packets to allow the sending node topause and listen to the channel and allow the receivingnode to resend an early ACK packet during that pause-is makes the sending node begin sending datapackets and the receiving node cut excessive preamblesand thus decrease per-hop latency and energy wastageon unnecessary waiting and transmissionPacketizing radios low-power listening is limited incapability for supporting packetizing radios butX-MACrsquos short strobed preambles can support all typesof digital radiosAdaptation to traffic load while a number of WSNapplications are reliable in traffic production they stillmust adapt to changing traffic Nodes are going to haveseparate sleep schedules depending on traffic loadConsidering systems that have time-fluctuating trafficall predetermined static schedules are suboptimalOptimality Buettner et al [32] model the expectationof energy as follows

Es (preamble energy + energy perACK listen)

lowast(expected preamble minus listen iterations)

+(energy to send parket)

Es PTxSp + PRxSal1113872 1113873 Rl + Rs( 1113857

Rl minus Sp+ SdPTx

(16)

where Es expected energy to send a packet PTx andPRx are the transmission and receiver power re-spectively Sp Sal and Sd are the duration of thesenderrsquos preamble ACK listen and data transmissionperiods respectively and Rl and Rs are the receiverrsquoslisten and sleep periods respectively Expected energyto receive a packet is given as

Er (listen cycle energy + sleep cycle energy)

lowast(expected iterations for preamble arrival)

+(energy to sendACK) +(energy to receive packet)

Er PsRs + PRxRl( 1113857

1 minus 1 minus Pd(t)( 1113857 Rl + Rs( 1113857+ RaPTx + RdPRx

(17)

And the one-hop expected latency is given as

Lat (preamble duration + ACK listen)

lowast(expected number of iterations)

+ duration to send packet

Lat Sp + Sal1113872 1113873 Rl + Rs( 1113857

Rl minus Sp

(18)

Advantages

(i) It increases energy efficiency(ii) It reduces data latency(iii) It increases throughput

2115 A-MAC (Advanced MAC) A-MAC is a TDMA-based protocol whose design is aimed at enhancing thelifetime of the network for low rate and reliability of datatransmission [39] To assign time slots it does not depend ona central controller like most protocols but it gathers in-formation from its neighbors so as to allocate the slotsthrough a distributed algorithm Energy is controlled byscheduling power when there is not a single event Its designcomprises many frames with each having many time slotswhich at the beginning have got a beacon transmitted tosynchronize and exchange information with neighborsNodes to participate in the subsequent session are selectedby the controlled node so they do not go into the sleep modewith others [13] -is protocol operates under four time slotdivisions namely initial wait discover and active states Anode in the initial state begins by listening to the medium forthe beacon packet from other nodes for network synchro-nization which when received makes it change the timerthrough subtraction of beacon transmission time from re-ception time To avoid synchronization problems due todrift and allow for continuity the node has to get thestrongest beacon signal It then enters the wait state afterchoosing a number of waiting frames so as to reduce theprobability that nodes get into the discover state simulta-neously In case the synchronization beacon is lost while stillin the wait state it will return to the initial state for anotherone On expiry of the waiting counter it enters the discoverstate and collects information from neighbors by listening totheir beacon messages On successful selection of a time slotit enters the active state where it endlessly transmits beaconsat the initiation of the slot A node goes to sleep when abeacon has been transmitted and there are no more data fortransmission or in case the neighborrsquos beacon received in-dicates no data packets are coming in [39]

Advantages

(i) It can enable multicasting(ii) It enhances the lifetime of the network(iii) It offers reliable data transmission

2116 L-MAC (Lightweight MAC) Based on TDMA thisprotocol is aimed at minimizing the number of transceiverswitches adapting the nodal sleep interval to data trafficamount and making implementation less complex [40] Apacket is transmitted comprising a control and a data partControl packets have a fixed size and the time slot con-trollerrsquos ID -ey show the distance from the transmittingnode for simplified routing to the network gateway as well asaddressing target node and reporting data unit length andinternode synchronization Adjacent nodes are more in-terested in getting control packets from nearby nodes -enodersquos transceivers are switched off until the subsequenttime slot unless the node is also addressed or the message isomnicast If so it listens to the data unit which may not fillthe remaining time slot in totality-e transceivers on eitherside are turned off when the transfer of packets is successfulEnergy wastage through idle listening in uncontrolled timeslots is avoided via a short time-out interval Network nodes

Journal of Computer Networks and Communications 13

are able to communicate without colliding-is saves energyand increases the lifetime of the network A node in L-MACcan only transmit one message per frame [13] -e networkis set up in such a way that all nodes are not synchronized onbeing switched on-e gateway takes control of a timeslot soas to have them synchronized and its immediate neighborgets the control messages thereby synchronizing their timerswith the gatewayrsquos Neighboring nodes learn of multiplegatewaysrsquo time slots in their vicinity just after one frame-ejust synchronized nodes randomly choose time slots forcontrolling as long as they are not occupied -e nodes thuspreserve a table for the neighboring nodes which makes atimeslot reusable after a minimum of 3 hops and preventsmessage collision All nodes keep their slots before thebattery is down or learn of an impending collision All nodescan track their hop distances to a selected gateway node andthe information is communicated in control messages [40]

Advantages

(i) It minimizes the number of transceiver switches(ii) It adapts the nodal sleep interval to data traffic

amount(ii) It makes implementation less complex(iii) It offers improved network lifetime(iv) Energy wastage through idle listening in un-

controlled time slots is avoided via a short time-out interval

Disadvantages

(i) L-MAC is not very good for mobile SNs since slotsare calculated only once

(ii) It suffers idle listening because nodes keep listeningto slotsrsquo control parts so as to get data and enableother nodes to join the network all the time [13]

2117 C-MAC -is was designed for WSNs with 3 seg-ments namely (i) aggressive RTS having double channelcheck to assess the channel (ii) anycast to initialize the flowand (iii) convergent packet forwarding to stabilize the flowIt promises an improvement in energy efficiency lowerlatency and enhanced throughput while avoiding syn-chronization overhead though it operates only in low duty-cycling apps especially where there is less traffic But whenthere is more traffic the protocol employs anycast-basedpacket forwarding for waking the nodes up or finding aforwarder quickly and converges from route-suboptimal toroute-optimal unicast After this energy could be preservedby the nodes using a synchronized wake-up plan [13]

Advantages

(i) It improves energy efficiency(ii) It lowers latency(iii) It enhances throughput avoiding synchronization

overhead

Disadvantages

(i) It operates only in low duty-cycling applicationsespecially where there is less traffic

22 MAC Protocols for WMSNs

221 Diff-MAC -is is a CSMACA-based QoS-awarehybrid-priority-based MAC protocol for WMSNs It targetsenhancing the channel utilization while making use of ef-ficient service differentiation to do coordination for mediumaccess of each traffic class while giving fair quick datatransfer It offers QoS using the following features [14]

(i) It balances energy consumption and delay throughsensor node duty-cycling adaptation depending onthe dominant traffic class

(ii) It has a feature for fragmentation and messagepassing that breaks lengthy video frames intosmaller packets that are transmitted as a burst thusreducing retransmission costs if there are any MACfailures

(iii) -ere is fair data delivery amongst the nodes andtraffic classes respectively resulting from intranodeand intraqueue prioritization reducing unbearableperformance

(iv) -e number of collisions can be reduced and packetlatencies are kept within limits by adjusting the sizeof the congestion window in this protocol as per thetraffic needsAdvantages

(i) Diff-MAC is very suitable for WMSNs since itadapts quickly to changes in networkconditions

(ii) It is a fair protocol(iii) It is scalable

Disadvantages

(i) It is quite difficult to monitor the statistics ofthe network

(ii) Adapting dynamically to the network condi-tions is not an easy task

(iii) It exhibits some degree of latency in packetdelivery

-e implementation of Diff-MAC on Imote2 indicatesthat resource requirements can be met on the sensorhardware Results of extensive simulation runs showed thatDiff-MAC outperforms its competitors

222 Cluster-Based On-Demand Multichannel MAC Pro-tocol for WMSN (COM-MAC) -is is an energy-efficientprotocol in WMSNs with high throughput and reliable datatransmission It achieves energy efficiency by schedulingmultichannel media access to reduce channel contentionand remove collision idle listening and overhearing [41]

Design it has a three-phased operation with the (i)request phase (ii) scheduling phase and (iii) datatransmission phaseRequest phase here all nodes have two protocols eachfor sending requests to the cluster head (CH) ie thecontention-based and contention-free TDMAFDMA

14 Journal of Computer Networks and Communications

protocols -e contention-based protocol allocates achannel per node for sending the request message andfree channels are useable as control channels Forcontention-free protocols a control slot is allocatedwhen a network is deployed Here a slot per channel isallocated to a different node enabling requests to be sentwithout interfering with transmissions from othernodes -e contention-based protocol operates undertwo steps (i) Control channel assignment phase achannel is assigned per node for transmitting a request(REQ) message A channel is assigned to multiplenodes and nodes are evenly distributed to availablechannels for congestion control (ii) Request trans-mission phase nodes with data send a REQmessage viathe assigned control channel to notify the CH whichresponds with an ACK message -e contention-freeTDMAFDMA protocol too has two phases includingthe control slot assignment and request transmissionphases For unreliable channel conditions and lowtraffic load the contention-based protocol is fit for usebecause of less collisions and congestion whereas thecontention-free protocol is preferable in heavy trafficload and reliable channel applicationsScheduling phase from the above phase a schedule isgenerated by the CH for coordination and this enablesnodes to transmit and broadcast their data via controlchannels-e schedule for broadcasting only comprisesdata for nodes assigned to a channel -e schedule isrebroadcast for transmission reliability enhancementbut for energy efficiency the nodersquos transceiver isswitched off when all the schedule is completely de-livered till it is the time for transmitting again-e timeslot and data communication radio channel are allincluded in the schedule generated for each nodeData transmission phase on getting the schedule thesensor node sends its data as per the time slot and thechannel assigned -e time slot is divided into datatransmission andACK sections ACK supports link layererror control Using implicit selective repeat ARQ theCH sends an ACK message for all received packets Ifthere is no ACK the packet is assumed lost and hence aretransmission during the subsequent interval is done-is enhances reliability of transmissions though excessdelay is possible here which is also intolerable for real-time critical apps like video or audio -is gives rise to ahybrid MAC protocol to exploit the unused spectrumduring data transmission sessions [41]Advantages

(i) It achieves energy efficiency by scheduling mul-tichannel media access to reduce channel con-tention and remove collision idle listening andoverhearing

(ii) It maximizes network throughput by dynamicallyallocating time slots and channels for nodes byexecution of a traffic-adaptive QoS-aware sched-uling algorithm depending on needs for QoS andtraffic data information

(iii) It increases transmission reliability by in-corporating spectrum-aware ARQ to avoid spec-trum wastage

223 EQoSA -is protocol is developed for video andimage transfer across WMSNs in a hybrid mode withprovision of QoS Concerning the active sensor nodes andtheir traffic loads it employs dynamic session sizes andchanges bitmap-assisted (BMA) MACrsquos fixed session sizeNodes declare any data available for transmission duringcontention time -is forces the cluster head to make aschedule of how slots have been assigned It communicatesthis schedule to the nodes giving each node a given numberof slots per session and so enabling the busty traffic to beaccommodated [14] -e protocol does not perform wellwith DiffServe mechanisms but does well with schemessupporting busty traffic loads and strong cluster heads arenecessary in WMSNs that can assign and broadcast slots

224 QoS-Supported Energy-Efficient MAC (QE-MAC)Based on the IEEE 80211e standard this protocol ensuresfairness and has low latency and jitter and better efficiencyon energy usage than other QoS-aware MAC protocols forWMSNs It operates under a two-phased routine

-e first phase introduces the innovating prioritymechanism in IEEE 80211e QoS and several data types areassigned priorities Its coordination function is a combi-nation of distributed contention-based channel allocationand centralized polling-based channel access mechanisms

-e second phase preserves the nodesrsquo energy usingdynamic duty cycling -ere is an exchange of RTSCTSpackets to supplement CSMACA by muting nodes near thesenderreceiver during packet duration [42]

Advantages

(i) It ensures fairness(ii) It has low latency and jitter and better efficiency on

energy usage than other QoS-awareMAC protocolsfor WMSNs

225 Black-Burst (BB) Contention A contention-basedMAC scheme for enhancing instantaneous QoS access tocarrier-sense WSNs is proposed in [42] In this schemenodes contend for the channel until it is free Real-timepackets are given higher priority over non-real-time packetsand if a node is transmitting real-time packets it is prior-itized over other nodes [15] Voice and video are some of thereal-time apps in consideration here which periodicallyaccess the common radio channel when transmitting -eseneed constrained end-to-end delay of packets at the MAClayer -is protocol majorly does the following [42]

(i) Prioritization of real-time traffic(ii) Implementation of the round-robin mechanism in

real-time nodes(iii) Bounded access delays among real-time packets

Journal of Computer Networks and Communications 15

(iv) Implementation of real-time communication withvarying needs of bandwidth good for multimediatraffic

Amechanism for dynamically assigning packet prioritiesbased on their deadlines and the traversed hops is proposedin [43] -e ReAllocative Priority (ReAP) mechanism ismeant to provide QoS at the MAC layer to video traffic -isscheme introduces adaptive TXOP by modifying the TXOPdynamically based on the queue length [15] Other protocolsin this category include dynamic duty cycle and adaptivecontention window-based MAC protocol frame sharing(FRASH) MAC protocol and real-time independentchannels (RICH) MAC protocol

Advantages

(i) It can implement real-time communication withvarying needs of bandwidth

23 Challenges and Opportunities(a) One of the challenges in satellite transmission is

cross-talk leading to interference in different fre-quency bands It is necessary to increase the rise andfall times as well as protect sensitive nodes as wecircumvent floating ones if we are to regulate cross-talk as in satellite communication therefore Wefurther need to investigate the possibility of usingdifferential signaling -is constitutes an opportu-nity for further research

(b) Another challenge is that the hidden and exposedterminal problems are still noticeable despite the useof RTSCTS solutions in MAC since collisions arestill possible from transmissions of separate nodesAt the MAC layer ACKs might block efficientchannel usage by prohibiting exposed terminalsfrom reusing the channel and may be eliminatedleaving no proof of sender data packet receptionUsing a directional antenna is one of the proposedtechniques of high-speed WSNs like IEEE 80211-is constitutes an opportunity for further study intothese issues for better solutions though

(c) Another challenge is that since nodes in contention-based MAC protocols are mobile there are manycollisions by the packets and schedule in-consistencies in schedule-based protocols in the two-hop neighborhood information when nodes areentering or leaving [44] Deng et al [45] propose amobility-based clustering (MBC) protocol for mo-bile nodes to enhance the packet delivery rate -isprotocol in comparison with CBR has 25 chancesof decreasing collision-induced packet losses and50 chances in comparison with low-energyadaptive clustering hierarchy (LEACH)

(d) -e process of analyzing and evaluating routingperformance for WMSNs is quite complex becauseof the bursty multimedia traffic Because of the wayencoding scheme frames are structured as well as thechanges in between the scenes that occur naturally

the compressed video frames display some substantialburstiness on several time scales -ere will also befrequent route recalculation that is as well energy in-tensive for nodes frequently capturing and transmittingenergy -e solution here could as well possibly be theuse of the MBC protocol Route recalculation tooshould be dynamic to cater for the mobile nodes whichmight be in and out of the range at different instants oftime Recently the use of smart dumpsters for effectivewaste management in smart cities has also been pro-posed as a possible solution [46]

(e) For WSNs their routing protocols do not look atmultimedia applications which require enormousbandwidth and processing energy and should betransmitted in real time with utmost fidelity Extrachallenges in routing are imposed in WMSNs indeployment of heterogeneous sensors since therecould be audio video and still pictures in addition toscalar data during transmission Heterogeneous datahandled also have different QoS requirements andwith many different business needs different ser-vices with varying requirements give rise to a bigchallenge for the routing design -ere is a need todesign protocols that can support WMSNs such asEQoSA [14] for this purpose

Table 2 summarizes the reviewed MAC protocols forWSNs and the ones designed for suitability in WMSNs

3 QoS Strategies at the Application Layer

We discuss many protocols used to solve different needs ofcommunication between machines

31 ConstrainedApplication Protocol (CoAP) It is a requestresponse protocol designed by the Internet Engineering TaskForce (IETF) for constrained devices such as those havinginsufficient RAM or CPU and WPANs with low-powerconstraints resulting in poor transfer of packets andenormous overhead -e protocol was designed for ma-chine-to-machine (M2M) apps and system automation toreduce bandwidth requirements increase packet transferdecrease the overhead and make work simpler leading tolightweight implementation HTTP commands like GETPOST PUT and DELETE are used in a client-server ar-chitecture Using the publishsubscribe architecture CoAPcan support multiple users and hence yield better perfor-mance via multicasting enabling asynchronous messageexchange -is protocol can support both multicasting andunicasting via UDP and hence is a reliable and simpleprotocol -is is done via the messaging sublayer (for re-liability based on UDP) and the requestresponse sublayer(for communication and interaction) [16]

It uses the following messages

(i) Confirmable message this guarantees reliabilityusing ACK messages sent by the receiver node-ese could be synchronous or asynchronous ifthere is a need for more computational time

16 Journal of Computer Networks and Communications

(ii) Nonconfirmable message here no ACK messagesare sent back

(iii) Acknowledgment (ACK) message this confirmsreceipt of a confirmable message

(iv) Reset message this confirms receipt of an un-processed message in case any critical parts nec-essary for interpreting the message are missing

(v) Piggybacked response the receiver directly re-sponds while getting the ACK message

(vi) Separate response the reply by the receiver is dis-crete from the ACK message

CoAP has no in-built security features It is secured bythe datagram transport layer security (DTLS) that runs ontop of UDP though it was not designed for the IoT -ismakes its suitability debatable since DTLS offers no supportfor multicasting It also needs more packets for handshakingthereby increasing the traffic volume and computationalresources and so shortening the lifespan of mobile devicessince they use batteries [17]

Advantages

(i) It can support multiple users and hence yieldbetter performance via multicasting

Table 2 Comparison of MAC layer protocols used in QoS provisioning

Protocol Features QoS parameter Comments Access Priorityassignment Energy eff Designed for

S-MAC[22] Collision avoidance Energy latency Requires periodic

synchronization CSMA Hybrid Medium WSNs

B-MAC[23] Collision avoidance Energy delay

throughput Flexible interface CSMA Dynamic High WSNs

T-MAC[25] Collision avoidance Energy Complex nonscalable CSMA Dynamic High WSNs

C-MAC[13]

Dynamic channelallocation Energy Multichannel MAC CSMA Dynamic Medium WSNs

CU-MAC[24]

Multichannelmechanism

-roughputlatency High packet delivery ratio TDMA Dynamic mdash WSNs

A-MAC[39]

Per flow serviceguarantees Delay m-ary tree model

dynamic prioritiesTDMACSMA Dynamic Medium WSNs

ER-MAC[30]

Emergencyresponse

Energy delaydelivery ratio

Adapts to traffic andtopology changes

TDMACSMA Hybrid Higher than Z-

MAC WSNs

RI-MAC[31] Collision detection Delay packet

delivery ratioHigh packet delivery ratio

and throughput CSMA Dynamic Medium WSNs

SRI-MAC[33] Collision avoidance Energy delay Good data transfer and

energy consumption TDMA Dynamic High WSNs

H-MAC[34] Slotted sleep time -roughput

latencyData transmission only in

sleep timeCSMAAloha Dynamic Medium WSNs

Z-MAC[36] Collision avoidance Energy delay

throughputVery good channel usage

and error syncCSMATDMA Dynamic Medium WSNs

Q-MAC[37]

Intrainternodescheduling Energy High QoS improved

energy efficiency CSMA Hybrid High WSNs

X-MAC[32]

Overhearingavoidance

Energy delaythroughput

Reduces excessivepreamble by strobing CSMA Adaptive

static Medium WSNs

L-MAC[40] Collision avoidance Network

lifetime energyReduces the number oftransceiver switches TDMA Static Medium WSNs

WiseMAC[38]

No downlinkchannel collision Power delay Significantly low-power

consumptionAlohaCSMA Static Higher than

ZigbeeInfrastructure

WSNsAR-MAC[27] Collision avoidance Energy delay Uses data packets and

control packets TDMA Static Higher thanIEEE 802154 WBANs

BB-MAC[42]

Real-time trafficprioritization Delay Employs round robin in

real-time nodesCSMATDMA Dynamic High WMSNs

COM-MAC [41] Collision avoidance Packet delay

throughputBalances reliability and

retransmissionTDMAFDMA Dynamic High WMSNs

EQoSA[14]

Busty traffic loadaccommodation Energy Provides QoS support for

video and imagesTDMACDMA mdash High WMSNs

QE-MAC[42] Collision avoidance Energy jitter

latencyRuns central polling-like

channel access CSMA Dynamic High WMSNs

R-MAC[26] Collision avoidance Energy delay Reservation based mdash Static High Underwater

SNs

Journal of Computer Networks and Communications 17

(ii) It enables asynchronous message exchange(iii) It is a reliable and simple protocol

32 Message Queue Telemetry Transport (MQTT)Developed by IBM and later improved by OASIS with anintention of decreasing bandwidth MQTT is a simple open-source publishsubscribe protocol analogous to the client-server version It is appropriate for lightweight IoT appli-cations and M2M transmissions in constrained situationslike limited power computation capability and memory orbandwidth -e network bandwidth and the computationalresource usage are decreasing since clients are not updatedmaking publishsubscribe protocols outperform requestresponse protocols via IoT requirements [16] -is protocolcan run over TCPIP and so has the following advantages

(i) Delivering of packets with reliable guarantee(ii) Enabling multicasting(iii) Remote device session establishment(iv) Reduction of transport overhead and protocol ex-

change to minimize network traffic(v) Provision of the notification mechanism when

anomalies occur [47]

MQTTrsquos design targets economical use of bandwidthand battery so it has a major application in FacebookMessenger MQTT ensures QoS and reliability of messagedelivery by providing the following three QoS levels

(1) Fire and forget a message is sent once depending onthe networkrsquos best effort and there is no need forACK (lowest QoS level)

(2) Delivered at least once a message is sent at least onceand an ACK message is needed to avoid duplicates

(3) Delivered exactly once by a four-way handshake itensures a message is sent exactly one time (highestQoS level)-e architecture it has three key components (i)publishersmdashlightweight nodes sending data to thebroker before returning to sleep (ii) broker whichorganizes sensory data in topics and sends them tointerested subscribers and (iii) subscribersmdashIoTapps interested in data sent by sensors through thebroker as shown in Figure 3

MQTT is a simple lightweight protocol that is betterthan CoAP via low and high packet loss probability interms of throughput and exhibits lower latency in lowsampling rates [48] It supports a number of imple-mentations for embedded devices and mobile applicationsand offers web-socket support too Nevertheless it hassome shortcomings

(i) It is only employed in simple data types(ii) It exhibits high latency in high sampling rates(iii) It is unsuitable for real-time apps in such cases(iv) It does not provide automatic discovery andor

sufficient security at the protocol level [49]

33 Secure Message Queue Telemetry Transport (SMQTT)-is is an extension of MQTT that is based on lightweightattribute-based encryption using elliptic curve cryptogra-phy -e protocol has three entities as shown in Figure 4(i) publisher that publishes information under a certaintopic (ii) subscriber that obtains the same topic in-formation via a broker and (iii) public key generator(PKG) or broker that acts as the trusted third party -eprotocol has four phases namely (a) set-up phase (b)encrypt phase (c) publish phase and (d) decrypt phase

(a) Set-up phase publisher and subscriber provide aunique ID with attributes for registration with thePKG -is generates a secret master key set andpublic parameters that it publishes with set U A1A2 An For ciphertext policy-attribute-basedencryption (CP-ABE) the PKG gives a key set todevices with attributes [50]

(b) Encrypt phase based on the access tree the publishercreates an access policy An access tree is sent by thepublisher to the PKG for generating the key-gen-eration policy in the key policy- (KP-) based ABEalgorithm -e publishers know beforehand thetopics and subscribers to access as well as the rulesfor access Subscribers are issued with private keysfor all communication sessions Publishers in theCP-ABE algorithm come up with access trees andrules where the payload is encrypted by the pub-lisher who also issues more info needed for de-cryption and the rules With the help of public keycryptography data are encrypted giving rise to ci-phertext on the basis of KPCP-ABE and the receivergets SUBACK from the broker [51]

(c) Publish phase ciphertext is embedded in theSPublish command as payload and topic name isdefined in the header by the publisher -e brokerreceives the SPublish packet and replies withPUBACK prompting the publisher to respond withthe PUBREL packet -e broker then broadcasts totopic subscribers and deletes the data forwarding thePUBCOMP packet to the sender node

(d) Decrypt phase the encrypted information isdecrypted by the subscriber with the help of a secretkey Considering KP-ABE on verification that itsatisfies the access policy the subscriber invokes thePKG for a corresponding key which it sends afterverifying the request With CP-ABE subscribers usethe private key for decryption after authenticationthereby enabling offline interaction [51]Advantages

(i) -e information being encrypted makes it moresecure

34 MQTT for Sensor Networks (MQTT-SN) -is is a ver-sion of MQTT designed to operate under environments oflow bandwidth high link failures and short message lengthIt is a publishsubscribe (pubsub) protocol designed for

18 Journal of Computer Networks and Communications

UDP-based WSNs to visualize communication for low-power devices [5152] It comprises a publisher device actingas the MQTT-SN client to forward communications to agateway that also changes the MQTT-SN message to theMQTTmessage before delivering it to the broker that sendsit to MQTT-SN subscribers [51]

It has a gateway for changing protocols from MQTT toMQTT-SN (Figure 5) [52] -is protocol supports nodes in

the sleeping mode using an offline keep-alive procedure thatcan be used by battery-operated devices to get into thesleeping mode To support sleeping clients MQTT-SN has anew offline keep-alive procedure with which battery-oper-ated devices can go to the sleeping state in which theirmessages are buffered at the server until they are awake InMQTT-SN clients with no preconfigured gateway addresscan figure out the real address of their operational gateway

Publisher

Subscriber

Subscriber

Broker

Publisher

4G

Figure 3 Architecture of the MQTT protocol

Register by mentioning URI Register by mentioning URI

Access tree

Set-up

Publisher ReceiverPKG(broker)

Key management Key management

Encrypt data using public parametersand random key This is done using

CP-ABE Publish it as

Subscribe (topics)SUBACK

S-Publish (TopicEncrypt (data))S-Publish (TopicEncrypt (data))

Encryptand publish

PUBACK

PUBACK

PUBRELPUBRELPUBCOMP

PUBCOMP

Figure 4 SMQTT protocol

Journal of Computer Networks and Communications 19

within the network through a discovery procedure Simul-taneously we may have several gateways in a single wirelessnetwork that can be cooperative via load sharing

MQTT-SN architecture it is composed of MQTT-SN clients gateways and forwarders (Figure 5)rough the MQTT-SN gateway (GW) the MQTT-SN protocol is used by the MQTT-SN clients toconnect to the MQTT server Integration betweenthe MQTT-SN GW and the MQTT-SN server mightbe possible or not Translating between the MQTT-SN and the MQTT for a stand-alone GW is doneusing the MQTT-SN protocol If a GW is detachedfrom the network it can be accessed by the MQTT-SN clients through a forwarder is just encapsu-lates the received MQTT-SN frames that are on thewireless side before forwarding them to the GW un-altered It then decapsulates the received frames tooresending them to the client as sent by the GW [52]ere are 2 types of gateways (a) transparent gatewayand (b) aggregating gateway as shown in Figure 6

(i) Transparent gateway this launches and sustains adirect MQTT connection between the MQTT-SNclient and the MQTT server which is entirely re-served for end-to-end client-server messaging enumber of servers and that of MQTT-SN clientsconnected to the GW are the same is transparent

GW translates syntax between these protocols and iseasier to implement compared to the aggregatingGW But the MQTT server must support dierentsessions for all connected clients Some MQTTservers are congured to enable few parallel sessions

(ii) Aggregating gateway this maintains a single MQTTconnection to the server on behalf of all clients Allexchanged messages from the MQTT-SN client to theaggregating GW are directed to the GW which sortsthe information to forward to the server e aggre-gating GW decreases the number of MQTT sessionssimultaneously supported by the server in WSNs withmany battery-operated sensors and actuators (SAs)having limited storage and processing capabilities It ishowever more complex to implement compared to atransparent GWe general format of the MQTT-SNprotocol message has two or four message headeroctets and n octets of variable header that is optionaldepending on the message in question MQTT-SN has256 message types (MsgType) with those from0X1Eminus 0XFD 0X19 and 0XFF reserved for futureusage [51]

Advantages

(i) e protocol can operate under environmentsof low bandwidth high link failures and shortmessage length

MQTT-SN gateway

MQTT-SN

MQTT

MQTT-SN

MQTT-SN

MQTT-SN

MQTT-SN client

MQTT-SN clients

MQTT broker

MQTT-SN client

MQTT-SN client

MQTT-SN gateway

MQTT-SN forwarder

Figure 5 MQTT-SN architecture

20 Journal of Computer Networks and Communications

(ii) It is a pubsub protocol designed for UDP-basedWSNs to visualize communication for low-power devices

(iii) It can be optimal for low-cost battery-operatednodes having low processing power and lessstorage

(iv) It supports nodes in the sleeping mode using theobrvbarine keep-alive procedure used by battery-operated devices

35e Extensible Messaging and Presence Protocol (XMPP)e standardization of this protocol by the IETF made itpopular for usage over the Internet especially in instantmessaging and chatting platforms It is however susceptibleto spamming attacks and its failure to have worldwidesupport has made Google too withdraw its support of thestandard [17] e protocol is mainly useable in the client-server architecture but can also be applicable in client-clientXMPP communication [53] Among the application layerprotocols XMPP is the only one that supports both asyn-chronous (publishsubscribe) and synchronous (requestresponse) services while running over TCP but can still runover other protocols

Given the simplicity of its implementation and ease ofusability in XMPP instant messaging (IM) is suitable forXMPP and is implemented using the message stanzaltmessagegt ese stanzas are usable in one-to-one ormultiuser sessions or chatting or in error reporting andsending headlines Attribute type of the stanza denes themessage type e attribute type could be a chat group chatheadline normal or error An instant messaging network inXMPP can support many XMPP clients and servers with agateway congured to enable other IM network users toconnect to other XMPP users e XMPP server mainlycontrols sessions and XML streams to clients and servers inaddition to routing XML stanzas in the streams [53] eTCP ports 5269 and 5222 have been respectively registeredfor the XMPP server-to-server connections and XMPP clientconnections by the Internet Assigned Numbers Authority(IANA) XMPP enables specication of XMPP ExtensionProtocols (XEP) to enhance its usefulness [17]

XMPP security clients connect to the server usingtransport layer security (TLS) and simple authentication andsecurity layer (SASL) protocols which are at the same timeused by the servers for interdomain communications Fordata condentiality and integrity TLS will encrypt thestreams of XML for all sessions To conrm the identity of anyclient that needs to access the server the SASL protocol doesthe authentication of the XML stream For any communi-cating session with a server to start the client needs to resolvethe serverrsquos DNS hostname For protection of the server of theclient from third-party attacks the server maintains the au-thentication details secret ie IP address and access methodand only original connections of the server are required

In server-to-server communications the SASL protocolis employed for authentication and condentiality Server-to-server communications may as well be deactivated from aserver by an administrator depending on the organizationalpolicy To guard against domain-spoong attacks serverdialback can be a good solution since servers support itsecuring more the XML stanzas Nevertheless it cannot beused to authenticate secure or encrypt in-server streamsand the server identication resulting therein is not strongenough to be relied on It further cannot secure against DNSpoisoning attacks and IP session hijacking for remote do-mains [53] High-security domains can use TLS and SASL

Advantages

(i) It is popular for usage over the Internet especially ininstant messaging and chatting platforms

Disadvantages

(i) It is susceptible to spamming attacks(ii) It has failed to have worldwide support making

Google withdraw support of the standard(iii) It is not secure against DNS poisoning attacks and

IP session hijacking for remote domains

36 Advanced Message Queuing Protocol (AMQP) isprotocol can be utilized by a number of platforms formessage exchange since it is open source and written inmany languages It is a publishsubscribe model depending

MQTT-SNclients Broker

MQTT

(a)

MQTT-SNclients Broker

(b)

Figure 6 (a) Transparent gateway and (b) aggregating gateway

Journal of Computer Networks and Communications 21

on a reliable and efficient messaging queue with capacity forinstant messaging or e-mail for different applications es-pecially in IoT environments It differs from others in thesense that message types can be specified and have theirsource traceable and how to do trade-off between perfor-mance security and reliability For reliability of messagetransmission AMQPmakes use of some delivery guaranteeslike at most once at least once and exactly once It can alsohowever utilize the TCP transport layer for guaranteeingreliability -e AMQPrsquos publishsubscribe method has twocomponents namely (i) exchange queue meant for routingof messages to an appropriate order in the queue and (ii)message queue that acts as storage pending their delivery toreceivers [16] In this protocol model the broker can makerouting decisions contrary to othermessaging systems whereapps using the queue have in them embedded the decision-making logic for recipients or senders -e developer mayneed to check each of the affected apps so as to change thelogic for routing and delivery of the messages [54]

Advantages

(i) It is convenient(ii) It has capacity for interoperability between dif-

ferent vendors(iii) It can use open standards to enable connection

between business partners(iv) It can enable innovations based on AMQP foun-

dations [55](v) It can enable supporting of mission-critical ap-

plications in e-commerce(vi) It can guarantee reliable delivery of messages

Disadvantages

(i) It is not very appropriate for real-timeapplications

(ii) It may not offer automatic discovery(iii) Interoperability is not guaranteed(iv) It lacks open-source libraries for constrained

gadgets [49]

37 Data Distribution Service (DDS) A product of theObject Management Group (OMG) DDS is a pubsubprotocol meant for M2M communications with two sub-layers -ese are the data-centric pubsub sublayer re-sponsible for connecting anonymous data publishers tosubscribers and the data-local reconstruction sublayer that isoptional and integrates DDS into the application layer [56]-e DDS architecture connects system components asshown in Figure 7 Publishers and subscribers are discon-nected with respect to (i) time (ii) space (iii) flow andbehavior (iv) platforms and (v) programming languages[57] A DDS data publisher generates topics and publisher-subscriber coupling is represented as topic name data typeschema and publishers and subscribersrsquo QoS attributesDelivering of data is done by the publisher layer -e datawriter and publishers together make a decision about anynecessary alterations for subsequent forwarding to thesubscribers who need to send data to IoT applications Data

readers are responsible for delivering topics to the sub-scribers after reading through them [56]

Some of the major components of the DDS architectureinclude the following [57]

(i) Topic this is a logical path specifying publicationand subscriptionrsquos data type between DataWritersand DataReaders For a successful session to begintopic names types and DataWriters and Data-Readersrsquo QoS should correspond

(ii) Domain this is a logical setting for transmissionthat can be utilized for separating and optimizingnetwork transmission in a setup of similar appli-cations If DDS apps share a domain ID then theycan exchange data among themselves

(iii) DataWriter and DataReader these are terminalobjects for writing and reading messages re-spectively to and from a universal data space

(iv) Participant this is an object that stands for apublisher or subscriber of the DDS app in a do-main acting like a container for other objects

Advantages

(i) It is highly reliable and offers QoS support(ii) It is highly scalable(iii) It is appropriate for real-time apps(iv) It is fault tolerant(v) It has high level of interoperability(vi) It does not need brokers that may act as

bottlenecks(vii) It offers automatic discovery(viii) It has a user-defined data structure for topic(ix) It supports publisher and subscriber decoupling

Disadvantages

(i) It is originally developed for only stand-aloneLANs

(ii) A lot of memory is needed(iii) It has no open-source libraries to support con-

strained devices(iv) It is hard to design and configure compared to

other protocols [49]

38 RESTful Services REpresentational State Transfer (REST)-is is an architectural style based on a set of principles thatdefine and address networked resources It offers web servicesthat enable transmission from the HTTP device to the devicewithin the IoT architecture and enables clients to reach theserver [16]-is protocol utilizes HTTP commands to backupthe requestresponse messaging model HTTP is a popularand secure protocol in the WWW using TLSSSL [16] ARESTful application has the following characteristics

(i) State and functionality are separated into distrib-uted resources

(ii) All resources are distinctively named by the use ofunique HTTP commands (GET POST PUT orDELETE over the Internet)

22 Journal of Computer Networks and Communications

(iii) It is a clientserver stateless layered protocol thatsupports caching REST uses the same methods asHTTP for all requests or responses with POST andGET for creating and retrieving resources PUT forupdating and changing the state of resources andDELETE for removing resourcesAdvantages of REST over arbitrary web services(SOAP) in the IoT context

(i) It reduces overhead(ii) It reduces parsing complexity(iii) It is stateless(iii) It integrates more tightly with HTTP [58](iv) It is easy to implement and learnis together

with its natural t over HTTP makes it apreferred method for Web 20 applications toshowcase their data and secure M2M sessionsin the IoT

(v) Applications that support RESTful web ser-vices exhibit better performance than the re-source-constrained WSN nodes [59]

(vi) REST is important for the IoT since it supportscommercial M2M cloud platforms

(vii) Its implementation in smart phones and tabletapps is also easy since it needs only an HTTPlibrary that is accessible to all operating systemplatforms [56]

Disadvantages

(i) Although it enjoys wide usage in commercialM2M platforms it is not likely to dominatebecause it is hard to implement

(ii) Its HTTP usage makes it incompatible withconstrained M2M devices leaving its usage toend applications

39 WebSockets It is a web-based protocol operating on asingle TCP channel and providing full-duplex transmissions

It permits secure duplex communication between a clientexecuting unreliable code in a restricted environment and adistant node communicating from that code while using theorigin-based security model employed by web browsers overTCP [60] e protocol is not a requestresponse model andis not a publishsubscribe model So unlike other protocolsa web-socket session is established by the client initiating ahandshake to the server to launch the session is onceestablished a full-duplex client-to-server connection beginsasynchronously and runs until terminated by either party[56] It gives a model for browser-based apps utilizing bi-directional communication with no need of many HTTPconnections to servers

WebSocket comprises two parts ie (i) the handshakehaving a clientrsquos message and the serverrsquos handshake responseand (ii) the data transfer HTTP headers should be repeated byapplications in all client requests and server responses arisingfrom the endless polling which might increase the commu-nication overhead subject to the application [61] WebSocketcan be used to come up with scalable real-time web appli-cations since it can enable full-duplex transmission via onesocket across the Web Its security can be ensured with thehelp of TLSSSL is protocol is suitable for resource-con-strained devices and IoTapplications will not have a good userexperience with it since it operates a clientserver architectureNevertheless it can outcompete all TCP-based protocolsbecause of the following advantages

(i) It is t for real-time transmission(ii) It has improved security(iii) It minimizes transmission overhead(iv) It can provide ecient messaging systems using

WebSocket ApplicationMessaging Protocol (WAMP)[17]

310 Streaming (Simple) Text-Oriented Messaging Protocol(STOMP) is is a TCP-based protocol designed for textmessage-oriented middleware (MOM) with the use of

Topic

Datareader

Datareader

Datareader

Datawriter

Datawriter

Datawriter

Subscriber PublisherPublisher

Data bus (DDS domain)

Subscriber

Topic Topic

Figure 7 DDS architecture

Journal of Computer Networks and Communications 23

HTTP-like commands for interoperability among platformslanguages and brokers Data communication takes placebetween the client and the broker in line with commands(like CONNECT DISCONNECT ACK NACK SUB-SCRIBE UNSUBSCRIBE SEND BEGIN COMMIT orABORT) then headers in the form ltkeygt ltvaluegt (one perline) then a blank line and then body content terminatingin a null character Server-client communication takes placeover a MESSAGE RECEIPT or ERROR frame with headerand body content format being analogous MOM productssupporting STOMP include Apache ActiveMQ or FuseMessage Broker HornetQ Open Message Queue(OpenMQ) RabbitMQ and syslog-ng through its STOMPdestination plugin [62] In addition to these other appli-cation layer protocols include simple media control protocol(SMCP) lightweight local automation protocol (LLAP)simple sensor interface (SSI) lightweight M2M (LWM2M)XMPP-IOT and simple object access protocol (SOAP)

Advantages

(i) It is lightweight for constrained networks(ii) It has flexibility to choose quality of services with

the given functionality(iii) It is standardized by OASIS Technical Committee(iv) It is easy and quick to implement

Challenges and opportunities

Here security threats are imminent because of the open-endedness of the application layer [63] -ese could includethe following

(i) -ere is poor or lack of a security design for anapplicationrsquos function Coming up with a gooddesign to address this could be a good ideaopportunity

(ii) Some of the programs might have backdoorsevading otherwise secure controls to allow un-authorized access to network resources We requiremechanisms to patch up such loopholes

(iii) Applications that lack or have weak authenticationmechanisms can be easily targeted by unlicensedusers to abuse the network Such mechanisms needto be strengthened further

(iv) Excessively complicated access control rules maybecause of their misunderstanding or poor writinggive unauthorized access allowing dubious activityrather than protecting the network they purport toprotect -ese should be simple and clear to un-derstand by all users

(v) In booting workstations without disks and managingnetwork devices we usually use the TFTP protocolalthough access does not require authenticating theuser by usernamepassword -is makes an intruderable to gain easy access to configaccess info as longas he can guess the filenames -is needs to be fixed

A few steps can be used to secure the application layerTo protect network data transmitted across the network

applications need to strengthen their encryption and au-thentication mechanisms Apps too need to enable strictcontrols of privileges for data access using good mechanismsto balance between usability and effectiveness [63] Appli-cations handling sensitive data need detailed logging andaudit capability as well as testing and reviewing

Table 3 summarizes the application layer protocols dis-cussed with their weaknesses and differences among them

4 Cross-Layer QoS Strategies

41 Service-Differentiated Real-Time Communication Scheme(SDRCS) -is is an event-driven routing protocol that routesreal-time traffic by featuring a cross-layer packet-forwardingdesign in which it integrates the real-time routing functionalitywith a new priority-based MAC scheme -e protocol ap-proximates distributed packet traversal speed for traffic clas-sification and admission control on the basis of this design Itfurther localizes decision-making by prioritizing packet for-warding to maximize the speed of packets [64]

Advantages

(i) It improves bandwidth utilization bandwidthdegradation due to unschedulable packets can beavoided

(ii) It offers a lightweight packet schedulability estima-tion mechanism Here it uses received signalstrength and admission control and then earlymissed-deadline packet-dropping policies are made

(iii) It delivers high end-to-end throughput via highersource data rates alongside strict end-to-end latencyneeds

(iv) Nodes which meet real-time requirements of lesstraffic load and better channel quality are givenhighest priority when it comes to packet forwarding

-e SDRCS operates via 5 mechanisms namely (1) RSS-(Rich Site SummaryReally Simple Syndication-) basedgrouping (2) admission control (3) prioritized queuing(4) real-time MAC and (5) dynamic forwarding -isprotocol has a queuing policy that employs per-hopdeadline-based queues and the nodes schedule theirpackets using FIFO priority-based queues [1] -e prin-cipal features of this protocol are mentioned as follows

(i) It is event driven and is easily adaptable to anynetwork changes

(ii) It does not need additional hardware for locali-zation or multichannel transmission

(iii) -e cross-layer packet-forwarding strategy can beused when transmitting multimedia traffic

(iv) It is able to efficiently circumvent any voids therein

Early deadline miss (EDM) policy drops all unscheduledpackets -e SDRCS circumvents wrong packet dropsdue to EDM since it adjusts to network changes easily

42 Sensor Fuzzy-Based Image Transport (SUIT) -is pro-tocol regulates congestion using a scheme based on fuzzy

24 Journal of Computer Networks and Communications

logic It transmits packets with reduced quality to allowablelevels if there is congestion enabling it to transmit lower-quality packetsframes without being dropped-e numberof packets delivered per second thus rises and while thisenhances video streaming apps it is more appropriatelymade for those that quickly transmit non-real-time videosuch as in video surveillance SUIT can stream JPEG pic-tures other than the traditional predictive video codingmethods that are not suitable for sensor devices because oftheir constraints of energy memory and computationalspeed Progressive JPEG-PJPEG performs better via packettransmission rate and delay though it degrades the imagequality because of congestion [65] -is protocol employscross-layer methods for interlayer data exchange though itstill employs some layering making it still possible topreserve all advantages of the layered design since it hasapplication and transport layers but has noMAC or routinglayer

For congestion detection and control SUIT bases on thepercentage number of in-bound and out-bound frames perwindow number of contenders and buffer occupancy of thenext-hop node Since buffer management has an effect oneffectiveness and QoS inWMSNs it is useful in the transportlayer so the protocol prioritizes packets for better QoS SUITreorders packets by tracking the sequence numbers of thedelivered packets since they all have a source ID framenumber and sequence number that it uses to detect anymissing packets -e protocol however does not providepacket reliability [65]

Advantages

(i) -e protocol prioritizes packets for better QoS(ii) It can transmit lower-quality packetsframes

without being dropped

Disadvantages

(i) It does not provide packet reliability

43 Network Layer QoS Support Enforced by a Cross-LayerController (NLQS) -is scheme permits packet-level servicedifferentiation as a function of throughput end-to-endpacket error rate and delay [66] (see Figure 8) -is en-hances network layer QoS It comprises a cross-layer controlunit (XLCU) for configuring and controlling networkingfunctionalities at physical MAC and network layers [18]-is depends upon the unified logic which affects choices forapplication layer needs and status of functional blocks whichdo implementation of networking functions [66]

Advantages

(i) Cross-layer interactions can be controlled withoutweakening the upgradability modularity andsimplicity of designing the system

44 Cross-Layer Signaling Shortcuts (CLASS) CLASS has ahigh level of effectiveness flexibility and comprehensive-ness According to [67] it has the following features

(i) Direct signaling among remote layers leading tofaster transmission

(ii) Lightweighted internal message format standard-ized protocols are not suitable for use in internalsignaling since they are not usually lightweight forinstance transmitting against faults in the networkCLASS requires only three fields namely destina-tion address event type and event contents

(iii) -e setup of its exterior messages is standardconsidering external signaling ICMP and TCPIPare suitable for general messages and short notifi-cations respectively

(iv) Message control protocol should ensure optimizedand organized exchange of dense simultaneousmessages across layers for high efficiency andavoidance of possible conflicts -e protocol appliesto many cross-layer signaling situations

Table 3 Differences and problems with application layer protocols

Protocol Architecture Transport QoS options Security Weaknesses

CoAP Requestresponse UDP DTLS No in-built security features DTLS does not supportmulticast

MQTT Publishsubscribe TCP TLSSSL Insufficient security at the protocol level sincepayload values are not encrypted

SMQTT Publishsubscribe TCP CPKP-ABE Key revocation and group pubsub for distributedSMQTT are still a challenge

MQTT-SN Publishsubscribe UDP TLSSSL SSLTLS suffers from attacks like BEAST CRIMERC4 and Heartbleed

XMPP RequestresponsePublishsubscribe TCP X TLSSSL Susceptible to spamming attacks and lacks worldwide

supportAMQP Publishsubscribe TCP TLSSSL Inappropriate for real-time applications

DDS Publishsubscribe TCPUDP TLSSSL Memory intensive and no open-source libraries forconstrained devices

REST Requestresponse HTTP X HTTPS Hard to implement Uses HTTP so incompatiblewith constrained apps

WebSockets PublishsubscribeClientserver TCP X TLSSSL Bad user experience for IoT apps since it runs the

clientserver architectureSTOMP Clientserver TCP HTTP -e broker can act as a bottleneck

Journal of Computer Networks and Communications 25

Advantages

(i) e protocol is scalable and has a very lowpropagation delay

(ii) It is very ecient and exible since it uses directsignaling between any two layers

(iii) It is very comprehensive and eective

45 Cross-Layer and Multipath-Based Video Transmission(CMVT) CMVT is a hybrid of application and networklayers At the application layer it carries out the encoding ofvideo streams to video data frames (I- P- and B-frames) bymeans of the MPEG-4 encoding format At the networklayer route discovery and data transfer take place Twoalgorithms greedy forwarding and rollback are used todiscover a number of paths from the source to the sink nodevia the route discovery method [20] A specied node icalculates the evaluation of its adjacent node j using thefollowing equation

fij (1 minus α)d2(j D) minus d2min(i)d2max(i) minus d

2min(i)

+ αeinit(j) minus eres(j)

einit(j)

(19)

where fij evaluation value for the node i to j d2(j D) isthe distance from the node j to the destination node D

d2min(i) and d2max(i) are the minimum and maximum dis-

tances of neighbors of the node i to D respectively einit(j) isthe initial energy of the node j eres(j) is the current residualenergy of the node j and α is the energy coecient given asfollows

α emax(i) minus emin(i)

emax(i) (20)

where emax(i) and emin(i) are respectively the maximumand minimum energy left of all node irsquos neighbors

e network layer further transmits video streams withCMVT doing status evaluation choosing an appropriatecommunication path for any packets and the QoS guaranteelevel for the path i is calculated as

fi (1 minus ω)hisum hi

+ ωnisum ni

(21)

where fi is the evaluation value for the path i hi are the hopsfor the path i ni is the summation of packets sent via isum ni isthe summation of packets sent by sources and ω is theenergy consumption factor [18]

Advantages

(i) e CMVT protocol is very superior in mediatransmission particularly large-scale WMSNs

To network peers

Flow requirements

-

Atimes

Admissioncontrol

Capacityassignment

Next-hopselection

Coding rate

Data rate

QoScontracts

Cross-layer controlunit (XLCU)

Localization subsystem

Synchronizationsubsystem

Networkcoordinationunit (NCU)

UWB modulator

Channel coder

QoS scheduler

Georouting

Source coder

Rate control

Figure 8 Cross-layer controller architecture

26 Journal of Computer Networks and Communications

46 Cross-Layer Cooperative MAC (CoopMAC)CoopMAC [68] is based on the IEEE 80211 DCF approach-ere is a reduction in interference of neighboring cellsleading to uniform coverage in densely deployed networks[68] A connection is established by two-way authenticationbetween the sender and the receiver When a Clear-to-Send(CTS) packet and short interframe space (SIFS) are obtainedby the sender packets are transmitted right to the receiverwith no valuable cooperation If available the two nodes findout whether they can exchange data -is is done by the useof a helper identification signal to establish how feasiblecooperative communication is It is established in thepresence of a signal and if it is not available direct com-munication is activated For distributed systems helper-initiated cooperation is preferable because of the presence ofthe RTSCTS packets [69]

Advantages

(i) CoopMAC has an advantage over others due to thespatial diversity amongst the three nodes [70]

(ii) It has the potential to achieve significantthroughput and delay with no complexity in thedesign of the system

47 Cooperative MAC Protocol for Multihop Networks (M-CMAC) Just as CoopMAC M-CMAC is made in such away that high data rate stations help low data rate stations toforward traffic for broadcasting [71] -ere are also helperschosen in such a way that 2 fast-hop transmissions replace aslow hop transmission -e helper via whom there existsleast delay from the sender to the receiver is considered bestbecause of theminimum two-hop transmission rate and so isthe chosen neighbor to the duo Here the working as-sumption is that all nodes have their position coordinatesknown and thus the Euclidean distances between all pairscan be calculated and converted to the data rate for that linkAll nodes have cooperative tables (CT) of potential helperswith destination and helper MAC address Euclidean dis-tance and total distance through the helper Any sender withdata to transmit checks out for the existence of any helper inthe CT for the destination and if available forwards RTS forchannel reservation for single-hop duration M-CMAC hasgot an RTS format having a shape with five fields similar tothe one given as follows

Framecontrol Duration Source

addressDestination

addressHelperaddress

-e destination nodersquos address is kept by the source in thepart for the helper address that is in turn saved as the desti-nation address Nodes inspect helper and destination addressfields on receipt of RTS whereby the node acts as the othernodersquos helper in case the helper address field is not the sameand the helper transmits CTS back to the source if at all it wantsto forward data On delivery of the CTS packet the sourcetransmits to the helper which in turn sends the packets to thedestination that also sends ACK to the helper on receipt of thepacket In M-CMAC there exists a higher level of channelreuse (parallel transmissions) due to the increased number of

nodes therein that subsequently raises availability of helpers fordata forwarding -is results in increased throughput incomparison with CoopMAC and IEEE 80211 DCF [71]

Advantages

(i) It promises a higher throughput compared toCoopMAC and IEEE 80211 DCF

48 Cluster-Based Cooperative Routing (CBCR) Protocol-e CBCR protocol has a multihop data-forwarding func-tion realized at the link layer with cooperative links that useM-CMAC -is protocol encompasses two stages namelythe routing relay selection phase and data forwarding phase

(i) Routing relay selection phase all nodes announcetheir presence to their neighbors by broadcastingperiodical beacon messages that carry the MAC ad-dress of the node Each of these constructs a relaytable that contains all neighboring nodes with whichit is able to communicate -e node further broad-casts its neighbor list in case of any change to itsentries since the previous broadcast -e MAC ad-dresses of nodes next to the node X are contained inits relay tablersquos column one and the neighbor nodersquosrow has MAC addresses for neighbors of the adjacentnode Based on its relay table each node in-dependently chooses routing relays Choosing a nodeas a relay node depends on the number of nodes itconnectsmdashit should be the highest number [71]

(ii) Data forwarding phase a node with packets totransmit needs to first verify if the receiver is in thesame cluster and if that receiver has a helper to whichthe packet is thence forwarded but in whose absencethe packet is delivered right to the receiver In casethe intended receiver is in a different cluster therelay table will be checked to see whether that re-ceiver is reachable via other routing relays If it isreachable packets will be sent to the routing relaydirectly or via the helper if at all the relay has one Incase of destination unreachable via relays packetsare multicast to all relays by the node [71]Advantages

(i) It enables multicasting

49 MAC-PHY Cross-Layer Protocol A cooperative cross-layer standard for cooperation at the physical layer in next-generation WMSNs is developed in [70] It provides acomplete MAC layer algorithm that gives a supportive shellfor the PHY-MAC layer Like CoopMAC its scheme de-pends on the middle node for communication betweennodes-eMAC layer protocol has been changed to regulateinformation exchanges at the physical layer -e receivernode receives duplicate packets from source and helpernodes to decode the data [70]

410 MAC-Centric Approach -is cross-layer protocoltargets multimedia applications by using the MPEG-4

Journal of Computer Networks and Communications 27

scheme [21] It is characterized by 4 access categories(AC3ndashAC0) according to their priority in transmission It ismeant for supporting different QoS needs in upcomingvideo applications and enabling MAC layer differentiationfor H264 partitioning [18] Because of low bandwidth delayand other QoS challenges that cause inefficiency in trans-mittingmultimedia data inWMSNs a number of algorithmsdepending on IEEE 80211e have been proposed to supporttransmission of quality videos [72] Selecting an AC dependson the measures for QoS such as the loss rate and delay-us AC3 having the highest priority is mapped to theparameter set concept since the stream is sensitive to bit lossespecially in video transmission [21]

411 Adaptive Cross-Layer Forward Error Correction(ACFEC) In the ACFEC [73] model data packets are ex-changed among nodes through the access point (AP) whichoperates the infrastructure mode in which it adds FEC to videodata [72] -ese data are dealt by encapsulation using astreaming server to the receiver through thewireless AP as RTPpackets An adaptive FEC controller senses the category ofpackets out of the RTP header and recovers the header ofpackets from UDP -e encoder generates some error-cor-recting packets whose number is determined by the sourcepacket number of the block Multimedia transmissions aremonitored by the controller using MAC failure data and itscounter is increased by one if there is a failed transmission-econtroller uses the failure counter to change the packet numberproduced after transmitting a block [18] In case there are lostpackets it changes the redundancy rates and produces extrapackets to replace the lost ones and satisfy the requirements ofthe receiving node [72] -e FEC packet number is enhancedor reduced to satisfy the receiverrsquos requirements and stoppacket losses-is is done through accurate detection of packetlosses and adjustment of redundancy rates If all video datapackets are well received there will not be generation of FECpackets [73]

Advantages

(i) It promises a good QoS through packet loss re-duction and redundancy rate adjustment

412 Balanced Cross-Layer Fuzzy Logic (BCFL) DesignRouting Algorithm in WSNs A new fuzzy logic-basedrouting algorithm (BCFL) [74] was designed using disper-sion of the cross-layer parameter as the fuzzy logic inferencesystem input Each cross-layer parameter has a dynamicweight depending on the value of dispersion -e designcomes with some innovations as follows (i) for fuzzy logicinference system input the absolute parameter value issubstituted by parameter dispersion thereby significantlyreducing algorithmic complexity (ii) dispersion does notchange with the order of magnitude according to the dis-persion formula and (iii) the weight of the parameter de-pends on the size of its dispersion and the two are inverselyproportional to each other -is enables BCFL to have somenobility unlike other algorithms as per the following dis-tinguishing properties making it advantageous

(i) It has got simple if-then rules which remain con-stant even when the constraints increase

(ii) It is capable of dealing with many constraints withno increment in complexity

(iii) It has capacity to yield a more balanced solutionthan other algorithms

(iv) It is easily adaptable to changes in network con-ditions and topology even when the changes arefrequent like in underwater WSNs -e algorithm isuseable in choosing the CH in cluster-based routingprotocols [74]

413 Minimum Hop Disjoint Multipath Routing Algorithmwith Time Slice Load-Balancing Congestion Control SchemeMHDMwTS is a two-phased routing protocol comprisingpath build-up and path acknowledgment phases It consists ofmultiple sources each of which has three build-up disjointpaths namely primary alternate and back-up paths Onactivation the source node requests to build up a route to thenearest hop neighbor which is the path build-up phase [19] Inthis phase step one has source activation in which therequested node for path building adds its number and time-stamp and sends to the least hop-count neighboring node-isgoes on until the least time latency sink having the informationneeded to construct the primary route is reached [75] In steptwo there is path extraction when the new package fromanother path arrives whereby the extracted path is comparedwith the primary path If the node is shared the package will berejected or an alternative path will be searched for to get abackup through comparison of the preceding two paths Phasetwo is path acknowledgment where we have the third step inwhich the sink returns the ACK packet to the sender with pathinformation having nodes and their time information after ithas computed it using the timestamp [1819]

Advantages

(i) -e protocol minimizes end-to-end latency(ii) It enhances congestion control

414 Cross-Layer Optimal Design (CLOD) Authors in [76]come up with a CLOD for scheduling at the data link layerrouting at the network layer and controlling congestion atthe transport layer with an assumption of fixed link capacity-rough congestion control energy efficiency is improvedTransport layer congestion at the nodes is minimized bycompressed sensing (CS) in which transmitted bits are re-duced whereas optimally allocating resources decreasescongestion on the links at the data link layer CLODpromises minimized computational complexity and betterperformance in light traffic scenarios [77] Generating andstoring CLOD is more efficient compared to Gaussianrandom matrices It prolongs network lifetime and savesenergy

Advantages

(i) It does congestion control subsequently increasingenergy efficiency

28 Journal of Computer Networks and Communications

(ii) It reduces computational complexity(iii) It improves performance in light traffic scenarios

415 Challenges and Opportunities of the Cross-Layer Design(CLD)

(i) Different layers in a real network assume differentfunctions andor services A layer only communi-cates with its neighboring layers -e layered modeldegrades system performance because of many

features in wireless transmission QoS parameters inthe layers of WMSNs are shared by routing pro-tocols to optimize the performance Neverthelessthere is a need to design cross-layer models forincreased efficiency in routing [1]

(ii) -e physical layer plays a highly vital role in CLDRate adaptation and channel allocation take place atthe physical layer via signal processing to enhanceQoS End-to-end performance in wireless media isaffected by the changes therein given the effect on

Table 4 A summary of some of the reviewed cross-layer models

Protocol Layers Aims Comments QoS parameters

SDRCS MACPHY Routing real-time traffic Transmits multimedia traffic viacross-layer packet forwarding

-roughputlatency

SUIT Applicationtransport Transmission of non-real-time video Uses fuzzy logic to regulate

congestionTransmission rate

delay

CLASS Any twoDesign serves as a framework fordifferent implementations ofdifferent application scenarios

Scalable very efficient flexible andhas very low propagation delay

Propagation delayjitter

NLQS PHYMACnetwork

Permits packet-level servicedifferentiation as a function of

throughput packet error rate anddelay

Network layer QoS is enhanced -roughput delaypacket error rate

CMVT Applicationnetwork

Encodes video streams to video dataframes via MPEG-4 encoding anddoes route discovery and data

transfer

Greedy forwarding and rollback areused to find source-to-sink paths Energy

CoopMAC PHYMAC Offers spatial diversity among thethree nodes

Helper ID signal finds out howfeasible cooperative communication

is

-roughput ratedelay

M-CMAC PHYMAC Increases end-to-end throughputand packet delivery ratio

Euclidean distances between nodesare calculated and converted to the

data rate for a link

-roughputpacket delivery

ratio

CBCR PHYMACMinimizes control overhead andtime consumed in establishing thecooperative paths than M-CMAC

Energy consumption is moreuniformly distributed in a network

enhancing network lifetime

-roughputpacket deliveryratio energy

MAC-PHY MACPHYMeant to enable PHY layer

cooperation and maximize gains ofcooperation at the MAC layer

Leverages both spatial diversity andcoding gain -roughput delay

MAC-centric MACAPPL

Meant to support QoS needs in newvideo apps and enable MAC layer

differentiation for H264partitioning

Targets multimedia applications byusing the MPEG-4 scheme

Delay packet lossrate

ACFEC MACnetwork(UDP)

Meant to enhance the quality ofvideo streaming over 80211WLANs

and overcome packet losses

Adjusts redundancy rates toovercome channel fluctuations and

detect and reduce packet lossPacket loss rate

MHDMwTS mdash

Meant to provide reliable datatransfer with the multipath routingand load-balancing congestioncontrol method in WMSNs

More reliable than basic routingschemes for transport multimedia

data

Latency packagetransmit rate

BCFL mdash

Introduces dispersion into fuzzylogic-based routing and sets every

cross-layer parameter with adynamic weight

Can be used to select a CH in cluster-based routing protocols andproposes a dispersion formula

Node utilitydispersion

CLODDatalinknetworktransport

Prolongs network lifetime andachieves congestion control anddesigned for lightly loaded WSNs

Assumes fixed link capacity andintegrates compressed sensing

technology

-roughputaverage energy CS

error ratio

Journal of Computer Networks and Communications 29

the purposefulness of the protocols at the networklayer CLD offers solutions to conserving powerminimizing energy and controlling both the flowand congestion in the network as well as fault tol-erance so it is an opportunity for designers con-sidering other layers We further desire to developCLDs that shall make an impact on network oper-ation and get close attention [18]

(iii) CLD interface standardization the architecturemust deliver module functionality although thereexist some queries amongst modules about possibleinterfaces predicted by the necessity to share in-formation among remote protocol layers Technicalchallenges include developing designing andstandardizing of cross-layer interfaces and algo-rithms that satisfy the cross-layer optimization re-quirements among protocol layers

Table 4 summarizes these cross-layer models with a fewcomments

5 Future Research Direction

51 At theMACLayer In WMSNs achieving duty cycling istricky since video traffic is volatile Listening to adjacentnodes wastes energy We need to switch states from awake tosleep -is requires further research

Self-organizing networks and green communicationalthough most CRWSN energy-saving algorithms aredesigned to reduce node transmission power energyconsumption is inevitable during the operational mode ofBSs because of internal processing Scholars are trying withpolicy-makers to encourage a shift to greenWSNs to reduceoperational costs and carbon footprints With self-orga-nizing CRWSNs for automatic switch-off when idle energyis saved which calls for cognitive switching algorithms Sointegrating cognitive switching with radio resource allo-cation in CRWSNs is a promising research area [78]

52 At the Application Layer Considering a WMSN ar-chitecture end users access the network via the BS to andfrom which they send their sensed data We desire to getguaranteed QoS and energy efficiency by designing a newarchitecture in the form of hardware and software whichneeds to be investigated further for a fix of the same [1]

QoS and energy efficiency are very crucial in WMSNsand mostly real-time apps that require guaranteed band-width and throughput in their network lifetime Mostprotocols ignore base station (BS) mobility and WMSNnodes Traffic management telemedicine and battlefieldsurveillance apps require mobile nodes or BSs so designingdynamic routing protocols to be adaptable in these cir-cumstances is necessary

53 Cross-Layer Security of data transmitted on WMSNseg in military surveillance and e-commerce is key withQoS and energy efficiency WMSNs are vulnerable to attacks

like worms and sinkhole Moreover their computationpower is not that high complicating implementation ofstrong secure protocols -is needs further attention

For fairness and priority issues in CRWSNs nodes areassigned diverse priorities depending on importance andurgency of their data For better network performance thereis a need to develop fair resource allocation mechanismswith priority-based fairness in sensors-ere is still potentialin this research area [78]

We must develop multiobjective adaptive protocols tooptimize QoS metrics involved in routing for best trans-mission results to get a trade-off between different opti-mization metrics We need to design standardized cross-layer algorithms which are able to meet cross-layer opti-mization standards

6 Conclusion

We have reviewed the different QoS strategies for WSNs inthe context of IoT from the MAC layer and application layeras well as the cross-layer paradigm For the MAC layer wehave reviewed protocols for WSNs such as the hybrid of thecontention-free and contention-based MAC protocols andthose for WMSNs We further reviewed a number of ap-plication layer protocols including many machine-to-ma-chine requestresponse and publishsubscribe protocols Forsystem optimization cross-layer QoS strategies are veryimportant in wireless communication We have reviewed anumber of cross-layer strategies For all categories reviewedchallenges and opportunities are discussed Finally somepossible future directions are discussed for research andapplication showing a promising potential for future re-search in this relatively new area especially as more WMSNapplications emerge recently

Conflicts of Interest

-e authors hereby declare no conflicts of interest regardingthe publication of this work

Acknowledgments

-is research was funded by the National Natural ScienceFoundation of China (Nos 61602398 61672447 and61711540306)

References

[1] V Bhandary A Malik and S Kumar ldquoRouting in wirelessmultimedia sensor networks a survey of existing protocolsand open research issuesrdquo Journal of Engineering vol 2016Article ID 9608757 27 pages 2016

[2] I F Akyildiz T Melodia and K R Chowdhury ldquoWirelessmultimedia sensor networks applications and testbedsrdquoProceedings of the IEEE vol 96 no 10 pp 1588ndash1605 2008

[3] M Iqbal M Naeem A Ahmed M Awais A Anpalagan andA Ahmad ldquoSwarm intelligence based resource managementfor cooperative cognitive radio network in smart hospitalsrdquoWireless Personal Communications vol 98 no 1 pp 571ndash5922018

30 Journal of Computer Networks and Communications

[4] S A Alvi B Afzal G A Shah L Atzori and W MahmoodldquoInternet of multimedia things vision and challengesrdquo AdHoc Networks vol 33 pp 87ndash111 2015

[5] Q F Hassan A U R Khan and S A Madani Internet ofings Challenges Advances and Applications CRC PressTaylor amp Francis Group A Chapman amp Hall Group BocaRaton FL USA 2018

[6] Q Wang Y Zhao W Wang et al ldquoMultimedia IoT systemsand applicationsrdquo in Proceedings of the 2017 Global Internet ofings Summit (GIoTS) vol 2 Geneva Switzerland June 2017

[7] A Aslam and E Curry ldquoTowards a generalized approach fordeep neural network based event processing for the internet ofmultimedia thingsrdquo IEEE Access vol 6 pp 25573ndash25587 2018

[8] M Jridi T Chapel V Dorez L Bougeant and A Le BotlanldquoSoC-based edge computing gateway in the context of theinternet of multimedia things experimental platformrdquoJournal of Low Power Electronics and Applications vol 8no 1 p 1 2018

[9] H Noura A Chehab L Sleem andM N Rapha ldquoOne roundcipher algorithm for multimedia IoT devicesrdquo MultimediaTools and Applications vol 77 no 14 pp 18383ndash18413 2018

[10] T Balan D Robu and F Sandu ldquoMultihoming for mobileinternet of multimedia thingsrdquo Mobile Information Systemsvol 2017 Article ID 6965028 16 pages 2017

[11] H Shen and G Bai ldquoRouting in wireless multimedia sensornetworks a survey and challenges aheadrdquo Journal of Networkand Computer Applications vol 71 pp 30ndash49 2016

[12] A Ahmad M H Rehmani H Tembine O A Mohammedand A Jamalipour ldquoIEEE access special section Editorial optimization for emerging wireless Networks IoT 5G andsmart grid communication networksrdquo IEEE Access vol 5pp 2096ndash2100 2017

[13] B A Muzakkari M A Mohamed M F A Kadir andZ Mohamad ldquoRecent advances in energy efficient-QoS awareMAC protocols for wireless sensor networksrdquo InternationalJournal of Advanced Computer Research vol 8 no 38pp 212ndash228 2018

[14] M A Yigitel O D Incel and C Ersoy ldquoQoS-aware MACprotocols for wireless sensor networks a surveyrdquo ComputerNetworks vol 55 no 8 pp 1982ndash2004 2011

[15] A M Abbas and O Kure ldquoQuality of Service in mobile ad hocnetworks a surveyrdquo International Journal of Ad Hoc andUbiquitous Computing vol 6 no 2 p 75 2010

[16] M Q Shatnawi ldquoApplication layer protocols for the internetof things a surveyrdquo in Proceedings of the InternationalConference on Engineering amp MIS (ICEMIS) Agadir Mo-rocco September 2016

[17] V Karagiannis P Chatzimisios F Vazquez-gallego andJ Alonso-zarate ldquoA survey on application layer protocols forthe internet of things research motivationrdquo InternationalJournal of Advanced Research in Computer Science vol 8no 3 pp 1ndash10 2015

[18] M S Bernard T Pei Z Li and K Li ldquoQoS strategies forwireless multimedia sensor networks in the context of IoTrdquoLecture Notes of the Institute for Computer Sciences SocialInformatics and Telecommunications Engineering vol 4pp 228ndash253 2019

[19] A AlAmri and M Abdullah ldquoCross-layer quality of serviceprotocols for wireless multimedia sensor networksrdquo in Pro-ceedings of the International Conference on CommunicationManagement and Information Technology pp 649ndash658Cosenza Italy April 2017

[20] J Guo L Sun and R Wang ldquoA cross-layer and multipathbased video transmission scheme for wireless multimedia

sensor networksrdquo Journal of Networks vol 7 no 9pp 1334ndash1340 2012

[21] A Ksentini A Naimi and M Gueroui ldquoToward an im-provement of H264 video transmission over IEEE 80211ethrough a cross-layer architecturerdquo IEEE CommunicationsMagazine vol 44 no 1 pp 107ndash114 2006

[22] W Ye J Heidemann and D Estrin ldquoAn energy-efficientMAC protocol for wireless sensor networksrdquo in Proceedings ofthe 21st International Annual Joint Conference of the IEEEComputer and Communications Societies (INFOCOM 2002)New York NY USA June 2002

[23] J Polastre J Hill and D Culler ldquoVersatile low power mediaaccess for wireless sensor networks categories and subjectdescriptorsrdquo in Proceedings of the Second InternationalConference on Embedded Networked Sensor Systems Sen-Sysrsquo04 pp 95ndash107 Baltimore MA USA November 2004

[24] T Danmanee K N Nakorn and K Rojviboonchai ldquoCU-MAC a duty-cycle MAC protocol for internet of things inwireless sensor networksrdquo ECTI Transactions on ElectricalEngineering Electronics and Communications vol 16 no 2pp 30ndash43 2018

[25] T Van Dam and K Langendoen ldquoAn adaptive energy-efficientMAC protocol for wireless sensor networksrdquo in Proceedings ofthe 1st International Conference on Embedded NetworkedSensor Systems Los Angeles CA USA November 2003

[26] P Xie and J Cui ldquoR-MAC an energy-efficient MAC protocolfor underwater sensor networksrdquo in International Conferenceon Wireless Algorithms Systems and Applications (WASA2007) Chicago IL USA August 2007

[27] A Rahim N Javaid M Aslam U Qasim and Z A KhanldquoAdaptive-reliable medium access control protocol forwireless body area networksrdquo in Proceedings of the AnnualIEEE Communications Society Conference on Sensor Meshand Ad Hoc Communications and Networks (SECON) SeoulKorea June 2012

[28] H S Kwak S Ullah D H Kwak and C H Lee ldquoA powerefficient MAC protocol for wireless body area networksrdquoJournal of the Korean Institute of Intelligent Systems vol 8no 6 pp 131ndash140 2009

[29] C Omeni A J Burdett and M Park ldquoEnergy efficientmedium access protocol for wireless medical body area sensornetworksrdquo in Proceedings of the 4th IEEEEMBS InternationalSummer School and Symposium on Medical Devices andBiosensors pp 29ndash32 Cambridge UK August 2007

[30] L Sitanayah C J Sreenan and K N Brown ldquoER-MAC ahybrid MAC protocol for emergency response wireless sensornetworksrdquo in Proceedings of the Fourth International Con-ference on Sensor Technologies and Applications pp 244ndash249Mestre Italy July 2010

[31] Y Sun and D B Johnson ldquoRI-MAC a receiver-initiatedasynchronous duty cycle MAC protocol for dynamic trafficloads in wireless sensor networksrdquo in Proceedings of the 6thACM conference on Embedded network sensorsystemsmdashSenSysrsquo08 pp 1ndash14 Raleigh NC USA November2008

[32] M Buettner G V Yee E Anderson and R Han ldquoX-MAC ashort preamble MAC protocol for duty-cycled wireless sensornetworksrdquo in Proceedings of the 4th International Conferenceon Embedded Networked Sensor Systems SenSys 2006pp 307ndash320 Boulder Colorado USA October 2006

[33] S Boulfekhar and M Benmohammed ldquoSynchronous receiverinitiated MAC protocol for long-lived sensor networksrdquoComputers amp Electrical Engineering vol 40 no 2 pp 504ndash516 2014

Journal of Computer Networks and Communications 31

[34] S Mehta and K S Kwak ldquoH-MAC a hybrid Mac protocol forwireless sensor networksrdquo International Journal of ComputerNetworks amp Communications vol 2 no 2 pp 108ndash117 2010

[35] I Rhee A Warrier M Aia J Min and M L Sichitiu ldquoZ-MAC a hybrid MAC for wireless sensor networksrdquo IEEEACM Transactions on Networking vol 16 no 3 pp 511ndash5242008

[36] S Arshad A Al-sadi and A Barnawi ldquoZ-MAC performanceevaluation and enhancementsrdquo Procedia Computer Sciencevol 21 pp 485ndash490 2013

[37] Y Liu I Elhanany and H Qi ldquoAn energy-efficient QoS-aware media access control protocol for wireless sensornetworksrdquo in Proceedings of the IEEE International Confer-ence on Mobile ad hoc and Sensor Systems pp 9ndash11 Wash-ington DC USA December 2005

[38] J Decotignie ldquoWiseMAC an ultra low power MAC protocolfor the downlink of infrastructure wireless sensor networksrdquoin Proceedings of the Ninth International Symposium onComputers And Communications pp 244ndash251 AlexandriaEgypt June 2004

[39] R A Rashid ldquoDevelopment of energy aware TDMA-basedMAC protocol for wireless sensor network systemrdquo EuropeanJournal of Scientific Research vol 30 no 4 pp 571ndash578 2009

[40] H P Van Hoesel ldquoA lightweight medium access protocol(LMAC) for wireless sensor networksrdquo in Proceedings of theInternational Workshop on Networked Sensing Systems Bal-timore MD USA November 2004

[41] C Li P Wang H Chen andM Guizani ldquoA cluster based on-demand multi-channel MAC protocol for wireless multi-media sensor networksrdquo in Proceedings of the IEEE In-ternational Conference on Communications pp 2371ndash2376Beijing China May 2008

[42] J L Sobrinho and A S Krishnakumar ldquoQuality-of-service inad hoc carrier sense multiple access wireless networksrdquo IEEEJournal on Selected Areas in Communications vol 17 no 8pp 1353ndash1368 1999

[43] J P John ldquoProviding MAC QoS for multimedia traffic in80211e based multi-hop ad hoc wireless networksrdquo ComputerNetworks vol 51 no 1 pp 153ndash176 2007

[44] S Lofty and N Padmavati ldquoA survey on mobility basedprotocols in WSNsrdquo in Proceedings of the InternationalConference Communication and Manufacturing pp 12ndash15Hangzhou China July 2014

[45] S Deng L Shen and J Li ldquoMobility-based clustering pro-tocol for wireless sensor networks with mobile nodesrdquo IETWireless Sensor Systems vol 1 no 1 pp 39ndash47 2011

[46] M Khalil A Khalid F U Khan and A Shabbir ldquoA review ofrouting protocol selection for wireless sensor networks insmart citiesrdquo in Proceedings of the 24th Asia-Pacific Confer-ence on Communications (APCC) Ningbo China November2014

[47] O Message Q Telemetry and T Mqtt Committee Specifi-cation Draft 02 April 2014

[48] M Collina M Bartolucci A Vanelli-coralli andG E Corazza ldquoInternet of things application layer protocolanalysis over error and delay prone linksrdquo in Proceedings ofthe 7th Advanced Satellite Multimedia Systems Conference andthe 13th Signal Processing for Space CommunicationsWorkshop Livorno Italy September 2014

[49] A Talaminos-barroso M A Estudillo-valderrama L M RoaJ Reina-tosina and F Ortega-ruiz ldquoA machine-to-machineprotocol benchmark for e health applicationsmdashuse case re-spiratory rehabilitationrdquo Computer Methods and Programs inBiomedicine vol 129 pp 1ndash11 2016

[50] V Goyal O Pandey A Sahai and B Waters ldquoAttribute-based encryption for fine-grained access control of encrypteddatardquo in Proceedings of the 13th ACM Conference on Com-puter and Communications Security pp 89ndash98 AlexandriaVA USA October 2006

[51] M Singh M A Rajan V L Shivraj and P BalamuralidharldquoSecure MQTT for internet of things (IoT)rdquo in Proceedings ofthe Fifth International Conference on Communication Systemsand Network Technologies Gwalior India April 2015

[52] A Stanford-Clark and H L Truong MQTT for Sensor Net-works (MQTT-SN) Protocol Specification httpmqttorgnewwp-contentuploads200906MQTT-SN_spec_v12pdf 2013

[53] RFC ldquoExtensible messaging and presence protocol (XMPP)rdquoin Proceedings of the IETF pp 1ndash9 San Diego CA USAAugust 2005

[54] S Vinoski and I Technologies ldquoQueuing protocolrdquo IEEEInternet Computing vol 10 no 6 pp 87ndash89 2006

[55] J L Fernandes I C Lopes J J P C Rodrigues and S UllahldquoPerformance evaluation of restful web services and AMQPprotocolrdquo in Proceedings of the Fifth International Conferenceon Ubiquitous and Future Networks (ICUFN) pp 2ndash7 DaNang Vietnam July 2013

[56] M Asim ldquoA survey on application layer protocols for internetof things (IoT)rdquo International Journal of Advanced Researchin Computer Science vol 8 no 3 pp 996ndash1000 2017

[57] K An A Gokhale D Schmidt S Tambe P Pazandak andG Pardo-castellote ldquoContent-based filtering discovery pro-tocol (CFDP) scalable and efficient OMG DDS discoveryprotocolrdquo in Proceedings of the 8th ACM InternationalConference on Distributed Event-Based Systems New YorkNY USA July 2014

[58] E Web What is a web service pp 52ndash57 2010[59] M Laine RESTful Web Services for the Internet of ings

pp 2ndash4 2012 httpmediaTkk[60] IETF -e WebSocket protocol pp 1ndash71 2011[61] V Pimentel ldquoCommunicating and displaying real-time data

with websocketrdquo IEEE Internet Computing vol 16 no 4pp 45ndash53 2012

[62] Wikipedia 2019 httpsenwikipediaorgwikiStreaming_Text_Oriented_Messaging_Protocol2019

[63] K Raghavendra and S Nireshwalya ldquoApplication layer se-curity issues and its solutionsrdquo IJCSET vol 2 no 6pp 1266ndash1269 2012

[64] Y Xue B Ramamurthy andM C Vuran ldquoSDRCS a service-differentiated real-time communication scheme for eventsensing in wireless sensor networksrdquo Computer Networksvol 55 no 15 pp 3287ndash3302 2011

[65] C Sonmez S Isik M Y Donmez O D Incel and C ErsoyldquoSUIT a cross layer image transport protocol with fuzzy logicbased congestion control for wireless multimedia sensornetworksrdquo in Proceedings of the 5th International Conferenceon New Technologies Mobility and Security (NTMS) pp 1ndash6IEEE Istanbul Turkey May 2012

[66] T Melodia and I Akyildiz ldquoCross-layer QoS-aware com-munication for ultra wide band wireless multimedia sensornetworksrdquo IEEE Journal on Selected Areas in Communica-tions vol 28 no 5 pp 653ndash663 2010

[67] Q Wang and M A Abu-Rgheff ldquoCross-layer signalling fornext-generation wireless systemsrdquo in Proceedings of the 2003IEEE Wireless Communications and Networking 2003WCNC 2003 vol 2 pp 1084ndash1089 IEEE New Orleans LAUSA March 2003

[68] P Liu Z Tao S Narayanan T Korakis and S PanwarldquoCoopMAC a cooperative MAC for wireless LANsrdquo IEEE

32 Journal of Computer Networks and Communications

Journal on Selected Areas in Communications vol 25 no 2pp 340ndash354 2007

[69] H Shan H T Cheng and W Zhuang ldquoCross-layer co-operative MAC protocol in distributed wireless networksrdquoIEEE Transactions onWireless Communications vol 10 no 8pp 2603ndash2615 2011

[70] F Liu T Korakis Z Tao and S Panwar ldquoAMAC-PHY cross-layer protocol for wireless Ad-Hoc networksrdquo in Proceedingsof the 2008 IEEE Wireless Communications and NetworkingConference pp 1792ndash1797 IEEE Las Vegas NV USA March2008

[71] L Jacob and H R Shamna ldquoEfficient cooperative MAC androuting in wireless networksrdquo Transactions on Networks andCommunications vol 3 no 5 2015

[72] S Rao and K Shama ldquoCross layer protocols for multimediatransmission in wireless networksrdquo International Journal ofComputer Science amp Engineering Survey vol 3 no 3pp 15ndash28 2012

[73] L Han S Park S Kang and H Peter ldquoAn adaptive cross-layer FEC mechanism for video transmission over 80211WLANsrdquo in Proceedings of the 2009 International Conferenceon Internet pp 209ndash215 Las Vegas NV USA December2009

[74] N Li J Martınez and V H Dıaz ldquo-e balanced cross-layerdesign routing algorithm in wireless sensor networks usingfuzzy logicrdquo Sensors vol 15 no 8 pp 19541ndash19559 2015

[75] G Sun J Qi Z Zang and Q Xu ldquoA reliable multipathrouting algorithm with related congestion control scheme inwireless multimedia sensor networksrdquo in Proceedings of the2011 3rd International Conference on Computer Research andDevelopment vol 4 pp 229ndash233 IEEE Shanghai ChinaMarch 2011

[76] M Li Y Jing and C Li ldquoA robust and efficient cross-layeroptimal design in wireless sensor networksrdquoWireless PersonalCommunications vol 72 no 4 pp 1889ndash1902 2013

[77] J Yan M Zhou and Z Ding ldquoRecent advances in energy-efficient routing protocols for wireless sensor networks areviewrdquo IEEE Access vol 4 pp 5673ndash5686 2016

[78] A Ahmad S Ahmad M H Rehmani and N U Hassan ldquoAsurvey on radio resource allocation in cognitive radio sensornetworksrdquo IEEE Communications Surveys amp Tutorials vol 17no 2 pp 888ndash917 2015

Journal of Computer Networks and Communications 33

International Journal of

AerospaceEngineeringHindawiwwwhindawicom Volume 2018

RoboticsJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Active and Passive Electronic Components

VLSI Design

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Shock and Vibration

Hindawiwwwhindawicom Volume 2018

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawiwwwhindawicom

Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Control Scienceand Engineering

Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Journal ofEngineeringVolume 2018

SensorsJournal of

Hindawiwwwhindawicom Volume 2018

International Journal of

RotatingMachinery

Hindawiwwwhindawicom Volume 2018

Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Navigation and Observation

International Journal of

Hindawi

wwwhindawicom Volume 2018

Advances in

Multimedia

Submit your manuscripts atwwwhindawicom

Page 2: ReviewArticle - downloads.hindawi.com · ReviewArticle QoSStrategiesforWirelessMultimediaSensorNetworksinthe ContextofIoTattheMACLayer,ApplicationLayer,and Cross-LayerAlgorithms MuwongeSsajjabbiBernard

systems traffic monitoring target tracking intrusion de-tection systems telemedicine for advanced health care etcIqbal et al for instance developed an efficient power al-location and personal wireless hub (PWH) placementstrategy for maximizing the data rate under the cognitiveradio interference constraint that has an efficacy with lowcomplexity to facilitate paramedic staff in next-generationhealth care facilities using multimedia in smart hospitals [3]

11 Classification of WSNs (Wireless Sensor Networks)Different types of WSNs include terrestrial WSNs un-derground WSNs underwater WSNs multimedia WSNsand mobile WSNs

111 Terrestrial WSNs -ese can communicate with basestations efficiently and comprise 100 s to 1000 s of wirelesssensor nodes deployed in the unstructured (ad hoc) orstructured (preplanned) style In the ad hoc mode nodes arerandomly distributed in the target region that is droppedfrom a fixed plane -e structured mode considers optimalplacement grid placement and 2D and 3D placementmodels Energy is conserved by use of low duty cycling delayminimization and optimal routing

112 Underground WSNs -ese are more expensive thanterrestrial WSNs in terms of deployment maintenanceequipment cost and careful planning -e WSNs comprisemany sensor nodes hidden in the ground to monitor un-derground conditions To relay information from the sensornodes to the base station additional sink nodes are locatedabove the ground -ese are highly affected by attenuationand signal loss and are very difficult to charge

113 Underwater WSNs -ese have multiple sensor nodesand vehicles deployed under water Autonomous un-derwater vehicles are used to gather data from the sensorLong propagation delay and bandwidth and sensor failuresare a major challenge here

114 Multimedia WSNs (WMSNs) -ese have been pro-posed to enable tracking and monitoring of events in theform of multimedia such as imaging video and audio-esenetworks comprise low-cost sensor nodes equipped withmicrophones and cameras -e nodes are interconnectedwith each other over a wireless connection for data com-pression retrieval and correlation -e challenges withWMSNs include high energy consumption data processingand compressing techniques and high bandwidth re-quirements for proper and easy content delivery

115 Mobile WSNs -ey comprise a collection of sensornodes which can move on their own and can interact with thephysical environment Mobile nodes have the ability to com-pute sense and communicate -e mobile WSNs are moreversatile than static sensor networks -e advantages ofWMSNs over the static WSNs include better and improved

coverage better energy efficiency and superior channelcapacity

12 e Architecture of a WMSN -e WMSN architecturehas the following 3 subdivisions

(i) Single-tier flat architecture this consists of homo-geneousmultimedia nodes which are able to executeany function to the sink via multihop routes

(ii) Single-tier clustered architecture this consists ofheterogeneous nodes passing sensed information tothe cluster head for processing

(iii) Multitier architecture this too consists of hetero-geneous nodes and does object sensing and targetcapturing and tracking

Figure 1 shows a typical example of a WMSN archi-tecture Figure 1(a) shows a single-tier flat and homogeneousarchitecture in which sensors having the same physicalabilities are utilized In Figure 1(b) we have a single-tierclustered and heterogeneous architecture having nodes withdifferent physical capabilities eg multimedia nodes andscalar nodes -e multitier clustered and heterogeneousarchitecture (Figure 1(c)) has several layers of nodes havingdiverse types and processing tasks per layer [2] WMSNs arewidely deployed to offer infrastructural support and sensoraccessibility rendering them suitable for Internet of multi-media things (IoMT) transmission [4] Most applications(apps) in the IoMT eg wearable devices utilize theWMSNtechnology

13e Internet of ings (IoT) In the IoTtoday we visualizea situation whereby smart devices connect to a single net-workmdashthe Internet It is becoming more popular than it hasbeen because of the multitude of connected devices availablerecently A number of IoT- and IoMT-based applications arecoming lately while attracting enormous attention

-ese applications include smart cities smart vehicleshomes factories (Figure 2) and GPS tracking devices [5 6]and have attracted many inventions in the Americas AsiaEurope etc A number of commercial and military appli-cations come up because of introduction of multimediaobjects in transmission of data such as remote patientmonitoring in telehealth and telemedicine and trafficmanagement systems enhanced by smart video camerasamong others [4] -is calls for an upgrade in functionalityof IoTsystems to the IoMT Figure 2 shows an illustration ofthe IoT architecture which could be aided by RFID opticaltags QR codes Bluetooth low energy Wi-Fi direct andLTE-Advanced among others Most of the scholars em-phasize improvement of efficiency for handling a lot of real-time information (info) but ignore multimedia transmissionaspects [7 8] -e direction of research is shifting from theordinary IoT to the multimedia-based IoT because of theneed to enable smart devices to efficiently observe sense andunderstand their environment through multimedia data[9 10] hence resulting in the emergence of the newer field ofInternet of multimedia things (IoMT)

2 Journal of Computer Networks and Communications

IoMT is ldquothe IoT-based paradigm that enables objects toconnect and exchange structured and unstructured data withone another to enable multimedia-based services and appli-cationsrdquo [3] ere is a need for vast processing powermemory and bandwidth for high QoS when transmittingmultimedia data in the IoMT in comparison with scalar datain a traditional IoT IoT system functionality shouldtherefore be upgraded to the IoMT We compare the two asdiscussed in [7]

(i) e IoT has standardized communication pro-tocols whereas the IoMTrsquos protocols arenonstandardized

(ii) In terms of QoS the IoT requires low bandwidthwhereas the IoMT requires higher bandwidth

(iii) e IoMT transmits heterogeneous multimediadata whereas IoT data transmitted have limitedheterogeneity

(iv) IoT sensor nodes consume less energy than IoMTsensor nodes

(v) IoT devices are deployed in application-dependentRFID tags but the IoMT is in video and audiosensors

(vi) In terms of service composition the IoMT has noavailable specialized middleware whereas the IoThas specialized service-oriented architecture-basedand event-based middleware

For improved QoS best eort services and higher en-ergy eciency in IoMT networks and applications enor-mous multimedia-supported routing is gaining ground inthe research arena in the WMSN area in routing protocolsalgorithms and techniques based on network architecturesand application requirements [11] Nevertheless accordingto Ahmad et al [12] the enormous resource-constrainedheterogeneous environment of the IoT challenges its ex-pansion and deployment is is because most existing IoTapps comprise overlaid deployments of wireless sensor andactuator networks in which apps cannot interact with eachother or share and reuse the few available resources Inaddition to that ecient sensing and propagation of info

C

Users

WMSN

Image sensor

CM

CM

Audio sensor

Scalar sensor

CC

Internet

Laptops

Laptop

Wireless gateways

Storage

Sink

Scalar sensor

Video sensor

Figure 1 Architecture of the WMSN (a) single-tier at and homogeneous (b) single-tier clustered and heterogeneous (c) multitierheterogeneous

Journal of Computer Networks and Communications 3

and swift response to changes in the physical world arechallenging IoT requirements [12]

14 What Has Been Done In wireless communicationsystems routing is one issue that is quite a challenge yetthere are very few surveys available especially regardingWMSN protocols [1] and more so the MAC layer appli-cation layer and cross-layer protocols although some sur-veys have been published on WSNs To the best of ourknowledge no survey combines all these aspects -is makesour survey important to add to the existing literature -eauthors in [1] discuss a number of WMSN routing tech-niques and the properties and shortcomings of theseHowever they do not discuss any application layer tech-niques Muzakkari et al [13] survey some recent WSNcontention-based scheduling-based and hybrid MACprotocols whereby they focus on the underlying principlesadvantages limitations and their applications But much asthey only focus on MAC protocols they still leave out anumber of them in their survey Yigitel et al [14] carry out asurvey on QoS-aware MAC protocols for WSNs -ey re-view the QoS challenges and views for WSNs study the QoSmechanisms categorize the state-of-the-art QoS-awareMAC protocols and also talk about the advantages anddrawbacks of the same But being a 2011 release a number ofkey protocols developed since then to date are not presented

therein Abbas and Kure [15] review various methods forQoS provisioning at the levels of routing MAC layer andcross-layer including the schemes for admission control andscheduling for QoS provisioning as well as the problems andchallenges involved But they leave out the protocols from2010 to date Shatnawi [16] and Karagiannis et al [17] reviewthe application layer protocols for the Internet of things Intheir survey they address a number of application layerprotocols which are employed for IoT for affirming a re-liable tie among objects and things -ey evaluate thereviewed protocols in terms of architecture communicationmodel security and QoS as well as the weaknesses andstrengths per reviewed protocol However they only con-centrate on application layer protocols and still leave outsome critical protocols In a recent survey Bernard et al [18]discuss a number of cross-layer QoS strategies for IoT Butbecause of space constraint we leave out some criticalprotocols and besides this paper does not talk about ap-plication layer and MAC layer protocols AlAmri andAbdullah [19] carry out a survey on cross-layer QoS pro-tocols for WMSNs in which they state that the cross-layerarchitecture is a novel idea that brings together a number oflayers for enabling integration and exchange of informationin between them with higher efficiency compared to thetraditional layered model -ey discuss and compare theavailable cross-layer WMSN protocols that cross the uses ofadjacent or nonadjacent layers But they concentrate on

Database

MMS user databases

MMS relayserver

Home location register

External servers(eg e-mail and fax)

MMS value-added servicesservers (premium content

e-mails web )

Foreign MMSC(MMS relay server)

MM

S us

er ag

ents

Internet

Figure 2 Example of the multimedia service architecture in the context of IoT

4 Journal of Computer Networks and Communications

cross-layer protocols and leave out the MAC layer andapplication layer protocols and also some cross-layer pro-tocols are left out -e surveys that have been reviewed arecompared in Table 1

-emajor contributions that make our survey importantare outlined below

15 Importance of the Survey

(i) We discuss and draw a comparison between the IoTand the IoMT by discussing the differences betweenthe two

(ii) We further discuss the newer paradigm of WMSNsand compareWSNs andWMSNs in our discussion

(iii) We make a comprehensive survey of the recentstate-of-the-art routing protocols focusing on theMAC layer the application layer and the cross-layer paradigm

(iv) We discuss the WMSN protocols as well as theWMSN-enabled protocols that have been de-veloped in the recent studies by several scholars

(v) In some of the reviewed protocols we talk about therelative advantages and disadvantages of some ofthe routing approaches for readers to be in positionto comprehend different techniques and so choosethe most appropriate technique depending on theuser requirements

(vi) In all these we cover the challenges and oppor-tunities existing for the discussed categories

(vii) We consider the possible future research trendsbefore concluding this paper

16 Challenges

(i) Since WSNs and WMSNs are usually distributed andad hoc networks their nodes are powered by batteriesmaking energy optimization a challenge that needs tobe addressed when designing their routing protocols

(ii) Considering WSNs their routing protocols do notput into consideration multimedia applicationswhich need a lot of bandwidth and processingenergy and must be transmitted in real time withutmost fidelity Extra challenges in routing areimposed in WMSNs in deployment of heteroge-neous sensors since there could be audio video andstill pictures in addition to scalar data duringtransmission Besides the heterogeneous datathereby handled all have different QoS re-quirements and with many different businessneeds different services with varying requirementsgive rise to a big challenge for the routing design

(iii) WMSNs have also got nonrechargeable batteriesplacing a limitation on the energy of the nodes yetmultimedia applications consume a lot of energyQuite a number of complications and limitingfactors exist practically despite the fact that sensorscan be self-powering as enabled by energy

harvesting Predictive video encoding using MPEG-4 and other standards is also an energy-consumingprocess and so causes enormous degradation inrouting efficiency complicating the design ofWMSN protocols even more

(iv) -e heterogeneous nature of the network is anotherchallenge since there is a difference in the categoriesof nodes needed for communication so as to enablethe facilitation of effective data collection andprocessing as well as efficient transmission It is noteasy to have a uniform communication protocolplatform because of different functionalities unlikethe case of the conventional WSNs compared to theWMSNs

(v) -ere must be a trade-off between the energy effi-ciency and the multimedia QoS when deciding onthe route taken Data aggregation and compressionrouting protocols may be applicable for energy-saving though they can lead to intolerable delays inWMSNs Furthermore there might be networkcongestion due to the many-to-many and mutualinterference in wireless routes and the scarceWMSN resources In case a given node gets so manyhigh-rate streams it lowers the performance of thenetwork and raises the possibility of node failuredue to energy drainage

-is papermainly discusses QoS strategies forWSNs andWMSNs in the context of IoT from the MAC layer appli-cation layer and cross-layer perspective -is paper presentsa state-of-the-art survey on routing in WSNs and WMSNs-e rest of this paper is organized as follows In Section 2 wediscuss the QoS strategies at the MAC layer In Section 3 wediscuss the QoS strategies at the application layer Section 4discusses the cross-layer QoS strategies and in Section 5 wegive some future research directions before concluding thispaper in Section 6

2 QoS Strategies at the MAC Layer

-e medium access control (MAC) protocol is meant toregulate admittance to the shared medium as well astransmission reliability and efficiency through error cor-rection in wireless transmissions among others It is alsoresponsible for performance of framing addressing andflow control -e MAC protocol design affects energy effi-ciency too -is is majorly categorized into contention-freeprotocols contention-based protocols and a hybrid of these

21 MAC Protocols forWSNs Hybrid MAC protocols thesecombine contention-free and contention-based MAC pro-tocols on whose advantages they capitalize Some of these arediscussed below

211 S-MAC (Sensor MAC) -is is a MAC protocoldesigned for WSNs -e main components of S-MAC in-clude (i) periodic listen and sleep (ii) collision and over-hearing avoidance and (iii) message passing Energy

Journal of Computer Networks and Communications 5

consumption and self-configuration are controlled usingthese three techniques Nodes periodically go into the sleepmode to control the energy consumed in listening to idlechannels and virtual clusters are formed by neighbor nodesfor sleep schedule autosynchronization -e protocol em-ploys in-channel signaling when the radio is in the sleepmode and message passing is applied to decrease contentionlatency for store-and-forwardWSN applications during datatransmission [22] For adjacent nodes to solve their clockdrift S-MAC requires periodic synchronization To reducesynchronization errors exchanged timestamps are relativenot absolute

Advantages

(i) S-MAC has a longer listening period than the clockerror

(ii) It has more capacity to conserve energy than IEEE80211

(iii) It can do trade-off between energy and latencydepending on traffic conditions [22]

212 B-MAC (Berkeley MAC) -is is a carrier-sense MACprotocol designed for WSNs featuring a simplistic scalableimplementation that does network variations and is effectivein channel valuation [23] -e protocol has an interface withsome flexibility leading to very low-power usage and con-trols collisions through collision avoidance mechanismsleading to effective use of the frequency band It uses asampling scheme that is adjustable to minimize duty cyclingand idle listening so as to reduce power wastage Nodes arescheduled with their duty cycle each within which they sendthe preamble to the channel in case of availability of data fortransmission In different duty-cycle schedules channelusage is checked by the destination nodes that is if thepreamble found is long it will remain powered on until thedestination address is obtained and in case it is the target itwill either await senderrsquos data or go to sleep-e protocol haslower latency than synchronous MAC protocols but has aproblem of overhearing where the neighbor node getslengthy preambles though not the intended receiver hencewasting a lot of power in doing so [24] It outperforms anumber of protocols via reconfiguration feedback andbidirectional interfaces for upper layer services It can run at

low duty cycles and its apps cannot experience synchro-nization overhead [23]

Advantages

(i) -e protocol features a simplistic scalableimplementation that does network variations andis effective in channel valuation

(ii) It has an interface with flexibility leading to verylow-power usage

(iii) It controls collisions through collision avoidancemechanisms leading to effective use of the fre-quency band

(iv) It uses an adjustable sampling scheme to minimizeduty cycling and idle listening reducing powerwastage

Disadvantages

(i) It has a problem of overhearing where the neighbornode gets lengthy preambles though not theintended receiver hence wasting a lot of power indoing so

213 T-MAC -is is a contention-based MAC protocol forWSNs that dynamically with little difficulty via fine-grainedtimeouts adapts a listensleep duty cycle It uses this dutycycle by ending its active part so as to vary load in time andposition thereby decreasing the energy wastage on idlelistening [25] When starting an active period it uses a shortlistening window thereby improving the use of energy byS-MAC -rough adaptive duty cycling T-MAC savespower but at a cost of less throughput and enhanced delayBoth S-MAC and T-MAC under similar conditions performin the same way though in variable workloads S-MAC uses 5times more power than T-MAC But T-MAC is complex andnot scalable If its active window is decreased its snoopingcapability on nearby traffic and adaptation to variableconditions of the network is reduced [23] All nodes in thisprotocol are configured to wake up periodically listen toadjacent nodes and go back into the sleep mode till the nextframe as more messages come into the queue Collisionavoidance and reliable communication are ensured throughRequest-to-Send (RTS) Clear-to-Send (CTS) and Ac-knowledgment (ACK) messages among nodes [25]

Table 1 Comparison of this survey with existing surveys

Author(s) Year CharacteristicsBhandary et al [1] 2016 Routing strategies in WMSNs properties and limitationsMuzakkari et al [13] 2018 WSN contention-based scheduling-based and hybridMAC protocolsAhmad et al [12] 2011 QoS-aware MAC protocols for WSNsMuzakkari et al [13] 2010 QoS provisioning at the levels of routing MAC layer and cross-layerShatnawi [14] 2016 IoT application layer protocolsKaragiannis et al [15] 2015 Application layer protocols for the Internet of thingsBernard et al [16] 2019 Cross-layer QoS strategies for the IoT in WMSNsAlAmri and Abdullah [17] 2017 Cross-layer QoS protocols for WMSNsGuo et al [20] 2012 Cross-layer and multipath-based video transmission for WMSNs

Ksentini et al [21] 2006 Toward an improvement of H264 video transmission over IEEE80211e through a cross-layer architecture for WMSNs

6 Journal of Computer Networks and Communications

T-MAC design the major forms of energy consump-tion in T-MAC are through (i) idle listening (ii) col-lisions (iii) protocol overhead and (iv) overhearing Incontrast in the active period the node remains lis-tening and sending data If no activation event occursfor a certain time t then the active period ends -eactivation event might be data reception on a radiocommunication sensing on a radio and end oftransmission for the data packet of the node or ACKknowing that the neighbor has finished transmitting

Clustering and synchronization on waking up a nodebegins by listening to the channel If after a certain timethere is nothing heard it chooses a schedule fortransmitting Synchronization (SYNC) packetsappended with the time of the next frame startup Ifduring this time it hears another nodersquos SYNC packetit follows its schedule subsequently transmitting itsown SYNC Following both schedules enables it to havean activation incident at the beginning of both frames

RTS and t choice if a node does not get a CTS messageafter sending an RTS then either (i) there is a collisionpreventing the destination node from hearing the RTS(ii) RTS or CTS is overheard and destination node barredfrom responding or (iii) the destination node is in thesleep mode So if after time t there is no answer thesource node switches to the sleepmode It should only doso if none of its neighbors is still communicating A newinterval t can be triggered by a neighborrsquos RTS or CTSreception A node may be out of range to hear theinitiating RTS-e length t of the interval should be longenough to get the CTS packetrsquos beginning [25] and t isthus given by the comparison in the following equation

tgtC + R + T (1)

where C contention interval length RRTS packet lengthand T turn-around time

Advantages

(i) It saves power through adaptive duty cycling

Disadvantages

(i) It is unscalable and more complex(ii) It offers less throughput and enhanced delay

214 CU-MAC -is protocol supports IoT standards re-quiring request-response transmissions It employs a multi-channel mechanism for constant bidirectional transmissionof packets at low duty cycle to resolve hidden-terminal issuesAccording to Danmanee et al [24] CU-MAC is an SI-MACprotocol that is aimed at improving bandwidth of the channelby initiating connections and transmitting data throughdifferent channels using the multichannel approach

Protocol design the protocol has the following threeproperties that make it perform more efficiently withWSNs in the IoT

(i) It enables multichannel transmission using 9useable channels in IEEE 802154

(ii) It can enable linked data transmission using oneconnection which reduces additional overheadand delay

(iii) It can support continuous transmission of packets-is is enabled depending on how the bufferoverflow protection policy is set in the buffer stack-is stack has 2 parts (a) bidirectional buffer slotsfor enabling sensor nodes to support linked datatransmission and (b) normal buffer slots fornormal data queuing Buffer stack policies are setas follows

(a) A packet is only dropped in the buffer when itexpires

(b) Unless the packet is from a target node to which areceiver is sending the data the receiver deniespacket reception when buffer slots fill up

(c) -e bidirectional buffer slot is used for bi-directional buffer transfer

CU-MAC can reduce power wastage by choosing toemploy only a single channel for transmitting both controlpackets and data in case there is only one sender in the sameregion of coverage It can further enable multiple trans-missions through another usable channel if there are othersending nodes in this area [24] -e protocol has two phases(i) advertising phase in which a source node announces toneighbors before transmission and (ii) data transfer phasewhich commences when the source node receives a Ready-to-Receive Information packet from the destination node

Advantages

(i) It has a higher packet delivery ratio compared toother MAC protocols

(ii) It performs well in high traffic situations since itsduty-cycle is lower

(iii) It reduces power wastage by employing only asingle channel for transmitting both controlpackets and data

(iv) It can enable multiple transmissions through an-other usable channel if there are other sendingnodes

215 R-MAC -is is a reservation-based MAC protocoldesigned majorly with an aim to be fair and efficient in termsof energy use To eliminate collision of data packets R-MAC[26] schedules control and data packet transmissions at thesource and destination nodes rather than using RTSCTSmessage exchanges Channel utilization in R-MAC is en-hanced by an ARQ method called burst-based acknowl-edgment which together with other scheduling algorithmsaddresses the exposed terminal problem besides enhancingthe network throughput

R-MAC design all nodes in this protocol intermittentlyfunction in listen and sleep states so as to avoid wastingenergy in overhearing and idle modes -e time takento listensleep is the same for all nodes each of which

Journal of Computer Networks and Communications 7

chooses a schedule for itself rendering central sched-uling and synchronization irrelevant So in the absenceof traffic in the locality the node simply does periodiclistening and sleeping and to eliminate collisionstransmissions are distributively synchronized using areservation-based model in case there is a sender totransmit data R-MAC differs from S-MAC in that ittransmits data to a number of nodes using the sleepmode [24] -ere are 3 stages in this protocol (i) la-tency detection (ii) period announcement and (iii)periodic operation of which (i) and (ii) are for nodesynchronization and (iii) is for listensleep procedures

(i) Latency detection phase here all nodes broadcast aNeighbor Discovery control packet (ND) at an ar-bitrarily chosen time since they are powered onWhen the ND is received from the neighbor nodethe receiver notes its arrival time and chooses arandom time to send an acknowledgment (ACK-ND)mdashthe same as the receivedND in sizemdashinwhichthe duration from ND arrival to ACK-ND trans-mission I2 is specified and the time from ND packettransmission to ACK-ND arrival I1 is computed

(a) Propagation latency between the 2 nodesL (I1 minus I2)2

(b) -us propagation latency can be defined as ldquotheinterval from the time the first node sends the firstbit of a packet to the time the second node receivesthe last bit of the packetrdquo [24]

(ii) Period announcement phase here nodes choosetheir own random listensleep time and start timeand broadcast these and when packets are re-ceived the node turns the received schedule intoits own All nodes will record their neighborsrsquoschedules in relation to theirs -ese two phasesmake a number of rounds to ensure all nodes areinformed about their neighboring nodes

(iii) Periodic operation phase here there is datatransmission [26] and nodes intermittently go intothe sleep mode and wake up One period is a listensleep cycle and nodes possess similar periods Twoparts make a period T0 listen window TL and sleepwindow TS as shown in the following equation

T0 TL + TS (2)

In this protocol the nodes make use of control packetsfor communication-ese include REV (reservation packet)ACKREV (acknowledgment packet for REV) and DATAACK-DATA (acknowledgment packet for data) messageexchanges all of which are similar in size

Advantages

(i) It is fair and efficient in terms of energy use(ii) It eliminates collision of data packets

216 AR-MAC (Adaptive-Reliable Media Access Control)-is is a TDMA-based MAC protocol for wireless body areanetworks (WBANs) designed to reduce energy consumption

as proposed by Rahim et al [27] -e protocol allocates aguaranteed time slot (GTS) to all nodes for transmissiondepending on their needs It utilizes periodic sleep andwakeup to minimize overhearing and idle listening based onthe needs of the node It has a central node (CN) having bigbatteries with higher computational power and single ordouble transceivers For double transceivers the sum of thetime frame TFrame is assigned for node transmission But inAR-MAC a single transceiver CN is considered with TFramehaving 3 subdivisions contention-free period (CFP) forsensor connection contention access period (CAP) foremergencies and time TMS to communicate with themonitoring station (MS)

Channel selection the CN first scans for redundantchannels from which it chooses one for transmissionand broadcasts its address and other information to thenodes -e destination nodes search for channels byscanning the radio frequency (RF) channels If free itswitches to the next and if busy it will wait for time TCPlistening to packets and again switch to the subsequentchannel if it does not get the channel packets When thenode finally succeeds in getting a channel packettransmission commences and an ACK packet is sent tothe CNTime slot assignment a Time Slot Request (TSR) packetis sent to the CN by the sensor node after it has selectedthe RF channel -e TSR packet has data rate and timeslot info for the node Static-size time slots with staticguard-band time TGB are proposed in [28] -isprotocol employs the adaptive algorithm of time slot(TS) and GB time and the CN allocates TS whilesending Time Slot Request Reply (TSRR) based on thenode traffic pattern -e size of the time slots variesdepending on the node requirements Based on thetransmissionmodel the data packet transmission ACKpacket reception and some amount of delay can betolerated by allocated TS To prevent interference be-tween adjacent time slots resulting from the node andCN clock drift TGB is inserted between them and itsvalue is given as follows

TGB1

FxTS1100

TGBn

FxTSn

100

TGBnn+1

F

100X12

TSn + TSn+11113858 1113859

(3)

where FGB factor depending on the mean driftvalue Nodes go to sleep after slots have been suc-cessfully assigned and wake up to transmit in assignedslots thus saving energy wasted in idle listening sinceeven allocated slots are collision freeSynchronization for periodic synchronization TDMAneeds additional energy [29] It needs a lot of energy tosynchronize nodes after several cycles To reduce en-ergy wastage and control collisions AR-MAC employs

8 Journal of Computer Networks and Communications

a new synchronization scheme When a data packetarrives the current and expected packet arrival timeswith allowable delay (D) are compared by the CN and adrift value (DV) is computed based on the differenceΔT -e DV is sent to the node in the ACK packet forchanging the slot in subsequent transmissions thoughthe value is based on allowable delay and F In caseΔTgtD the CN will send the DV in the SYNC-ACKpacket for subsequent SYNC to nodes or else a mereACK packet is sent by the CN for the received packetFor any forthcoming data transmissions the node willchange its wake-up time plan based on the DV Withsuch a sync approach a sleeping node will not lose syncfor all cycles N Acceptable delay (D) is given as

D min TS1 TSn( 1113857XF

100 (4)

where F is the guard-band factor

ΔT expected arrival time minus current arrival time

DV 0 if |ΔT|ltD

ΔT if |ΔT|gtD1113896

(5)

Frame format the AR-MAC protocol employs twoforms of packets namely (i) data packets in which anode transmits in its assigned time slot and employsCAP for emergency data and (ii) control packets thatinclude channel packet Time Slot Request (TSR)packet Time Slot Request Reply (TSRR) packet Syn-chronization-Acknowledgment (SYNC-ACK) packetData Request (DR) packet and Acknowledgment(ACK) packet [27]Energy consumption if N is the number of cyclesenergy consumption for these is measured as follows

ETotal 1113944N

k1ESleepk

+ 1113944N

k1EActivek

ESleep TSleep x ISleep x V

TSleep Tframe minus Tactive

(6)

where ISleep current from the voltage source V whenthe node is asleep If Esw is the switching energy Etransthe transmission energy Erec the receiving energy andETout the time-out energy then

EActive 2Esw + Etrans + Erec + ETout (7)

Advantages

(i) It reduces energy consumption(ii) It minimizes overhearing and idle listening based

on the needs of the node via periodic sleep andwakeup

217 ER-MAC It is a combination of TDMA and CSMAmethods making it flexible and adaptable to topological andtraffic changes and hence applicable in rapid response

situations inWSNs Collision-free slots can be planned usingTDMA [30] ER-MAC is designed in such a way that thefollowing are met

(a) It has a higher ratio of packet delivery and low delayunder lower energy consumption levels out-performing Z-MAC

(b) It keeps two priority queues in order to distinguishhigh from least priority packets

(c) It permits contention in TDMA slots so as to handleenormous traffic volumes

(d) It has a harmonized slot shape and the nodes areable to locally change their plans enabling themconnect to or exit the network with easeProtocol design it discovers the topology in such away that it uses the simple flooding mechanism tobuild the tree as initiated by the base station (BS) -etopology discovery aims at neighbor discovery andchange tracking in addition to setting up a routingtree A TOPOLOGY_DISCOVERY message com-prising hop_count new_parent_id and old_-parent_id is generated by the BS A node broadcasts itto discover if it has any potential children and aresponse to the parent In case of the desire to changethe parent the former parent is notified For all nodesthe hop count to the BS parent ID children and one-hop neighbor are registered with the BS at this stageTDMA slot assignment when a BS sends a SYNCmessage it switches to the TDMAmode which whenreceived by the child uses the broadcast slot for itschildrenrsquos synchronization Neighboring nodescannot share a slot since slots are assigned andschedules are exchanged accordingly Slot assign-ment begins with a leaf node in a bottom-up modesuch that the flow of messages to the BS may besmooth Before getting the child-node schedules anonleaf node will not transmit its data via a unicastslot descendantrsquos data via many unicast slots andchild synchronization via a broadcast slot Childnodes send SCHEDULE_NOTIFICATION mes-sages to the BS to end this phase

Local time synchronization is done via parent-childbroadcast sync since they both need to operate the same clockso that when one is transmitting the other is ready to receivePacket prioritization happens in such a way that a highpriority queue is first emptied before a lower priority queue isallowed to transmit In case a queue is filled up the shortestslack packet is dropped and then fairness over source isconsidered as queue modification is done to enable the BSbalance info from different nodes Energy conservation is bysender turning off the radio in absence of any info to transmitand the receiver will switch to the sleep mode in case nopackets are received after a certain timeout [30]

Advantages

(i) It is flexible and adaptable to topological and trafficchanges

Journal of Computer Networks and Communications 9

(ii) It is applicable in rapid response situations inWSNs

(iii) It has a higher ratio of packet delivery and lowdelay under lower energy consumption levels

218 RI-MAC (Receiver-Initiated MAC) -is is an asyn-chronous duty-cycle protocol that tries to reduce timeoccupied by the communicating nodes on the medium toagree on time for data exchange-e sending node waits forthe receiving node to signify the beginning of the com-munication session through a short beacon frame that itsends -e channel is only occupied by the beacon and datatransmission reducing channel occupancy and thus en-abling data exchange for other nodes too -is implies animprovement in channel utilization that subsequently leadsto increment in throughput packet delivery ratio andpower efficiency over a wide-ranging traffic load [31] Beinga receiver-initiated model it ensures substantial reductionin overhearing and lowers the probability of colliding andthe cost of recovery compared to B-MAC [23] or X-MAC[32] RI-MAC is further capable of using a beacon withACK and Ready-to-Receive packets though it has aproblem if a sending node with Ready packets needs to havethe radio on till a ready-to-receive node is awake [13] whichthen sends it a beacon and transmission begins immediatelyand later is acknowledged by another beacon Nodes in-termittently wake up to listen to any possible frames meantfor them depending on their schedules A node announcesits readiness to receive a frame through a broadcast beaconon waking up when it switches on the radio When a dataframe has been received the receiver will spend some moretime known as ldquodwell timerdquo in the active state so as toreceive queued packets which is the time in RI-MAC thatdepends on the number of contending senders If there areseveral contending senders that might not have beenpredicted there is a challenge in trying to reduce the re-ceiverrsquos active time to optimize power efficiency and reducethe cost of collision detection and lost data recovery even inthe case of hidden senders For this protocol this can beaddressed by a beacon frame from the receiving node forcoordinating data frame transmissions of contendingsending nodes-e protocol also beats B-MAC and X-MACin minimizing the cost of collision detection and datarecovery through coordination of the receiversrsquo dataframes -is is because the receiving node is aware of themaximum delay prior to the arrival of the frame-e abilityto broadcast is supported by RI-MAC when it sends data asunicast to each senderrsquos neighboring node or continuouslytransmits the frame in a back-to-back mode for as long asthe sleep interval [31]

Advantages

(i) It improves channel utilization leading to in-crement in throughput packet delivery ratio andpower efficiency over a wide-ranging traffic load

(ii) It ensures substantial reduction in overhearing andlowers the probability of colliding and the cost ofrecovery

219 SRI-MAC (Synchronous Receiver-Initiated MAC)-is is a synchronous duty-cycle protocol that implementsthe method of receiver-initiated data transmission with anaim of enhancing network lifetime through energy wastageavoidance via collision overhearing and idle listening To dothis and also minimize duty cycle it uses a sequence ofadaptive beacons and RTSCTS packets In this protocolsending nodes do not go to sleep till they have got CTSpackets from receiving nodes to begin transmitting exceptfor nonsending nodes that return into the sleep mode rightaway as if the channel was idle [33] Since sending nodeshave separate schedules it reduces the contention proba-bility leading to more energy conservation A time frame isdivided into the information period the allocation periodand the communication period which are discussed asfollows

Information period at frame initiation a beacon is sentby the receiving node declaring its return from sleepand readiness to accept data packets in case the channelis not busy -e beacon is a small packet alerting otherlistening nodes of the transmitterrsquos readiness to get arequest for channel It consists of the receiverrsquos ID andthe duration allocation period (DAP) that relies onnumber of the receiving nodersquos neighbors [13]Allocation period here the receiving node stays awake forsome little time after broadcasting its beacon to confirm ifthere is any node having anRTS packet to transmitWhenthe beacon has been received an RTS packet (with 3fieldsmdashsender ID receiver ID and data size) is sent by thesender prompting the receiver to send a CTS for SYNC asthere is neighbor information exchangeCommunication period SRI-MAC is capable of or-ganizing and scheduling transmissions by means ofsensor node IDs -e receiver assigns order andtransmitting time for every sender such that they switchoff their radios till their time to transmit is dueHowever the receiving node remains in the wake-upmode to get all sending nodersquos data SRI-MAC canenable sensors have lengthy sleeping times duringtransmission such that energy can be saved as long as itis neither sending nor receiving data -e protocolrsquoscommunication period is premised on the TDMAtechnique which conserves energy since the radiorsquosduty cycle is lowered and collision is eliminated [33]

Advantages

(i) Since it employs the TDMA principle it eliminatescollisions while maintaining energy conservationtoo

(ii) Noncommunicating nodes switch off their radioand get into the sleep mode which eliminatesoverhearing

(iii) Since receiving nodes broadcast SYNC signals toall nodes in the form of CTS packets overemittingis avoided

(iv) It decreases duty cycling as sleep time increasesand lowers idle listening

10 Journal of Computer Networks and Communications

2110 H-MAC -is protocol is built on the power savingmechanism (PSM) of IEEE 80211 and slotted Aloha pro-tocols to enhance performance using multiple slots Timehere comprises large frames each having an active and asleeping part [34] If any node has got data to send it will askfor slots from the receiver node during active time and latertransmit the packets in the prenegotiated slots Nodes gointo the sleep mode during sleep-time slots if they do nothave data to transmit

Latency in IEEE 80211 if N is the number of hops nthe value for the current hop tcsn the back-off time andttx the transmission delay the total latency is calculatedusing the following equation

D(N) 1113944N

n1tcsn + ttx1113872 1113873 (8)

and the average latency is given by the followingequation

E[D(N)] N tcsn + ttx1113872 1113873 (9)

-roughput in H-MAC is given by the followingequation

Th npnm

tact + Cts (10)

where np is the number of transmitted packets nm is themaximum number of nodes connected in a time frameC is the number of slots with the same length tact is thelistening time and ts is the sleep delayAdvantages

(i) It enhances performance using multiple slots

2111 Z-MAC (ZebraMAC) -is rides on the strengths andoffsets weaknesses of TDMA and CSMA It yields very goodchannel usage and low delay under low contention and lowerscollision among two-hop neighbors at a reduced cost Itsperformance is strong via error synchronization failure inassigning slots and channel conditions that change with timeIn the set-up phase the protocol operates through certainsteps including neighbor discovery slot assignment localframe exchange and global time synchronization whichunless the topology of the network changes significantly donot run again When the data are being transmitted there isimprovement in network throughput and energy efficiencywhich covers up for the initial upfront operational costs [35]

Disadvantages

(i) Initial slot assignment is quite a cumbersome taskhere since it is done at the beginning yet this maynot be at the same time in reality It further as-sumes static links until setup happens again whichtoo is impractical

(ii) It also suffers from the hidden-terminal problemsince slot owners have a small contention windowand priority for slots allocated to them due to theECN messages

(iii) Synchronizing time also poses a challenge in thisprotocol due to a large clock drift between nodes

(iv) Source and destination nodes too are not wellcoordinated in Z-MAC [36]

2112 Q-MAC (QoS-Aware Media Access Control) -isprotocol offers high QoS with improved energy efficiencyand accesses the wireless channel using MACAW as theunderlying protocol [37] Traffic from different nodes isprioritized for QoS satisfaction depending on how criticalthe data are It mainly has two scheduling algorithms inWSNs (i) intranode and (ii) internode scheduling -eintranode scheduling scheme adopts a first-in first-out-(FIFO-) based queuing algorithm based on application andMAC layer abstraction for data packet classification In-ternode scheduling reduces idle listening and collision on thechannel for improved energy consumption

Advantages

(i) It gives a high QoS with improved energy efficiency

2113 WiseMAC (Wireless Sensor MAC) WiseMAC isbased on the Aloha protocol and employs preamble sam-pling in achieving low-power transmissions in infrastructuresensor networks by occasionally sampling the channel tocheck for activity [3238] -e protocol employs the samemethod as B-MAC except that the sender shortens theextended preamblersquos length by scheduling its transmissionsaccordingly after learning the receiverrsquos awake schedules-is is done by the receiver node putting its subsequentawake time on the ACK packet such that when the senderneeds to send to it again it starts the preamble shortly beforeit wakes up so as to avoid energy wastage in transmitting thepreamble [32] -e data frame is repeated instead of theextended preamble for less traffic transmissions that have alengthier preamble than the data frame in WiseMAC Afterprocessing the frame a node that is not the target recipientwill go back into the sleep mode while the recipient staysawake till the session ends thereby sending an ACK packetFor an idle channel the power consumption is rather lowwith this method Nevertheless there is a problem of longwake-up preamble limiting throughput and causing anenormous power consumption overhead in reception [38]All nodes overhearing a transmission bear the overheadtogether with the recipient node -e protocol can addressissues concerning low-power transmission but has nomechanism of nodes adapting to varying traffic patterns[32] To compensate for drift between clock at the accesspoint and on the sensor node there is a need to compute thewake-up preamble duration -e drift varies directly withtime since previous resynchronization and is given by thefollowing equation

TP min 4θl TW( 1113857 (11)

where θ is the time-base quartz frequency tolerance and l isthe interval between two transmissions Increase in trafficvolume naturally reduces overhearing

Journal of Computer Networks and Communications 11

Radiomodel for protocols with lower power transitiondelays between transceiver states and their powerconsumption are modeled Some of the states includeDOZE where the transceiver neither transmits norreceives but is prepared to power on to the receivetransmit state very fast RXwhere the transceiver listensto the channel receives data or demodulates data fromthe noisyidle channel and TX which is the trans-mission state by the transceiver Let TS be the set-uptime needed to turn the transceiver from the DOZE tothe RX or TX state TT the turn-around time needed toswitch between RX and TX states and B the trans-ceiverrsquos bit rate PZ PR and PT are the values of powerconsumed in the DOZE RX and TX statesrespectively

1113954PR PR minus PZ

1113954PT PT minus PZ(12)

where 1113954PR and 1113954PT are the changes in power as a result ofbeing in RX and TX states respectivelyTraffic model let N be the number of sensor nodesconnected to an access point (AP) and λ be the global rateat which the downlink Poisson traffic arrives at the APfrom the fixed network Traffic toward each sensor node isthe same and is given by λN Average packet interarrivaltime Lwith which data packets arrive at a node is given byL Nλ If TC and TD are the control packets and datapackets respectively assuming a small traffic volume theglobal interarrival 1λ far exceeds the summation ofdurations for the data packet and control packet and theturn-around time Mathematically it is given as

1λ≫TD + TC + TT (13)

-e average power consumed by WiseMAC andtransmission delay incurred are given by equations (14)and (15) [38]

PW PZ +1113954PR TS +(1B)( 1113857

TW+

1113954PR X + TD + TT( 1113857

L

+ 1113954PR(N minus 1)Y

L

(14)

where

X 2θL 1 minus eminus TD4θL( )1113874 1113875

Y T2D + 12TDθL

2TW1 minus e

minus TW4θL( )1113874 1113875

DW TD +TW

21 minus e

minus TW4θL( )1113874 1113875

+ 2θL 2 minus eminus TD4θL( ) minus e

minus TW4θL( )1113874 1113875

(15)

where PW and DW are the power consumed byWiseMAC and the transmission delay for WiseMACrespectivelyAdvantages

(i) It achieves low-power transmissions in in-frastructure sensor networks through preamblesampling while checking for activity

(ii) It reduces overhearing due to increased trafficvolume

Disadvantages

(i) It has a problem of long wake-up preamble limitingthroughput and causing enormous power con-sumption overhead

2114 X-MAC It is a simple minimal transmission powerMAC protocol that uses a short preamble with decouplingfor the sending and receiving node sleep plans inWSNs-isleads to a significant reduction in energy usage at sendingand receiving nodes and decrease in per-hop latency andgives flexibility in adapting to bursty and periodic sensordata sources [32]

Protocol design X-MACrsquos design is aimed at increasingenergy efficiency reducing data latency and increasingthroughput -is enables application in different kindsof packetized and bit stream digitized radios as well assimplified low-overhead distributed application -edesign is as such in order to address the low-powerlistening problems including overhearing excessivepreamble and incompatibility with packetizing radiosas in [32]Asynchronous duty cycling a node with information totransmit will send the packets after sending a preamblewhereas the remaining nodes will preserve their sleepschedules that are not synchronized -e receiver willsample the link on waking up and stays awake if it detectsa preamble and if it is the target it will return to sleep if itis not the target after getting the whole preambleEmbedding the target ID in the preamble to avoidoverhearing the lengthy preamble is divided into manyshortened packets embedded with the receiver nodersquosID so as to eliminate overhearing One lengthy pre-amble is constituted by the shortened preamble packetstream Any node on receiving the short preamble afterwaking up will examine the target ID on the packet Itgoes back into the sleepmode right away resuming dutycycling if it is not the targeted recipient or stay awakefor successive packets if it is the one -is approachchecks energy wastage due to network density It canalso be used on several types of radiosReducing excessive preamble using strobing it ispossible to enable low-power transmission and alsosave energy by reducing time for preamble trans-mission through the use of enhanced preambles inaddition to preamble sampling X-MAC does not sendan endless preamble packet stream but puts slight

12 Journal of Computer Networks and Communications

pauses in between packets to allow the sending node topause and listen to the channel and allow the receivingnode to resend an early ACK packet during that pause-is makes the sending node begin sending datapackets and the receiving node cut excessive preamblesand thus decrease per-hop latency and energy wastageon unnecessary waiting and transmissionPacketizing radios low-power listening is limited incapability for supporting packetizing radios butX-MACrsquos short strobed preambles can support all typesof digital radiosAdaptation to traffic load while a number of WSNapplications are reliable in traffic production they stillmust adapt to changing traffic Nodes are going to haveseparate sleep schedules depending on traffic loadConsidering systems that have time-fluctuating trafficall predetermined static schedules are suboptimalOptimality Buettner et al [32] model the expectationof energy as follows

Es (preamble energy + energy perACK listen)

lowast(expected preamble minus listen iterations)

+(energy to send parket)

Es PTxSp + PRxSal1113872 1113873 Rl + Rs( 1113857

Rl minus Sp+ SdPTx

(16)

where Es expected energy to send a packet PTx andPRx are the transmission and receiver power re-spectively Sp Sal and Sd are the duration of thesenderrsquos preamble ACK listen and data transmissionperiods respectively and Rl and Rs are the receiverrsquoslisten and sleep periods respectively Expected energyto receive a packet is given as

Er (listen cycle energy + sleep cycle energy)

lowast(expected iterations for preamble arrival)

+(energy to sendACK) +(energy to receive packet)

Er PsRs + PRxRl( 1113857

1 minus 1 minus Pd(t)( 1113857 Rl + Rs( 1113857+ RaPTx + RdPRx

(17)

And the one-hop expected latency is given as

Lat (preamble duration + ACK listen)

lowast(expected number of iterations)

+ duration to send packet

Lat Sp + Sal1113872 1113873 Rl + Rs( 1113857

Rl minus Sp

(18)

Advantages

(i) It increases energy efficiency(ii) It reduces data latency(iii) It increases throughput

2115 A-MAC (Advanced MAC) A-MAC is a TDMA-based protocol whose design is aimed at enhancing thelifetime of the network for low rate and reliability of datatransmission [39] To assign time slots it does not depend ona central controller like most protocols but it gathers in-formation from its neighbors so as to allocate the slotsthrough a distributed algorithm Energy is controlled byscheduling power when there is not a single event Its designcomprises many frames with each having many time slotswhich at the beginning have got a beacon transmitted tosynchronize and exchange information with neighborsNodes to participate in the subsequent session are selectedby the controlled node so they do not go into the sleep modewith others [13] -is protocol operates under four time slotdivisions namely initial wait discover and active states Anode in the initial state begins by listening to the medium forthe beacon packet from other nodes for network synchro-nization which when received makes it change the timerthrough subtraction of beacon transmission time from re-ception time To avoid synchronization problems due todrift and allow for continuity the node has to get thestrongest beacon signal It then enters the wait state afterchoosing a number of waiting frames so as to reduce theprobability that nodes get into the discover state simulta-neously In case the synchronization beacon is lost while stillin the wait state it will return to the initial state for anotherone On expiry of the waiting counter it enters the discoverstate and collects information from neighbors by listening totheir beacon messages On successful selection of a time slotit enters the active state where it endlessly transmits beaconsat the initiation of the slot A node goes to sleep when abeacon has been transmitted and there are no more data fortransmission or in case the neighborrsquos beacon received in-dicates no data packets are coming in [39]

Advantages

(i) It can enable multicasting(ii) It enhances the lifetime of the network(iii) It offers reliable data transmission

2116 L-MAC (Lightweight MAC) Based on TDMA thisprotocol is aimed at minimizing the number of transceiverswitches adapting the nodal sleep interval to data trafficamount and making implementation less complex [40] Apacket is transmitted comprising a control and a data partControl packets have a fixed size and the time slot con-trollerrsquos ID -ey show the distance from the transmittingnode for simplified routing to the network gateway as well asaddressing target node and reporting data unit length andinternode synchronization Adjacent nodes are more in-terested in getting control packets from nearby nodes -enodersquos transceivers are switched off until the subsequenttime slot unless the node is also addressed or the message isomnicast If so it listens to the data unit which may not fillthe remaining time slot in totality-e transceivers on eitherside are turned off when the transfer of packets is successfulEnergy wastage through idle listening in uncontrolled timeslots is avoided via a short time-out interval Network nodes

Journal of Computer Networks and Communications 13

are able to communicate without colliding-is saves energyand increases the lifetime of the network A node in L-MACcan only transmit one message per frame [13] -e networkis set up in such a way that all nodes are not synchronized onbeing switched on-e gateway takes control of a timeslot soas to have them synchronized and its immediate neighborgets the control messages thereby synchronizing their timerswith the gatewayrsquos Neighboring nodes learn of multiplegatewaysrsquo time slots in their vicinity just after one frame-ejust synchronized nodes randomly choose time slots forcontrolling as long as they are not occupied -e nodes thuspreserve a table for the neighboring nodes which makes atimeslot reusable after a minimum of 3 hops and preventsmessage collision All nodes keep their slots before thebattery is down or learn of an impending collision All nodescan track their hop distances to a selected gateway node andthe information is communicated in control messages [40]

Advantages

(i) It minimizes the number of transceiver switches(ii) It adapts the nodal sleep interval to data traffic

amount(ii) It makes implementation less complex(iii) It offers improved network lifetime(iv) Energy wastage through idle listening in un-

controlled time slots is avoided via a short time-out interval

Disadvantages

(i) L-MAC is not very good for mobile SNs since slotsare calculated only once

(ii) It suffers idle listening because nodes keep listeningto slotsrsquo control parts so as to get data and enableother nodes to join the network all the time [13]

2117 C-MAC -is was designed for WSNs with 3 seg-ments namely (i) aggressive RTS having double channelcheck to assess the channel (ii) anycast to initialize the flowand (iii) convergent packet forwarding to stabilize the flowIt promises an improvement in energy efficiency lowerlatency and enhanced throughput while avoiding syn-chronization overhead though it operates only in low duty-cycling apps especially where there is less traffic But whenthere is more traffic the protocol employs anycast-basedpacket forwarding for waking the nodes up or finding aforwarder quickly and converges from route-suboptimal toroute-optimal unicast After this energy could be preservedby the nodes using a synchronized wake-up plan [13]

Advantages

(i) It improves energy efficiency(ii) It lowers latency(iii) It enhances throughput avoiding synchronization

overhead

Disadvantages

(i) It operates only in low duty-cycling applicationsespecially where there is less traffic

22 MAC Protocols for WMSNs

221 Diff-MAC -is is a CSMACA-based QoS-awarehybrid-priority-based MAC protocol for WMSNs It targetsenhancing the channel utilization while making use of ef-ficient service differentiation to do coordination for mediumaccess of each traffic class while giving fair quick datatransfer It offers QoS using the following features [14]

(i) It balances energy consumption and delay throughsensor node duty-cycling adaptation depending onthe dominant traffic class

(ii) It has a feature for fragmentation and messagepassing that breaks lengthy video frames intosmaller packets that are transmitted as a burst thusreducing retransmission costs if there are any MACfailures

(iii) -ere is fair data delivery amongst the nodes andtraffic classes respectively resulting from intranodeand intraqueue prioritization reducing unbearableperformance

(iv) -e number of collisions can be reduced and packetlatencies are kept within limits by adjusting the sizeof the congestion window in this protocol as per thetraffic needsAdvantages

(i) Diff-MAC is very suitable for WMSNs since itadapts quickly to changes in networkconditions

(ii) It is a fair protocol(iii) It is scalable

Disadvantages

(i) It is quite difficult to monitor the statistics ofthe network

(ii) Adapting dynamically to the network condi-tions is not an easy task

(iii) It exhibits some degree of latency in packetdelivery

-e implementation of Diff-MAC on Imote2 indicatesthat resource requirements can be met on the sensorhardware Results of extensive simulation runs showed thatDiff-MAC outperforms its competitors

222 Cluster-Based On-Demand Multichannel MAC Pro-tocol for WMSN (COM-MAC) -is is an energy-efficientprotocol in WMSNs with high throughput and reliable datatransmission It achieves energy efficiency by schedulingmultichannel media access to reduce channel contentionand remove collision idle listening and overhearing [41]

Design it has a three-phased operation with the (i)request phase (ii) scheduling phase and (iii) datatransmission phaseRequest phase here all nodes have two protocols eachfor sending requests to the cluster head (CH) ie thecontention-based and contention-free TDMAFDMA

14 Journal of Computer Networks and Communications

protocols -e contention-based protocol allocates achannel per node for sending the request message andfree channels are useable as control channels Forcontention-free protocols a control slot is allocatedwhen a network is deployed Here a slot per channel isallocated to a different node enabling requests to be sentwithout interfering with transmissions from othernodes -e contention-based protocol operates undertwo steps (i) Control channel assignment phase achannel is assigned per node for transmitting a request(REQ) message A channel is assigned to multiplenodes and nodes are evenly distributed to availablechannels for congestion control (ii) Request trans-mission phase nodes with data send a REQmessage viathe assigned control channel to notify the CH whichresponds with an ACK message -e contention-freeTDMAFDMA protocol too has two phases includingthe control slot assignment and request transmissionphases For unreliable channel conditions and lowtraffic load the contention-based protocol is fit for usebecause of less collisions and congestion whereas thecontention-free protocol is preferable in heavy trafficload and reliable channel applicationsScheduling phase from the above phase a schedule isgenerated by the CH for coordination and this enablesnodes to transmit and broadcast their data via controlchannels-e schedule for broadcasting only comprisesdata for nodes assigned to a channel -e schedule isrebroadcast for transmission reliability enhancementbut for energy efficiency the nodersquos transceiver isswitched off when all the schedule is completely de-livered till it is the time for transmitting again-e timeslot and data communication radio channel are allincluded in the schedule generated for each nodeData transmission phase on getting the schedule thesensor node sends its data as per the time slot and thechannel assigned -e time slot is divided into datatransmission andACK sections ACK supports link layererror control Using implicit selective repeat ARQ theCH sends an ACK message for all received packets Ifthere is no ACK the packet is assumed lost and hence aretransmission during the subsequent interval is done-is enhances reliability of transmissions though excessdelay is possible here which is also intolerable for real-time critical apps like video or audio -is gives rise to ahybrid MAC protocol to exploit the unused spectrumduring data transmission sessions [41]Advantages

(i) It achieves energy efficiency by scheduling mul-tichannel media access to reduce channel con-tention and remove collision idle listening andoverhearing

(ii) It maximizes network throughput by dynamicallyallocating time slots and channels for nodes byexecution of a traffic-adaptive QoS-aware sched-uling algorithm depending on needs for QoS andtraffic data information

(iii) It increases transmission reliability by in-corporating spectrum-aware ARQ to avoid spec-trum wastage

223 EQoSA -is protocol is developed for video andimage transfer across WMSNs in a hybrid mode withprovision of QoS Concerning the active sensor nodes andtheir traffic loads it employs dynamic session sizes andchanges bitmap-assisted (BMA) MACrsquos fixed session sizeNodes declare any data available for transmission duringcontention time -is forces the cluster head to make aschedule of how slots have been assigned It communicatesthis schedule to the nodes giving each node a given numberof slots per session and so enabling the busty traffic to beaccommodated [14] -e protocol does not perform wellwith DiffServe mechanisms but does well with schemessupporting busty traffic loads and strong cluster heads arenecessary in WMSNs that can assign and broadcast slots

224 QoS-Supported Energy-Efficient MAC (QE-MAC)Based on the IEEE 80211e standard this protocol ensuresfairness and has low latency and jitter and better efficiencyon energy usage than other QoS-aware MAC protocols forWMSNs It operates under a two-phased routine

-e first phase introduces the innovating prioritymechanism in IEEE 80211e QoS and several data types areassigned priorities Its coordination function is a combi-nation of distributed contention-based channel allocationand centralized polling-based channel access mechanisms

-e second phase preserves the nodesrsquo energy usingdynamic duty cycling -ere is an exchange of RTSCTSpackets to supplement CSMACA by muting nodes near thesenderreceiver during packet duration [42]

Advantages

(i) It ensures fairness(ii) It has low latency and jitter and better efficiency on

energy usage than other QoS-awareMAC protocolsfor WMSNs

225 Black-Burst (BB) Contention A contention-basedMAC scheme for enhancing instantaneous QoS access tocarrier-sense WSNs is proposed in [42] In this schemenodes contend for the channel until it is free Real-timepackets are given higher priority over non-real-time packetsand if a node is transmitting real-time packets it is prior-itized over other nodes [15] Voice and video are some of thereal-time apps in consideration here which periodicallyaccess the common radio channel when transmitting -eseneed constrained end-to-end delay of packets at the MAClayer -is protocol majorly does the following [42]

(i) Prioritization of real-time traffic(ii) Implementation of the round-robin mechanism in

real-time nodes(iii) Bounded access delays among real-time packets

Journal of Computer Networks and Communications 15

(iv) Implementation of real-time communication withvarying needs of bandwidth good for multimediatraffic

Amechanism for dynamically assigning packet prioritiesbased on their deadlines and the traversed hops is proposedin [43] -e ReAllocative Priority (ReAP) mechanism ismeant to provide QoS at the MAC layer to video traffic -isscheme introduces adaptive TXOP by modifying the TXOPdynamically based on the queue length [15] Other protocolsin this category include dynamic duty cycle and adaptivecontention window-based MAC protocol frame sharing(FRASH) MAC protocol and real-time independentchannels (RICH) MAC protocol

Advantages

(i) It can implement real-time communication withvarying needs of bandwidth

23 Challenges and Opportunities(a) One of the challenges in satellite transmission is

cross-talk leading to interference in different fre-quency bands It is necessary to increase the rise andfall times as well as protect sensitive nodes as wecircumvent floating ones if we are to regulate cross-talk as in satellite communication therefore Wefurther need to investigate the possibility of usingdifferential signaling -is constitutes an opportu-nity for further research

(b) Another challenge is that the hidden and exposedterminal problems are still noticeable despite the useof RTSCTS solutions in MAC since collisions arestill possible from transmissions of separate nodesAt the MAC layer ACKs might block efficientchannel usage by prohibiting exposed terminalsfrom reusing the channel and may be eliminatedleaving no proof of sender data packet receptionUsing a directional antenna is one of the proposedtechniques of high-speed WSNs like IEEE 80211-is constitutes an opportunity for further study intothese issues for better solutions though

(c) Another challenge is that since nodes in contention-based MAC protocols are mobile there are manycollisions by the packets and schedule in-consistencies in schedule-based protocols in the two-hop neighborhood information when nodes areentering or leaving [44] Deng et al [45] propose amobility-based clustering (MBC) protocol for mo-bile nodes to enhance the packet delivery rate -isprotocol in comparison with CBR has 25 chancesof decreasing collision-induced packet losses and50 chances in comparison with low-energyadaptive clustering hierarchy (LEACH)

(d) -e process of analyzing and evaluating routingperformance for WMSNs is quite complex becauseof the bursty multimedia traffic Because of the wayencoding scheme frames are structured as well as thechanges in between the scenes that occur naturally

the compressed video frames display some substantialburstiness on several time scales -ere will also befrequent route recalculation that is as well energy in-tensive for nodes frequently capturing and transmittingenergy -e solution here could as well possibly be theuse of the MBC protocol Route recalculation tooshould be dynamic to cater for the mobile nodes whichmight be in and out of the range at different instants oftime Recently the use of smart dumpsters for effectivewaste management in smart cities has also been pro-posed as a possible solution [46]

(e) For WSNs their routing protocols do not look atmultimedia applications which require enormousbandwidth and processing energy and should betransmitted in real time with utmost fidelity Extrachallenges in routing are imposed in WMSNs indeployment of heterogeneous sensors since therecould be audio video and still pictures in addition toscalar data during transmission Heterogeneous datahandled also have different QoS requirements andwith many different business needs different ser-vices with varying requirements give rise to a bigchallenge for the routing design -ere is a need todesign protocols that can support WMSNs such asEQoSA [14] for this purpose

Table 2 summarizes the reviewed MAC protocols forWSNs and the ones designed for suitability in WMSNs

3 QoS Strategies at the Application Layer

We discuss many protocols used to solve different needs ofcommunication between machines

31 ConstrainedApplication Protocol (CoAP) It is a requestresponse protocol designed by the Internet Engineering TaskForce (IETF) for constrained devices such as those havinginsufficient RAM or CPU and WPANs with low-powerconstraints resulting in poor transfer of packets andenormous overhead -e protocol was designed for ma-chine-to-machine (M2M) apps and system automation toreduce bandwidth requirements increase packet transferdecrease the overhead and make work simpler leading tolightweight implementation HTTP commands like GETPOST PUT and DELETE are used in a client-server ar-chitecture Using the publishsubscribe architecture CoAPcan support multiple users and hence yield better perfor-mance via multicasting enabling asynchronous messageexchange -is protocol can support both multicasting andunicasting via UDP and hence is a reliable and simpleprotocol -is is done via the messaging sublayer (for re-liability based on UDP) and the requestresponse sublayer(for communication and interaction) [16]

It uses the following messages

(i) Confirmable message this guarantees reliabilityusing ACK messages sent by the receiver node-ese could be synchronous or asynchronous ifthere is a need for more computational time

16 Journal of Computer Networks and Communications

(ii) Nonconfirmable message here no ACK messagesare sent back

(iii) Acknowledgment (ACK) message this confirmsreceipt of a confirmable message

(iv) Reset message this confirms receipt of an un-processed message in case any critical parts nec-essary for interpreting the message are missing

(v) Piggybacked response the receiver directly re-sponds while getting the ACK message

(vi) Separate response the reply by the receiver is dis-crete from the ACK message

CoAP has no in-built security features It is secured bythe datagram transport layer security (DTLS) that runs ontop of UDP though it was not designed for the IoT -ismakes its suitability debatable since DTLS offers no supportfor multicasting It also needs more packets for handshakingthereby increasing the traffic volume and computationalresources and so shortening the lifespan of mobile devicessince they use batteries [17]

Advantages

(i) It can support multiple users and hence yieldbetter performance via multicasting

Table 2 Comparison of MAC layer protocols used in QoS provisioning

Protocol Features QoS parameter Comments Access Priorityassignment Energy eff Designed for

S-MAC[22] Collision avoidance Energy latency Requires periodic

synchronization CSMA Hybrid Medium WSNs

B-MAC[23] Collision avoidance Energy delay

throughput Flexible interface CSMA Dynamic High WSNs

T-MAC[25] Collision avoidance Energy Complex nonscalable CSMA Dynamic High WSNs

C-MAC[13]

Dynamic channelallocation Energy Multichannel MAC CSMA Dynamic Medium WSNs

CU-MAC[24]

Multichannelmechanism

-roughputlatency High packet delivery ratio TDMA Dynamic mdash WSNs

A-MAC[39]

Per flow serviceguarantees Delay m-ary tree model

dynamic prioritiesTDMACSMA Dynamic Medium WSNs

ER-MAC[30]

Emergencyresponse

Energy delaydelivery ratio

Adapts to traffic andtopology changes

TDMACSMA Hybrid Higher than Z-

MAC WSNs

RI-MAC[31] Collision detection Delay packet

delivery ratioHigh packet delivery ratio

and throughput CSMA Dynamic Medium WSNs

SRI-MAC[33] Collision avoidance Energy delay Good data transfer and

energy consumption TDMA Dynamic High WSNs

H-MAC[34] Slotted sleep time -roughput

latencyData transmission only in

sleep timeCSMAAloha Dynamic Medium WSNs

Z-MAC[36] Collision avoidance Energy delay

throughputVery good channel usage

and error syncCSMATDMA Dynamic Medium WSNs

Q-MAC[37]

Intrainternodescheduling Energy High QoS improved

energy efficiency CSMA Hybrid High WSNs

X-MAC[32]

Overhearingavoidance

Energy delaythroughput

Reduces excessivepreamble by strobing CSMA Adaptive

static Medium WSNs

L-MAC[40] Collision avoidance Network

lifetime energyReduces the number oftransceiver switches TDMA Static Medium WSNs

WiseMAC[38]

No downlinkchannel collision Power delay Significantly low-power

consumptionAlohaCSMA Static Higher than

ZigbeeInfrastructure

WSNsAR-MAC[27] Collision avoidance Energy delay Uses data packets and

control packets TDMA Static Higher thanIEEE 802154 WBANs

BB-MAC[42]

Real-time trafficprioritization Delay Employs round robin in

real-time nodesCSMATDMA Dynamic High WMSNs

COM-MAC [41] Collision avoidance Packet delay

throughputBalances reliability and

retransmissionTDMAFDMA Dynamic High WMSNs

EQoSA[14]

Busty traffic loadaccommodation Energy Provides QoS support for

video and imagesTDMACDMA mdash High WMSNs

QE-MAC[42] Collision avoidance Energy jitter

latencyRuns central polling-like

channel access CSMA Dynamic High WMSNs

R-MAC[26] Collision avoidance Energy delay Reservation based mdash Static High Underwater

SNs

Journal of Computer Networks and Communications 17

(ii) It enables asynchronous message exchange(iii) It is a reliable and simple protocol

32 Message Queue Telemetry Transport (MQTT)Developed by IBM and later improved by OASIS with anintention of decreasing bandwidth MQTT is a simple open-source publishsubscribe protocol analogous to the client-server version It is appropriate for lightweight IoT appli-cations and M2M transmissions in constrained situationslike limited power computation capability and memory orbandwidth -e network bandwidth and the computationalresource usage are decreasing since clients are not updatedmaking publishsubscribe protocols outperform requestresponse protocols via IoT requirements [16] -is protocolcan run over TCPIP and so has the following advantages

(i) Delivering of packets with reliable guarantee(ii) Enabling multicasting(iii) Remote device session establishment(iv) Reduction of transport overhead and protocol ex-

change to minimize network traffic(v) Provision of the notification mechanism when

anomalies occur [47]

MQTTrsquos design targets economical use of bandwidthand battery so it has a major application in FacebookMessenger MQTT ensures QoS and reliability of messagedelivery by providing the following three QoS levels

(1) Fire and forget a message is sent once depending onthe networkrsquos best effort and there is no need forACK (lowest QoS level)

(2) Delivered at least once a message is sent at least onceand an ACK message is needed to avoid duplicates

(3) Delivered exactly once by a four-way handshake itensures a message is sent exactly one time (highestQoS level)-e architecture it has three key components (i)publishersmdashlightweight nodes sending data to thebroker before returning to sleep (ii) broker whichorganizes sensory data in topics and sends them tointerested subscribers and (iii) subscribersmdashIoTapps interested in data sent by sensors through thebroker as shown in Figure 3

MQTT is a simple lightweight protocol that is betterthan CoAP via low and high packet loss probability interms of throughput and exhibits lower latency in lowsampling rates [48] It supports a number of imple-mentations for embedded devices and mobile applicationsand offers web-socket support too Nevertheless it hassome shortcomings

(i) It is only employed in simple data types(ii) It exhibits high latency in high sampling rates(iii) It is unsuitable for real-time apps in such cases(iv) It does not provide automatic discovery andor

sufficient security at the protocol level [49]

33 Secure Message Queue Telemetry Transport (SMQTT)-is is an extension of MQTT that is based on lightweightattribute-based encryption using elliptic curve cryptogra-phy -e protocol has three entities as shown in Figure 4(i) publisher that publishes information under a certaintopic (ii) subscriber that obtains the same topic in-formation via a broker and (iii) public key generator(PKG) or broker that acts as the trusted third party -eprotocol has four phases namely (a) set-up phase (b)encrypt phase (c) publish phase and (d) decrypt phase

(a) Set-up phase publisher and subscriber provide aunique ID with attributes for registration with thePKG -is generates a secret master key set andpublic parameters that it publishes with set U A1A2 An For ciphertext policy-attribute-basedencryption (CP-ABE) the PKG gives a key set todevices with attributes [50]

(b) Encrypt phase based on the access tree the publishercreates an access policy An access tree is sent by thepublisher to the PKG for generating the key-gen-eration policy in the key policy- (KP-) based ABEalgorithm -e publishers know beforehand thetopics and subscribers to access as well as the rulesfor access Subscribers are issued with private keysfor all communication sessions Publishers in theCP-ABE algorithm come up with access trees andrules where the payload is encrypted by the pub-lisher who also issues more info needed for de-cryption and the rules With the help of public keycryptography data are encrypted giving rise to ci-phertext on the basis of KPCP-ABE and the receivergets SUBACK from the broker [51]

(c) Publish phase ciphertext is embedded in theSPublish command as payload and topic name isdefined in the header by the publisher -e brokerreceives the SPublish packet and replies withPUBACK prompting the publisher to respond withthe PUBREL packet -e broker then broadcasts totopic subscribers and deletes the data forwarding thePUBCOMP packet to the sender node

(d) Decrypt phase the encrypted information isdecrypted by the subscriber with the help of a secretkey Considering KP-ABE on verification that itsatisfies the access policy the subscriber invokes thePKG for a corresponding key which it sends afterverifying the request With CP-ABE subscribers usethe private key for decryption after authenticationthereby enabling offline interaction [51]Advantages

(i) -e information being encrypted makes it moresecure

34 MQTT for Sensor Networks (MQTT-SN) -is is a ver-sion of MQTT designed to operate under environments oflow bandwidth high link failures and short message lengthIt is a publishsubscribe (pubsub) protocol designed for

18 Journal of Computer Networks and Communications

UDP-based WSNs to visualize communication for low-power devices [5152] It comprises a publisher device actingas the MQTT-SN client to forward communications to agateway that also changes the MQTT-SN message to theMQTTmessage before delivering it to the broker that sendsit to MQTT-SN subscribers [51]

It has a gateway for changing protocols from MQTT toMQTT-SN (Figure 5) [52] -is protocol supports nodes in

the sleeping mode using an offline keep-alive procedure thatcan be used by battery-operated devices to get into thesleeping mode To support sleeping clients MQTT-SN has anew offline keep-alive procedure with which battery-oper-ated devices can go to the sleeping state in which theirmessages are buffered at the server until they are awake InMQTT-SN clients with no preconfigured gateway addresscan figure out the real address of their operational gateway

Publisher

Subscriber

Subscriber

Broker

Publisher

4G

Figure 3 Architecture of the MQTT protocol

Register by mentioning URI Register by mentioning URI

Access tree

Set-up

Publisher ReceiverPKG(broker)

Key management Key management

Encrypt data using public parametersand random key This is done using

CP-ABE Publish it as

Subscribe (topics)SUBACK

S-Publish (TopicEncrypt (data))S-Publish (TopicEncrypt (data))

Encryptand publish

PUBACK

PUBACK

PUBRELPUBRELPUBCOMP

PUBCOMP

Figure 4 SMQTT protocol

Journal of Computer Networks and Communications 19

within the network through a discovery procedure Simul-taneously we may have several gateways in a single wirelessnetwork that can be cooperative via load sharing

MQTT-SN architecture it is composed of MQTT-SN clients gateways and forwarders (Figure 5)rough the MQTT-SN gateway (GW) the MQTT-SN protocol is used by the MQTT-SN clients toconnect to the MQTT server Integration betweenthe MQTT-SN GW and the MQTT-SN server mightbe possible or not Translating between the MQTT-SN and the MQTT for a stand-alone GW is doneusing the MQTT-SN protocol If a GW is detachedfrom the network it can be accessed by the MQTT-SN clients through a forwarder is just encapsu-lates the received MQTT-SN frames that are on thewireless side before forwarding them to the GW un-altered It then decapsulates the received frames tooresending them to the client as sent by the GW [52]ere are 2 types of gateways (a) transparent gatewayand (b) aggregating gateway as shown in Figure 6

(i) Transparent gateway this launches and sustains adirect MQTT connection between the MQTT-SNclient and the MQTT server which is entirely re-served for end-to-end client-server messaging enumber of servers and that of MQTT-SN clientsconnected to the GW are the same is transparent

GW translates syntax between these protocols and iseasier to implement compared to the aggregatingGW But the MQTT server must support dierentsessions for all connected clients Some MQTTservers are congured to enable few parallel sessions

(ii) Aggregating gateway this maintains a single MQTTconnection to the server on behalf of all clients Allexchanged messages from the MQTT-SN client to theaggregating GW are directed to the GW which sortsthe information to forward to the server e aggre-gating GW decreases the number of MQTT sessionssimultaneously supported by the server in WSNs withmany battery-operated sensors and actuators (SAs)having limited storage and processing capabilities It ishowever more complex to implement compared to atransparent GWe general format of the MQTT-SNprotocol message has two or four message headeroctets and n octets of variable header that is optionaldepending on the message in question MQTT-SN has256 message types (MsgType) with those from0X1Eminus 0XFD 0X19 and 0XFF reserved for futureusage [51]

Advantages

(i) e protocol can operate under environmentsof low bandwidth high link failures and shortmessage length

MQTT-SN gateway

MQTT-SN

MQTT

MQTT-SN

MQTT-SN

MQTT-SN

MQTT-SN client

MQTT-SN clients

MQTT broker

MQTT-SN client

MQTT-SN client

MQTT-SN gateway

MQTT-SN forwarder

Figure 5 MQTT-SN architecture

20 Journal of Computer Networks and Communications

(ii) It is a pubsub protocol designed for UDP-basedWSNs to visualize communication for low-power devices

(iii) It can be optimal for low-cost battery-operatednodes having low processing power and lessstorage

(iv) It supports nodes in the sleeping mode using theobrvbarine keep-alive procedure used by battery-operated devices

35e Extensible Messaging and Presence Protocol (XMPP)e standardization of this protocol by the IETF made itpopular for usage over the Internet especially in instantmessaging and chatting platforms It is however susceptibleto spamming attacks and its failure to have worldwidesupport has made Google too withdraw its support of thestandard [17] e protocol is mainly useable in the client-server architecture but can also be applicable in client-clientXMPP communication [53] Among the application layerprotocols XMPP is the only one that supports both asyn-chronous (publishsubscribe) and synchronous (requestresponse) services while running over TCP but can still runover other protocols

Given the simplicity of its implementation and ease ofusability in XMPP instant messaging (IM) is suitable forXMPP and is implemented using the message stanzaltmessagegt ese stanzas are usable in one-to-one ormultiuser sessions or chatting or in error reporting andsending headlines Attribute type of the stanza denes themessage type e attribute type could be a chat group chatheadline normal or error An instant messaging network inXMPP can support many XMPP clients and servers with agateway congured to enable other IM network users toconnect to other XMPP users e XMPP server mainlycontrols sessions and XML streams to clients and servers inaddition to routing XML stanzas in the streams [53] eTCP ports 5269 and 5222 have been respectively registeredfor the XMPP server-to-server connections and XMPP clientconnections by the Internet Assigned Numbers Authority(IANA) XMPP enables specication of XMPP ExtensionProtocols (XEP) to enhance its usefulness [17]

XMPP security clients connect to the server usingtransport layer security (TLS) and simple authentication andsecurity layer (SASL) protocols which are at the same timeused by the servers for interdomain communications Fordata condentiality and integrity TLS will encrypt thestreams of XML for all sessions To conrm the identity of anyclient that needs to access the server the SASL protocol doesthe authentication of the XML stream For any communi-cating session with a server to start the client needs to resolvethe serverrsquos DNS hostname For protection of the server of theclient from third-party attacks the server maintains the au-thentication details secret ie IP address and access methodand only original connections of the server are required

In server-to-server communications the SASL protocolis employed for authentication and condentiality Server-to-server communications may as well be deactivated from aserver by an administrator depending on the organizationalpolicy To guard against domain-spoong attacks serverdialback can be a good solution since servers support itsecuring more the XML stanzas Nevertheless it cannot beused to authenticate secure or encrypt in-server streamsand the server identication resulting therein is not strongenough to be relied on It further cannot secure against DNSpoisoning attacks and IP session hijacking for remote do-mains [53] High-security domains can use TLS and SASL

Advantages

(i) It is popular for usage over the Internet especially ininstant messaging and chatting platforms

Disadvantages

(i) It is susceptible to spamming attacks(ii) It has failed to have worldwide support making

Google withdraw support of the standard(iii) It is not secure against DNS poisoning attacks and

IP session hijacking for remote domains

36 Advanced Message Queuing Protocol (AMQP) isprotocol can be utilized by a number of platforms formessage exchange since it is open source and written inmany languages It is a publishsubscribe model depending

MQTT-SNclients Broker

MQTT

(a)

MQTT-SNclients Broker

(b)

Figure 6 (a) Transparent gateway and (b) aggregating gateway

Journal of Computer Networks and Communications 21

on a reliable and efficient messaging queue with capacity forinstant messaging or e-mail for different applications es-pecially in IoT environments It differs from others in thesense that message types can be specified and have theirsource traceable and how to do trade-off between perfor-mance security and reliability For reliability of messagetransmission AMQPmakes use of some delivery guaranteeslike at most once at least once and exactly once It can alsohowever utilize the TCP transport layer for guaranteeingreliability -e AMQPrsquos publishsubscribe method has twocomponents namely (i) exchange queue meant for routingof messages to an appropriate order in the queue and (ii)message queue that acts as storage pending their delivery toreceivers [16] In this protocol model the broker can makerouting decisions contrary to othermessaging systems whereapps using the queue have in them embedded the decision-making logic for recipients or senders -e developer mayneed to check each of the affected apps so as to change thelogic for routing and delivery of the messages [54]

Advantages

(i) It is convenient(ii) It has capacity for interoperability between dif-

ferent vendors(iii) It can use open standards to enable connection

between business partners(iv) It can enable innovations based on AMQP foun-

dations [55](v) It can enable supporting of mission-critical ap-

plications in e-commerce(vi) It can guarantee reliable delivery of messages

Disadvantages

(i) It is not very appropriate for real-timeapplications

(ii) It may not offer automatic discovery(iii) Interoperability is not guaranteed(iv) It lacks open-source libraries for constrained

gadgets [49]

37 Data Distribution Service (DDS) A product of theObject Management Group (OMG) DDS is a pubsubprotocol meant for M2M communications with two sub-layers -ese are the data-centric pubsub sublayer re-sponsible for connecting anonymous data publishers tosubscribers and the data-local reconstruction sublayer that isoptional and integrates DDS into the application layer [56]-e DDS architecture connects system components asshown in Figure 7 Publishers and subscribers are discon-nected with respect to (i) time (ii) space (iii) flow andbehavior (iv) platforms and (v) programming languages[57] A DDS data publisher generates topics and publisher-subscriber coupling is represented as topic name data typeschema and publishers and subscribersrsquo QoS attributesDelivering of data is done by the publisher layer -e datawriter and publishers together make a decision about anynecessary alterations for subsequent forwarding to thesubscribers who need to send data to IoT applications Data

readers are responsible for delivering topics to the sub-scribers after reading through them [56]

Some of the major components of the DDS architectureinclude the following [57]

(i) Topic this is a logical path specifying publicationand subscriptionrsquos data type between DataWritersand DataReaders For a successful session to begintopic names types and DataWriters and Data-Readersrsquo QoS should correspond

(ii) Domain this is a logical setting for transmissionthat can be utilized for separating and optimizingnetwork transmission in a setup of similar appli-cations If DDS apps share a domain ID then theycan exchange data among themselves

(iii) DataWriter and DataReader these are terminalobjects for writing and reading messages re-spectively to and from a universal data space

(iv) Participant this is an object that stands for apublisher or subscriber of the DDS app in a do-main acting like a container for other objects

Advantages

(i) It is highly reliable and offers QoS support(ii) It is highly scalable(iii) It is appropriate for real-time apps(iv) It is fault tolerant(v) It has high level of interoperability(vi) It does not need brokers that may act as

bottlenecks(vii) It offers automatic discovery(viii) It has a user-defined data structure for topic(ix) It supports publisher and subscriber decoupling

Disadvantages

(i) It is originally developed for only stand-aloneLANs

(ii) A lot of memory is needed(iii) It has no open-source libraries to support con-

strained devices(iv) It is hard to design and configure compared to

other protocols [49]

38 RESTful Services REpresentational State Transfer (REST)-is is an architectural style based on a set of principles thatdefine and address networked resources It offers web servicesthat enable transmission from the HTTP device to the devicewithin the IoT architecture and enables clients to reach theserver [16]-is protocol utilizes HTTP commands to backupthe requestresponse messaging model HTTP is a popularand secure protocol in the WWW using TLSSSL [16] ARESTful application has the following characteristics

(i) State and functionality are separated into distrib-uted resources

(ii) All resources are distinctively named by the use ofunique HTTP commands (GET POST PUT orDELETE over the Internet)

22 Journal of Computer Networks and Communications

(iii) It is a clientserver stateless layered protocol thatsupports caching REST uses the same methods asHTTP for all requests or responses with POST andGET for creating and retrieving resources PUT forupdating and changing the state of resources andDELETE for removing resourcesAdvantages of REST over arbitrary web services(SOAP) in the IoT context

(i) It reduces overhead(ii) It reduces parsing complexity(iii) It is stateless(iii) It integrates more tightly with HTTP [58](iv) It is easy to implement and learnis together

with its natural t over HTTP makes it apreferred method for Web 20 applications toshowcase their data and secure M2M sessionsin the IoT

(v) Applications that support RESTful web ser-vices exhibit better performance than the re-source-constrained WSN nodes [59]

(vi) REST is important for the IoT since it supportscommercial M2M cloud platforms

(vii) Its implementation in smart phones and tabletapps is also easy since it needs only an HTTPlibrary that is accessible to all operating systemplatforms [56]

Disadvantages

(i) Although it enjoys wide usage in commercialM2M platforms it is not likely to dominatebecause it is hard to implement

(ii) Its HTTP usage makes it incompatible withconstrained M2M devices leaving its usage toend applications

39 WebSockets It is a web-based protocol operating on asingle TCP channel and providing full-duplex transmissions

It permits secure duplex communication between a clientexecuting unreliable code in a restricted environment and adistant node communicating from that code while using theorigin-based security model employed by web browsers overTCP [60] e protocol is not a requestresponse model andis not a publishsubscribe model So unlike other protocolsa web-socket session is established by the client initiating ahandshake to the server to launch the session is onceestablished a full-duplex client-to-server connection beginsasynchronously and runs until terminated by either party[56] It gives a model for browser-based apps utilizing bi-directional communication with no need of many HTTPconnections to servers

WebSocket comprises two parts ie (i) the handshakehaving a clientrsquos message and the serverrsquos handshake responseand (ii) the data transfer HTTP headers should be repeated byapplications in all client requests and server responses arisingfrom the endless polling which might increase the commu-nication overhead subject to the application [61] WebSocketcan be used to come up with scalable real-time web appli-cations since it can enable full-duplex transmission via onesocket across the Web Its security can be ensured with thehelp of TLSSSL is protocol is suitable for resource-con-strained devices and IoTapplications will not have a good userexperience with it since it operates a clientserver architectureNevertheless it can outcompete all TCP-based protocolsbecause of the following advantages

(i) It is t for real-time transmission(ii) It has improved security(iii) It minimizes transmission overhead(iv) It can provide ecient messaging systems using

WebSocket ApplicationMessaging Protocol (WAMP)[17]

310 Streaming (Simple) Text-Oriented Messaging Protocol(STOMP) is is a TCP-based protocol designed for textmessage-oriented middleware (MOM) with the use of

Topic

Datareader

Datareader

Datareader

Datawriter

Datawriter

Datawriter

Subscriber PublisherPublisher

Data bus (DDS domain)

Subscriber

Topic Topic

Figure 7 DDS architecture

Journal of Computer Networks and Communications 23

HTTP-like commands for interoperability among platformslanguages and brokers Data communication takes placebetween the client and the broker in line with commands(like CONNECT DISCONNECT ACK NACK SUB-SCRIBE UNSUBSCRIBE SEND BEGIN COMMIT orABORT) then headers in the form ltkeygt ltvaluegt (one perline) then a blank line and then body content terminatingin a null character Server-client communication takes placeover a MESSAGE RECEIPT or ERROR frame with headerand body content format being analogous MOM productssupporting STOMP include Apache ActiveMQ or FuseMessage Broker HornetQ Open Message Queue(OpenMQ) RabbitMQ and syslog-ng through its STOMPdestination plugin [62] In addition to these other appli-cation layer protocols include simple media control protocol(SMCP) lightweight local automation protocol (LLAP)simple sensor interface (SSI) lightweight M2M (LWM2M)XMPP-IOT and simple object access protocol (SOAP)

Advantages

(i) It is lightweight for constrained networks(ii) It has flexibility to choose quality of services with

the given functionality(iii) It is standardized by OASIS Technical Committee(iv) It is easy and quick to implement

Challenges and opportunities

Here security threats are imminent because of the open-endedness of the application layer [63] -ese could includethe following

(i) -ere is poor or lack of a security design for anapplicationrsquos function Coming up with a gooddesign to address this could be a good ideaopportunity

(ii) Some of the programs might have backdoorsevading otherwise secure controls to allow un-authorized access to network resources We requiremechanisms to patch up such loopholes

(iii) Applications that lack or have weak authenticationmechanisms can be easily targeted by unlicensedusers to abuse the network Such mechanisms needto be strengthened further

(iv) Excessively complicated access control rules maybecause of their misunderstanding or poor writinggive unauthorized access allowing dubious activityrather than protecting the network they purport toprotect -ese should be simple and clear to un-derstand by all users

(v) In booting workstations without disks and managingnetwork devices we usually use the TFTP protocolalthough access does not require authenticating theuser by usernamepassword -is makes an intruderable to gain easy access to configaccess info as longas he can guess the filenames -is needs to be fixed

A few steps can be used to secure the application layerTo protect network data transmitted across the network

applications need to strengthen their encryption and au-thentication mechanisms Apps too need to enable strictcontrols of privileges for data access using good mechanismsto balance between usability and effectiveness [63] Appli-cations handling sensitive data need detailed logging andaudit capability as well as testing and reviewing

Table 3 summarizes the application layer protocols dis-cussed with their weaknesses and differences among them

4 Cross-Layer QoS Strategies

41 Service-Differentiated Real-Time Communication Scheme(SDRCS) -is is an event-driven routing protocol that routesreal-time traffic by featuring a cross-layer packet-forwardingdesign in which it integrates the real-time routing functionalitywith a new priority-based MAC scheme -e protocol ap-proximates distributed packet traversal speed for traffic clas-sification and admission control on the basis of this design Itfurther localizes decision-making by prioritizing packet for-warding to maximize the speed of packets [64]

Advantages

(i) It improves bandwidth utilization bandwidthdegradation due to unschedulable packets can beavoided

(ii) It offers a lightweight packet schedulability estima-tion mechanism Here it uses received signalstrength and admission control and then earlymissed-deadline packet-dropping policies are made

(iii) It delivers high end-to-end throughput via highersource data rates alongside strict end-to-end latencyneeds

(iv) Nodes which meet real-time requirements of lesstraffic load and better channel quality are givenhighest priority when it comes to packet forwarding

-e SDRCS operates via 5 mechanisms namely (1) RSS-(Rich Site SummaryReally Simple Syndication-) basedgrouping (2) admission control (3) prioritized queuing(4) real-time MAC and (5) dynamic forwarding -isprotocol has a queuing policy that employs per-hopdeadline-based queues and the nodes schedule theirpackets using FIFO priority-based queues [1] -e prin-cipal features of this protocol are mentioned as follows

(i) It is event driven and is easily adaptable to anynetwork changes

(ii) It does not need additional hardware for locali-zation or multichannel transmission

(iii) -e cross-layer packet-forwarding strategy can beused when transmitting multimedia traffic

(iv) It is able to efficiently circumvent any voids therein

Early deadline miss (EDM) policy drops all unscheduledpackets -e SDRCS circumvents wrong packet dropsdue to EDM since it adjusts to network changes easily

42 Sensor Fuzzy-Based Image Transport (SUIT) -is pro-tocol regulates congestion using a scheme based on fuzzy

24 Journal of Computer Networks and Communications

logic It transmits packets with reduced quality to allowablelevels if there is congestion enabling it to transmit lower-quality packetsframes without being dropped-e numberof packets delivered per second thus rises and while thisenhances video streaming apps it is more appropriatelymade for those that quickly transmit non-real-time videosuch as in video surveillance SUIT can stream JPEG pic-tures other than the traditional predictive video codingmethods that are not suitable for sensor devices because oftheir constraints of energy memory and computationalspeed Progressive JPEG-PJPEG performs better via packettransmission rate and delay though it degrades the imagequality because of congestion [65] -is protocol employscross-layer methods for interlayer data exchange though itstill employs some layering making it still possible topreserve all advantages of the layered design since it hasapplication and transport layers but has noMAC or routinglayer

For congestion detection and control SUIT bases on thepercentage number of in-bound and out-bound frames perwindow number of contenders and buffer occupancy of thenext-hop node Since buffer management has an effect oneffectiveness and QoS inWMSNs it is useful in the transportlayer so the protocol prioritizes packets for better QoS SUITreorders packets by tracking the sequence numbers of thedelivered packets since they all have a source ID framenumber and sequence number that it uses to detect anymissing packets -e protocol however does not providepacket reliability [65]

Advantages

(i) -e protocol prioritizes packets for better QoS(ii) It can transmit lower-quality packetsframes

without being dropped

Disadvantages

(i) It does not provide packet reliability

43 Network Layer QoS Support Enforced by a Cross-LayerController (NLQS) -is scheme permits packet-level servicedifferentiation as a function of throughput end-to-endpacket error rate and delay [66] (see Figure 8) -is en-hances network layer QoS It comprises a cross-layer controlunit (XLCU) for configuring and controlling networkingfunctionalities at physical MAC and network layers [18]-is depends upon the unified logic which affects choices forapplication layer needs and status of functional blocks whichdo implementation of networking functions [66]

Advantages

(i) Cross-layer interactions can be controlled withoutweakening the upgradability modularity andsimplicity of designing the system

44 Cross-Layer Signaling Shortcuts (CLASS) CLASS has ahigh level of effectiveness flexibility and comprehensive-ness According to [67] it has the following features

(i) Direct signaling among remote layers leading tofaster transmission

(ii) Lightweighted internal message format standard-ized protocols are not suitable for use in internalsignaling since they are not usually lightweight forinstance transmitting against faults in the networkCLASS requires only three fields namely destina-tion address event type and event contents

(iii) -e setup of its exterior messages is standardconsidering external signaling ICMP and TCPIPare suitable for general messages and short notifi-cations respectively

(iv) Message control protocol should ensure optimizedand organized exchange of dense simultaneousmessages across layers for high efficiency andavoidance of possible conflicts -e protocol appliesto many cross-layer signaling situations

Table 3 Differences and problems with application layer protocols

Protocol Architecture Transport QoS options Security Weaknesses

CoAP Requestresponse UDP DTLS No in-built security features DTLS does not supportmulticast

MQTT Publishsubscribe TCP TLSSSL Insufficient security at the protocol level sincepayload values are not encrypted

SMQTT Publishsubscribe TCP CPKP-ABE Key revocation and group pubsub for distributedSMQTT are still a challenge

MQTT-SN Publishsubscribe UDP TLSSSL SSLTLS suffers from attacks like BEAST CRIMERC4 and Heartbleed

XMPP RequestresponsePublishsubscribe TCP X TLSSSL Susceptible to spamming attacks and lacks worldwide

supportAMQP Publishsubscribe TCP TLSSSL Inappropriate for real-time applications

DDS Publishsubscribe TCPUDP TLSSSL Memory intensive and no open-source libraries forconstrained devices

REST Requestresponse HTTP X HTTPS Hard to implement Uses HTTP so incompatiblewith constrained apps

WebSockets PublishsubscribeClientserver TCP X TLSSSL Bad user experience for IoT apps since it runs the

clientserver architectureSTOMP Clientserver TCP HTTP -e broker can act as a bottleneck

Journal of Computer Networks and Communications 25

Advantages

(i) e protocol is scalable and has a very lowpropagation delay

(ii) It is very ecient and exible since it uses directsignaling between any two layers

(iii) It is very comprehensive and eective

45 Cross-Layer and Multipath-Based Video Transmission(CMVT) CMVT is a hybrid of application and networklayers At the application layer it carries out the encoding ofvideo streams to video data frames (I- P- and B-frames) bymeans of the MPEG-4 encoding format At the networklayer route discovery and data transfer take place Twoalgorithms greedy forwarding and rollback are used todiscover a number of paths from the source to the sink nodevia the route discovery method [20] A specied node icalculates the evaluation of its adjacent node j using thefollowing equation

fij (1 minus α)d2(j D) minus d2min(i)d2max(i) minus d

2min(i)

+ αeinit(j) minus eres(j)

einit(j)

(19)

where fij evaluation value for the node i to j d2(j D) isthe distance from the node j to the destination node D

d2min(i) and d2max(i) are the minimum and maximum dis-

tances of neighbors of the node i to D respectively einit(j) isthe initial energy of the node j eres(j) is the current residualenergy of the node j and α is the energy coecient given asfollows

α emax(i) minus emin(i)

emax(i) (20)

where emax(i) and emin(i) are respectively the maximumand minimum energy left of all node irsquos neighbors

e network layer further transmits video streams withCMVT doing status evaluation choosing an appropriatecommunication path for any packets and the QoS guaranteelevel for the path i is calculated as

fi (1 minus ω)hisum hi

+ ωnisum ni

(21)

where fi is the evaluation value for the path i hi are the hopsfor the path i ni is the summation of packets sent via isum ni isthe summation of packets sent by sources and ω is theenergy consumption factor [18]

Advantages

(i) e CMVT protocol is very superior in mediatransmission particularly large-scale WMSNs

To network peers

Flow requirements

-

Atimes

Admissioncontrol

Capacityassignment

Next-hopselection

Coding rate

Data rate

QoScontracts

Cross-layer controlunit (XLCU)

Localization subsystem

Synchronizationsubsystem

Networkcoordinationunit (NCU)

UWB modulator

Channel coder

QoS scheduler

Georouting

Source coder

Rate control

Figure 8 Cross-layer controller architecture

26 Journal of Computer Networks and Communications

46 Cross-Layer Cooperative MAC (CoopMAC)CoopMAC [68] is based on the IEEE 80211 DCF approach-ere is a reduction in interference of neighboring cellsleading to uniform coverage in densely deployed networks[68] A connection is established by two-way authenticationbetween the sender and the receiver When a Clear-to-Send(CTS) packet and short interframe space (SIFS) are obtainedby the sender packets are transmitted right to the receiverwith no valuable cooperation If available the two nodes findout whether they can exchange data -is is done by the useof a helper identification signal to establish how feasiblecooperative communication is It is established in thepresence of a signal and if it is not available direct com-munication is activated For distributed systems helper-initiated cooperation is preferable because of the presence ofthe RTSCTS packets [69]

Advantages

(i) CoopMAC has an advantage over others due to thespatial diversity amongst the three nodes [70]

(ii) It has the potential to achieve significantthroughput and delay with no complexity in thedesign of the system

47 Cooperative MAC Protocol for Multihop Networks (M-CMAC) Just as CoopMAC M-CMAC is made in such away that high data rate stations help low data rate stations toforward traffic for broadcasting [71] -ere are also helperschosen in such a way that 2 fast-hop transmissions replace aslow hop transmission -e helper via whom there existsleast delay from the sender to the receiver is considered bestbecause of theminimum two-hop transmission rate and so isthe chosen neighbor to the duo Here the working as-sumption is that all nodes have their position coordinatesknown and thus the Euclidean distances between all pairscan be calculated and converted to the data rate for that linkAll nodes have cooperative tables (CT) of potential helperswith destination and helper MAC address Euclidean dis-tance and total distance through the helper Any sender withdata to transmit checks out for the existence of any helper inthe CT for the destination and if available forwards RTS forchannel reservation for single-hop duration M-CMAC hasgot an RTS format having a shape with five fields similar tothe one given as follows

Framecontrol Duration Source

addressDestination

addressHelperaddress

-e destination nodersquos address is kept by the source in thepart for the helper address that is in turn saved as the desti-nation address Nodes inspect helper and destination addressfields on receipt of RTS whereby the node acts as the othernodersquos helper in case the helper address field is not the sameand the helper transmits CTS back to the source if at all it wantsto forward data On delivery of the CTS packet the sourcetransmits to the helper which in turn sends the packets to thedestination that also sends ACK to the helper on receipt of thepacket In M-CMAC there exists a higher level of channelreuse (parallel transmissions) due to the increased number of

nodes therein that subsequently raises availability of helpers fordata forwarding -is results in increased throughput incomparison with CoopMAC and IEEE 80211 DCF [71]

Advantages

(i) It promises a higher throughput compared toCoopMAC and IEEE 80211 DCF

48 Cluster-Based Cooperative Routing (CBCR) Protocol-e CBCR protocol has a multihop data-forwarding func-tion realized at the link layer with cooperative links that useM-CMAC -is protocol encompasses two stages namelythe routing relay selection phase and data forwarding phase

(i) Routing relay selection phase all nodes announcetheir presence to their neighbors by broadcastingperiodical beacon messages that carry the MAC ad-dress of the node Each of these constructs a relaytable that contains all neighboring nodes with whichit is able to communicate -e node further broad-casts its neighbor list in case of any change to itsentries since the previous broadcast -e MAC ad-dresses of nodes next to the node X are contained inits relay tablersquos column one and the neighbor nodersquosrow has MAC addresses for neighbors of the adjacentnode Based on its relay table each node in-dependently chooses routing relays Choosing a nodeas a relay node depends on the number of nodes itconnectsmdashit should be the highest number [71]

(ii) Data forwarding phase a node with packets totransmit needs to first verify if the receiver is in thesame cluster and if that receiver has a helper to whichthe packet is thence forwarded but in whose absencethe packet is delivered right to the receiver In casethe intended receiver is in a different cluster therelay table will be checked to see whether that re-ceiver is reachable via other routing relays If it isreachable packets will be sent to the routing relaydirectly or via the helper if at all the relay has one Incase of destination unreachable via relays packetsare multicast to all relays by the node [71]Advantages

(i) It enables multicasting

49 MAC-PHY Cross-Layer Protocol A cooperative cross-layer standard for cooperation at the physical layer in next-generation WMSNs is developed in [70] It provides acomplete MAC layer algorithm that gives a supportive shellfor the PHY-MAC layer Like CoopMAC its scheme de-pends on the middle node for communication betweennodes-eMAC layer protocol has been changed to regulateinformation exchanges at the physical layer -e receivernode receives duplicate packets from source and helpernodes to decode the data [70]

410 MAC-Centric Approach -is cross-layer protocoltargets multimedia applications by using the MPEG-4

Journal of Computer Networks and Communications 27

scheme [21] It is characterized by 4 access categories(AC3ndashAC0) according to their priority in transmission It ismeant for supporting different QoS needs in upcomingvideo applications and enabling MAC layer differentiationfor H264 partitioning [18] Because of low bandwidth delayand other QoS challenges that cause inefficiency in trans-mittingmultimedia data inWMSNs a number of algorithmsdepending on IEEE 80211e have been proposed to supporttransmission of quality videos [72] Selecting an AC dependson the measures for QoS such as the loss rate and delay-us AC3 having the highest priority is mapped to theparameter set concept since the stream is sensitive to bit lossespecially in video transmission [21]

411 Adaptive Cross-Layer Forward Error Correction(ACFEC) In the ACFEC [73] model data packets are ex-changed among nodes through the access point (AP) whichoperates the infrastructure mode in which it adds FEC to videodata [72] -ese data are dealt by encapsulation using astreaming server to the receiver through thewireless AP as RTPpackets An adaptive FEC controller senses the category ofpackets out of the RTP header and recovers the header ofpackets from UDP -e encoder generates some error-cor-recting packets whose number is determined by the sourcepacket number of the block Multimedia transmissions aremonitored by the controller using MAC failure data and itscounter is increased by one if there is a failed transmission-econtroller uses the failure counter to change the packet numberproduced after transmitting a block [18] In case there are lostpackets it changes the redundancy rates and produces extrapackets to replace the lost ones and satisfy the requirements ofthe receiving node [72] -e FEC packet number is enhancedor reduced to satisfy the receiverrsquos requirements and stoppacket losses-is is done through accurate detection of packetlosses and adjustment of redundancy rates If all video datapackets are well received there will not be generation of FECpackets [73]

Advantages

(i) It promises a good QoS through packet loss re-duction and redundancy rate adjustment

412 Balanced Cross-Layer Fuzzy Logic (BCFL) DesignRouting Algorithm in WSNs A new fuzzy logic-basedrouting algorithm (BCFL) [74] was designed using disper-sion of the cross-layer parameter as the fuzzy logic inferencesystem input Each cross-layer parameter has a dynamicweight depending on the value of dispersion -e designcomes with some innovations as follows (i) for fuzzy logicinference system input the absolute parameter value issubstituted by parameter dispersion thereby significantlyreducing algorithmic complexity (ii) dispersion does notchange with the order of magnitude according to the dis-persion formula and (iii) the weight of the parameter de-pends on the size of its dispersion and the two are inverselyproportional to each other -is enables BCFL to have somenobility unlike other algorithms as per the following dis-tinguishing properties making it advantageous

(i) It has got simple if-then rules which remain con-stant even when the constraints increase

(ii) It is capable of dealing with many constraints withno increment in complexity

(iii) It has capacity to yield a more balanced solutionthan other algorithms

(iv) It is easily adaptable to changes in network con-ditions and topology even when the changes arefrequent like in underwater WSNs -e algorithm isuseable in choosing the CH in cluster-based routingprotocols [74]

413 Minimum Hop Disjoint Multipath Routing Algorithmwith Time Slice Load-Balancing Congestion Control SchemeMHDMwTS is a two-phased routing protocol comprisingpath build-up and path acknowledgment phases It consists ofmultiple sources each of which has three build-up disjointpaths namely primary alternate and back-up paths Onactivation the source node requests to build up a route to thenearest hop neighbor which is the path build-up phase [19] Inthis phase step one has source activation in which therequested node for path building adds its number and time-stamp and sends to the least hop-count neighboring node-isgoes on until the least time latency sink having the informationneeded to construct the primary route is reached [75] In steptwo there is path extraction when the new package fromanother path arrives whereby the extracted path is comparedwith the primary path If the node is shared the package will berejected or an alternative path will be searched for to get abackup through comparison of the preceding two paths Phasetwo is path acknowledgment where we have the third step inwhich the sink returns the ACK packet to the sender with pathinformation having nodes and their time information after ithas computed it using the timestamp [1819]

Advantages

(i) -e protocol minimizes end-to-end latency(ii) It enhances congestion control

414 Cross-Layer Optimal Design (CLOD) Authors in [76]come up with a CLOD for scheduling at the data link layerrouting at the network layer and controlling congestion atthe transport layer with an assumption of fixed link capacity-rough congestion control energy efficiency is improvedTransport layer congestion at the nodes is minimized bycompressed sensing (CS) in which transmitted bits are re-duced whereas optimally allocating resources decreasescongestion on the links at the data link layer CLODpromises minimized computational complexity and betterperformance in light traffic scenarios [77] Generating andstoring CLOD is more efficient compared to Gaussianrandom matrices It prolongs network lifetime and savesenergy

Advantages

(i) It does congestion control subsequently increasingenergy efficiency

28 Journal of Computer Networks and Communications

(ii) It reduces computational complexity(iii) It improves performance in light traffic scenarios

415 Challenges and Opportunities of the Cross-Layer Design(CLD)

(i) Different layers in a real network assume differentfunctions andor services A layer only communi-cates with its neighboring layers -e layered modeldegrades system performance because of many

features in wireless transmission QoS parameters inthe layers of WMSNs are shared by routing pro-tocols to optimize the performance Neverthelessthere is a need to design cross-layer models forincreased efficiency in routing [1]

(ii) -e physical layer plays a highly vital role in CLDRate adaptation and channel allocation take place atthe physical layer via signal processing to enhanceQoS End-to-end performance in wireless media isaffected by the changes therein given the effect on

Table 4 A summary of some of the reviewed cross-layer models

Protocol Layers Aims Comments QoS parameters

SDRCS MACPHY Routing real-time traffic Transmits multimedia traffic viacross-layer packet forwarding

-roughputlatency

SUIT Applicationtransport Transmission of non-real-time video Uses fuzzy logic to regulate

congestionTransmission rate

delay

CLASS Any twoDesign serves as a framework fordifferent implementations ofdifferent application scenarios

Scalable very efficient flexible andhas very low propagation delay

Propagation delayjitter

NLQS PHYMACnetwork

Permits packet-level servicedifferentiation as a function of

throughput packet error rate anddelay

Network layer QoS is enhanced -roughput delaypacket error rate

CMVT Applicationnetwork

Encodes video streams to video dataframes via MPEG-4 encoding anddoes route discovery and data

transfer

Greedy forwarding and rollback areused to find source-to-sink paths Energy

CoopMAC PHYMAC Offers spatial diversity among thethree nodes

Helper ID signal finds out howfeasible cooperative communication

is

-roughput ratedelay

M-CMAC PHYMAC Increases end-to-end throughputand packet delivery ratio

Euclidean distances between nodesare calculated and converted to the

data rate for a link

-roughputpacket delivery

ratio

CBCR PHYMACMinimizes control overhead andtime consumed in establishing thecooperative paths than M-CMAC

Energy consumption is moreuniformly distributed in a network

enhancing network lifetime

-roughputpacket deliveryratio energy

MAC-PHY MACPHYMeant to enable PHY layer

cooperation and maximize gains ofcooperation at the MAC layer

Leverages both spatial diversity andcoding gain -roughput delay

MAC-centric MACAPPL

Meant to support QoS needs in newvideo apps and enable MAC layer

differentiation for H264partitioning

Targets multimedia applications byusing the MPEG-4 scheme

Delay packet lossrate

ACFEC MACnetwork(UDP)

Meant to enhance the quality ofvideo streaming over 80211WLANs

and overcome packet losses

Adjusts redundancy rates toovercome channel fluctuations and

detect and reduce packet lossPacket loss rate

MHDMwTS mdash

Meant to provide reliable datatransfer with the multipath routingand load-balancing congestioncontrol method in WMSNs

More reliable than basic routingschemes for transport multimedia

data

Latency packagetransmit rate

BCFL mdash

Introduces dispersion into fuzzylogic-based routing and sets every

cross-layer parameter with adynamic weight

Can be used to select a CH in cluster-based routing protocols andproposes a dispersion formula

Node utilitydispersion

CLODDatalinknetworktransport

Prolongs network lifetime andachieves congestion control anddesigned for lightly loaded WSNs

Assumes fixed link capacity andintegrates compressed sensing

technology

-roughputaverage energy CS

error ratio

Journal of Computer Networks and Communications 29

the purposefulness of the protocols at the networklayer CLD offers solutions to conserving powerminimizing energy and controlling both the flowand congestion in the network as well as fault tol-erance so it is an opportunity for designers con-sidering other layers We further desire to developCLDs that shall make an impact on network oper-ation and get close attention [18]

(iii) CLD interface standardization the architecturemust deliver module functionality although thereexist some queries amongst modules about possibleinterfaces predicted by the necessity to share in-formation among remote protocol layers Technicalchallenges include developing designing andstandardizing of cross-layer interfaces and algo-rithms that satisfy the cross-layer optimization re-quirements among protocol layers

Table 4 summarizes these cross-layer models with a fewcomments

5 Future Research Direction

51 At theMACLayer In WMSNs achieving duty cycling istricky since video traffic is volatile Listening to adjacentnodes wastes energy We need to switch states from awake tosleep -is requires further research

Self-organizing networks and green communicationalthough most CRWSN energy-saving algorithms aredesigned to reduce node transmission power energyconsumption is inevitable during the operational mode ofBSs because of internal processing Scholars are trying withpolicy-makers to encourage a shift to greenWSNs to reduceoperational costs and carbon footprints With self-orga-nizing CRWSNs for automatic switch-off when idle energyis saved which calls for cognitive switching algorithms Sointegrating cognitive switching with radio resource allo-cation in CRWSNs is a promising research area [78]

52 At the Application Layer Considering a WMSN ar-chitecture end users access the network via the BS to andfrom which they send their sensed data We desire to getguaranteed QoS and energy efficiency by designing a newarchitecture in the form of hardware and software whichneeds to be investigated further for a fix of the same [1]

QoS and energy efficiency are very crucial in WMSNsand mostly real-time apps that require guaranteed band-width and throughput in their network lifetime Mostprotocols ignore base station (BS) mobility and WMSNnodes Traffic management telemedicine and battlefieldsurveillance apps require mobile nodes or BSs so designingdynamic routing protocols to be adaptable in these cir-cumstances is necessary

53 Cross-Layer Security of data transmitted on WMSNseg in military surveillance and e-commerce is key withQoS and energy efficiency WMSNs are vulnerable to attacks

like worms and sinkhole Moreover their computationpower is not that high complicating implementation ofstrong secure protocols -is needs further attention

For fairness and priority issues in CRWSNs nodes areassigned diverse priorities depending on importance andurgency of their data For better network performance thereis a need to develop fair resource allocation mechanismswith priority-based fairness in sensors-ere is still potentialin this research area [78]

We must develop multiobjective adaptive protocols tooptimize QoS metrics involved in routing for best trans-mission results to get a trade-off between different opti-mization metrics We need to design standardized cross-layer algorithms which are able to meet cross-layer opti-mization standards

6 Conclusion

We have reviewed the different QoS strategies for WSNs inthe context of IoT from the MAC layer and application layeras well as the cross-layer paradigm For the MAC layer wehave reviewed protocols for WSNs such as the hybrid of thecontention-free and contention-based MAC protocols andthose for WMSNs We further reviewed a number of ap-plication layer protocols including many machine-to-ma-chine requestresponse and publishsubscribe protocols Forsystem optimization cross-layer QoS strategies are veryimportant in wireless communication We have reviewed anumber of cross-layer strategies For all categories reviewedchallenges and opportunities are discussed Finally somepossible future directions are discussed for research andapplication showing a promising potential for future re-search in this relatively new area especially as more WMSNapplications emerge recently

Conflicts of Interest

-e authors hereby declare no conflicts of interest regardingthe publication of this work

Acknowledgments

-is research was funded by the National Natural ScienceFoundation of China (Nos 61602398 61672447 and61711540306)

References

[1] V Bhandary A Malik and S Kumar ldquoRouting in wirelessmultimedia sensor networks a survey of existing protocolsand open research issuesrdquo Journal of Engineering vol 2016Article ID 9608757 27 pages 2016

[2] I F Akyildiz T Melodia and K R Chowdhury ldquoWirelessmultimedia sensor networks applications and testbedsrdquoProceedings of the IEEE vol 96 no 10 pp 1588ndash1605 2008

[3] M Iqbal M Naeem A Ahmed M Awais A Anpalagan andA Ahmad ldquoSwarm intelligence based resource managementfor cooperative cognitive radio network in smart hospitalsrdquoWireless Personal Communications vol 98 no 1 pp 571ndash5922018

30 Journal of Computer Networks and Communications

[4] S A Alvi B Afzal G A Shah L Atzori and W MahmoodldquoInternet of multimedia things vision and challengesrdquo AdHoc Networks vol 33 pp 87ndash111 2015

[5] Q F Hassan A U R Khan and S A Madani Internet ofings Challenges Advances and Applications CRC PressTaylor amp Francis Group A Chapman amp Hall Group BocaRaton FL USA 2018

[6] Q Wang Y Zhao W Wang et al ldquoMultimedia IoT systemsand applicationsrdquo in Proceedings of the 2017 Global Internet ofings Summit (GIoTS) vol 2 Geneva Switzerland June 2017

[7] A Aslam and E Curry ldquoTowards a generalized approach fordeep neural network based event processing for the internet ofmultimedia thingsrdquo IEEE Access vol 6 pp 25573ndash25587 2018

[8] M Jridi T Chapel V Dorez L Bougeant and A Le BotlanldquoSoC-based edge computing gateway in the context of theinternet of multimedia things experimental platformrdquoJournal of Low Power Electronics and Applications vol 8no 1 p 1 2018

[9] H Noura A Chehab L Sleem andM N Rapha ldquoOne roundcipher algorithm for multimedia IoT devicesrdquo MultimediaTools and Applications vol 77 no 14 pp 18383ndash18413 2018

[10] T Balan D Robu and F Sandu ldquoMultihoming for mobileinternet of multimedia thingsrdquo Mobile Information Systemsvol 2017 Article ID 6965028 16 pages 2017

[11] H Shen and G Bai ldquoRouting in wireless multimedia sensornetworks a survey and challenges aheadrdquo Journal of Networkand Computer Applications vol 71 pp 30ndash49 2016

[12] A Ahmad M H Rehmani H Tembine O A Mohammedand A Jamalipour ldquoIEEE access special section Editorial optimization for emerging wireless Networks IoT 5G andsmart grid communication networksrdquo IEEE Access vol 5pp 2096ndash2100 2017

[13] B A Muzakkari M A Mohamed M F A Kadir andZ Mohamad ldquoRecent advances in energy efficient-QoS awareMAC protocols for wireless sensor networksrdquo InternationalJournal of Advanced Computer Research vol 8 no 38pp 212ndash228 2018

[14] M A Yigitel O D Incel and C Ersoy ldquoQoS-aware MACprotocols for wireless sensor networks a surveyrdquo ComputerNetworks vol 55 no 8 pp 1982ndash2004 2011

[15] A M Abbas and O Kure ldquoQuality of Service in mobile ad hocnetworks a surveyrdquo International Journal of Ad Hoc andUbiquitous Computing vol 6 no 2 p 75 2010

[16] M Q Shatnawi ldquoApplication layer protocols for the internetof things a surveyrdquo in Proceedings of the InternationalConference on Engineering amp MIS (ICEMIS) Agadir Mo-rocco September 2016

[17] V Karagiannis P Chatzimisios F Vazquez-gallego andJ Alonso-zarate ldquoA survey on application layer protocols forthe internet of things research motivationrdquo InternationalJournal of Advanced Research in Computer Science vol 8no 3 pp 1ndash10 2015

[18] M S Bernard T Pei Z Li and K Li ldquoQoS strategies forwireless multimedia sensor networks in the context of IoTrdquoLecture Notes of the Institute for Computer Sciences SocialInformatics and Telecommunications Engineering vol 4pp 228ndash253 2019

[19] A AlAmri and M Abdullah ldquoCross-layer quality of serviceprotocols for wireless multimedia sensor networksrdquo in Pro-ceedings of the International Conference on CommunicationManagement and Information Technology pp 649ndash658Cosenza Italy April 2017

[20] J Guo L Sun and R Wang ldquoA cross-layer and multipathbased video transmission scheme for wireless multimedia

sensor networksrdquo Journal of Networks vol 7 no 9pp 1334ndash1340 2012

[21] A Ksentini A Naimi and M Gueroui ldquoToward an im-provement of H264 video transmission over IEEE 80211ethrough a cross-layer architecturerdquo IEEE CommunicationsMagazine vol 44 no 1 pp 107ndash114 2006

[22] W Ye J Heidemann and D Estrin ldquoAn energy-efficientMAC protocol for wireless sensor networksrdquo in Proceedings ofthe 21st International Annual Joint Conference of the IEEEComputer and Communications Societies (INFOCOM 2002)New York NY USA June 2002

[23] J Polastre J Hill and D Culler ldquoVersatile low power mediaaccess for wireless sensor networks categories and subjectdescriptorsrdquo in Proceedings of the Second InternationalConference on Embedded Networked Sensor Systems Sen-Sysrsquo04 pp 95ndash107 Baltimore MA USA November 2004

[24] T Danmanee K N Nakorn and K Rojviboonchai ldquoCU-MAC a duty-cycle MAC protocol for internet of things inwireless sensor networksrdquo ECTI Transactions on ElectricalEngineering Electronics and Communications vol 16 no 2pp 30ndash43 2018

[25] T Van Dam and K Langendoen ldquoAn adaptive energy-efficientMAC protocol for wireless sensor networksrdquo in Proceedings ofthe 1st International Conference on Embedded NetworkedSensor Systems Los Angeles CA USA November 2003

[26] P Xie and J Cui ldquoR-MAC an energy-efficient MAC protocolfor underwater sensor networksrdquo in International Conferenceon Wireless Algorithms Systems and Applications (WASA2007) Chicago IL USA August 2007

[27] A Rahim N Javaid M Aslam U Qasim and Z A KhanldquoAdaptive-reliable medium access control protocol forwireless body area networksrdquo in Proceedings of the AnnualIEEE Communications Society Conference on Sensor Meshand Ad Hoc Communications and Networks (SECON) SeoulKorea June 2012

[28] H S Kwak S Ullah D H Kwak and C H Lee ldquoA powerefficient MAC protocol for wireless body area networksrdquoJournal of the Korean Institute of Intelligent Systems vol 8no 6 pp 131ndash140 2009

[29] C Omeni A J Burdett and M Park ldquoEnergy efficientmedium access protocol for wireless medical body area sensornetworksrdquo in Proceedings of the 4th IEEEEMBS InternationalSummer School and Symposium on Medical Devices andBiosensors pp 29ndash32 Cambridge UK August 2007

[30] L Sitanayah C J Sreenan and K N Brown ldquoER-MAC ahybrid MAC protocol for emergency response wireless sensornetworksrdquo in Proceedings of the Fourth International Con-ference on Sensor Technologies and Applications pp 244ndash249Mestre Italy July 2010

[31] Y Sun and D B Johnson ldquoRI-MAC a receiver-initiatedasynchronous duty cycle MAC protocol for dynamic trafficloads in wireless sensor networksrdquo in Proceedings of the 6thACM conference on Embedded network sensorsystemsmdashSenSysrsquo08 pp 1ndash14 Raleigh NC USA November2008

[32] M Buettner G V Yee E Anderson and R Han ldquoX-MAC ashort preamble MAC protocol for duty-cycled wireless sensornetworksrdquo in Proceedings of the 4th International Conferenceon Embedded Networked Sensor Systems SenSys 2006pp 307ndash320 Boulder Colorado USA October 2006

[33] S Boulfekhar and M Benmohammed ldquoSynchronous receiverinitiated MAC protocol for long-lived sensor networksrdquoComputers amp Electrical Engineering vol 40 no 2 pp 504ndash516 2014

Journal of Computer Networks and Communications 31

[34] S Mehta and K S Kwak ldquoH-MAC a hybrid Mac protocol forwireless sensor networksrdquo International Journal of ComputerNetworks amp Communications vol 2 no 2 pp 108ndash117 2010

[35] I Rhee A Warrier M Aia J Min and M L Sichitiu ldquoZ-MAC a hybrid MAC for wireless sensor networksrdquo IEEEACM Transactions on Networking vol 16 no 3 pp 511ndash5242008

[36] S Arshad A Al-sadi and A Barnawi ldquoZ-MAC performanceevaluation and enhancementsrdquo Procedia Computer Sciencevol 21 pp 485ndash490 2013

[37] Y Liu I Elhanany and H Qi ldquoAn energy-efficient QoS-aware media access control protocol for wireless sensornetworksrdquo in Proceedings of the IEEE International Confer-ence on Mobile ad hoc and Sensor Systems pp 9ndash11 Wash-ington DC USA December 2005

[38] J Decotignie ldquoWiseMAC an ultra low power MAC protocolfor the downlink of infrastructure wireless sensor networksrdquoin Proceedings of the Ninth International Symposium onComputers And Communications pp 244ndash251 AlexandriaEgypt June 2004

[39] R A Rashid ldquoDevelopment of energy aware TDMA-basedMAC protocol for wireless sensor network systemrdquo EuropeanJournal of Scientific Research vol 30 no 4 pp 571ndash578 2009

[40] H P Van Hoesel ldquoA lightweight medium access protocol(LMAC) for wireless sensor networksrdquo in Proceedings of theInternational Workshop on Networked Sensing Systems Bal-timore MD USA November 2004

[41] C Li P Wang H Chen andM Guizani ldquoA cluster based on-demand multi-channel MAC protocol for wireless multi-media sensor networksrdquo in Proceedings of the IEEE In-ternational Conference on Communications pp 2371ndash2376Beijing China May 2008

[42] J L Sobrinho and A S Krishnakumar ldquoQuality-of-service inad hoc carrier sense multiple access wireless networksrdquo IEEEJournal on Selected Areas in Communications vol 17 no 8pp 1353ndash1368 1999

[43] J P John ldquoProviding MAC QoS for multimedia traffic in80211e based multi-hop ad hoc wireless networksrdquo ComputerNetworks vol 51 no 1 pp 153ndash176 2007

[44] S Lofty and N Padmavati ldquoA survey on mobility basedprotocols in WSNsrdquo in Proceedings of the InternationalConference Communication and Manufacturing pp 12ndash15Hangzhou China July 2014

[45] S Deng L Shen and J Li ldquoMobility-based clustering pro-tocol for wireless sensor networks with mobile nodesrdquo IETWireless Sensor Systems vol 1 no 1 pp 39ndash47 2011

[46] M Khalil A Khalid F U Khan and A Shabbir ldquoA review ofrouting protocol selection for wireless sensor networks insmart citiesrdquo in Proceedings of the 24th Asia-Pacific Confer-ence on Communications (APCC) Ningbo China November2014

[47] O Message Q Telemetry and T Mqtt Committee Specifi-cation Draft 02 April 2014

[48] M Collina M Bartolucci A Vanelli-coralli andG E Corazza ldquoInternet of things application layer protocolanalysis over error and delay prone linksrdquo in Proceedings ofthe 7th Advanced Satellite Multimedia Systems Conference andthe 13th Signal Processing for Space CommunicationsWorkshop Livorno Italy September 2014

[49] A Talaminos-barroso M A Estudillo-valderrama L M RoaJ Reina-tosina and F Ortega-ruiz ldquoA machine-to-machineprotocol benchmark for e health applicationsmdashuse case re-spiratory rehabilitationrdquo Computer Methods and Programs inBiomedicine vol 129 pp 1ndash11 2016

[50] V Goyal O Pandey A Sahai and B Waters ldquoAttribute-based encryption for fine-grained access control of encrypteddatardquo in Proceedings of the 13th ACM Conference on Com-puter and Communications Security pp 89ndash98 AlexandriaVA USA October 2006

[51] M Singh M A Rajan V L Shivraj and P BalamuralidharldquoSecure MQTT for internet of things (IoT)rdquo in Proceedings ofthe Fifth International Conference on Communication Systemsand Network Technologies Gwalior India April 2015

[52] A Stanford-Clark and H L Truong MQTT for Sensor Net-works (MQTT-SN) Protocol Specification httpmqttorgnewwp-contentuploads200906MQTT-SN_spec_v12pdf 2013

[53] RFC ldquoExtensible messaging and presence protocol (XMPP)rdquoin Proceedings of the IETF pp 1ndash9 San Diego CA USAAugust 2005

[54] S Vinoski and I Technologies ldquoQueuing protocolrdquo IEEEInternet Computing vol 10 no 6 pp 87ndash89 2006

[55] J L Fernandes I C Lopes J J P C Rodrigues and S UllahldquoPerformance evaluation of restful web services and AMQPprotocolrdquo in Proceedings of the Fifth International Conferenceon Ubiquitous and Future Networks (ICUFN) pp 2ndash7 DaNang Vietnam July 2013

[56] M Asim ldquoA survey on application layer protocols for internetof things (IoT)rdquo International Journal of Advanced Researchin Computer Science vol 8 no 3 pp 996ndash1000 2017

[57] K An A Gokhale D Schmidt S Tambe P Pazandak andG Pardo-castellote ldquoContent-based filtering discovery pro-tocol (CFDP) scalable and efficient OMG DDS discoveryprotocolrdquo in Proceedings of the 8th ACM InternationalConference on Distributed Event-Based Systems New YorkNY USA July 2014

[58] E Web What is a web service pp 52ndash57 2010[59] M Laine RESTful Web Services for the Internet of ings

pp 2ndash4 2012 httpmediaTkk[60] IETF -e WebSocket protocol pp 1ndash71 2011[61] V Pimentel ldquoCommunicating and displaying real-time data

with websocketrdquo IEEE Internet Computing vol 16 no 4pp 45ndash53 2012

[62] Wikipedia 2019 httpsenwikipediaorgwikiStreaming_Text_Oriented_Messaging_Protocol2019

[63] K Raghavendra and S Nireshwalya ldquoApplication layer se-curity issues and its solutionsrdquo IJCSET vol 2 no 6pp 1266ndash1269 2012

[64] Y Xue B Ramamurthy andM C Vuran ldquoSDRCS a service-differentiated real-time communication scheme for eventsensing in wireless sensor networksrdquo Computer Networksvol 55 no 15 pp 3287ndash3302 2011

[65] C Sonmez S Isik M Y Donmez O D Incel and C ErsoyldquoSUIT a cross layer image transport protocol with fuzzy logicbased congestion control for wireless multimedia sensornetworksrdquo in Proceedings of the 5th International Conferenceon New Technologies Mobility and Security (NTMS) pp 1ndash6IEEE Istanbul Turkey May 2012

[66] T Melodia and I Akyildiz ldquoCross-layer QoS-aware com-munication for ultra wide band wireless multimedia sensornetworksrdquo IEEE Journal on Selected Areas in Communica-tions vol 28 no 5 pp 653ndash663 2010

[67] Q Wang and M A Abu-Rgheff ldquoCross-layer signalling fornext-generation wireless systemsrdquo in Proceedings of the 2003IEEE Wireless Communications and Networking 2003WCNC 2003 vol 2 pp 1084ndash1089 IEEE New Orleans LAUSA March 2003

[68] P Liu Z Tao S Narayanan T Korakis and S PanwarldquoCoopMAC a cooperative MAC for wireless LANsrdquo IEEE

32 Journal of Computer Networks and Communications

Journal on Selected Areas in Communications vol 25 no 2pp 340ndash354 2007

[69] H Shan H T Cheng and W Zhuang ldquoCross-layer co-operative MAC protocol in distributed wireless networksrdquoIEEE Transactions onWireless Communications vol 10 no 8pp 2603ndash2615 2011

[70] F Liu T Korakis Z Tao and S Panwar ldquoAMAC-PHY cross-layer protocol for wireless Ad-Hoc networksrdquo in Proceedingsof the 2008 IEEE Wireless Communications and NetworkingConference pp 1792ndash1797 IEEE Las Vegas NV USA March2008

[71] L Jacob and H R Shamna ldquoEfficient cooperative MAC androuting in wireless networksrdquo Transactions on Networks andCommunications vol 3 no 5 2015

[72] S Rao and K Shama ldquoCross layer protocols for multimediatransmission in wireless networksrdquo International Journal ofComputer Science amp Engineering Survey vol 3 no 3pp 15ndash28 2012

[73] L Han S Park S Kang and H Peter ldquoAn adaptive cross-layer FEC mechanism for video transmission over 80211WLANsrdquo in Proceedings of the 2009 International Conferenceon Internet pp 209ndash215 Las Vegas NV USA December2009

[74] N Li J Martınez and V H Dıaz ldquo-e balanced cross-layerdesign routing algorithm in wireless sensor networks usingfuzzy logicrdquo Sensors vol 15 no 8 pp 19541ndash19559 2015

[75] G Sun J Qi Z Zang and Q Xu ldquoA reliable multipathrouting algorithm with related congestion control scheme inwireless multimedia sensor networksrdquo in Proceedings of the2011 3rd International Conference on Computer Research andDevelopment vol 4 pp 229ndash233 IEEE Shanghai ChinaMarch 2011

[76] M Li Y Jing and C Li ldquoA robust and efficient cross-layeroptimal design in wireless sensor networksrdquoWireless PersonalCommunications vol 72 no 4 pp 1889ndash1902 2013

[77] J Yan M Zhou and Z Ding ldquoRecent advances in energy-efficient routing protocols for wireless sensor networks areviewrdquo IEEE Access vol 4 pp 5673ndash5686 2016

[78] A Ahmad S Ahmad M H Rehmani and N U Hassan ldquoAsurvey on radio resource allocation in cognitive radio sensornetworksrdquo IEEE Communications Surveys amp Tutorials vol 17no 2 pp 888ndash917 2015

Journal of Computer Networks and Communications 33

International Journal of

AerospaceEngineeringHindawiwwwhindawicom Volume 2018

RoboticsJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Active and Passive Electronic Components

VLSI Design

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Shock and Vibration

Hindawiwwwhindawicom Volume 2018

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawiwwwhindawicom

Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Control Scienceand Engineering

Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Journal ofEngineeringVolume 2018

SensorsJournal of

Hindawiwwwhindawicom Volume 2018

International Journal of

RotatingMachinery

Hindawiwwwhindawicom Volume 2018

Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Navigation and Observation

International Journal of

Hindawi

wwwhindawicom Volume 2018

Advances in

Multimedia

Submit your manuscripts atwwwhindawicom

Page 3: ReviewArticle - downloads.hindawi.com · ReviewArticle QoSStrategiesforWirelessMultimediaSensorNetworksinthe ContextofIoTattheMACLayer,ApplicationLayer,and Cross-LayerAlgorithms MuwongeSsajjabbiBernard

IoMT is ldquothe IoT-based paradigm that enables objects toconnect and exchange structured and unstructured data withone another to enable multimedia-based services and appli-cationsrdquo [3] ere is a need for vast processing powermemory and bandwidth for high QoS when transmittingmultimedia data in the IoMT in comparison with scalar datain a traditional IoT IoT system functionality shouldtherefore be upgraded to the IoMT We compare the two asdiscussed in [7]

(i) e IoT has standardized communication pro-tocols whereas the IoMTrsquos protocols arenonstandardized

(ii) In terms of QoS the IoT requires low bandwidthwhereas the IoMT requires higher bandwidth

(iii) e IoMT transmits heterogeneous multimediadata whereas IoT data transmitted have limitedheterogeneity

(iv) IoT sensor nodes consume less energy than IoMTsensor nodes

(v) IoT devices are deployed in application-dependentRFID tags but the IoMT is in video and audiosensors

(vi) In terms of service composition the IoMT has noavailable specialized middleware whereas the IoThas specialized service-oriented architecture-basedand event-based middleware

For improved QoS best eort services and higher en-ergy eciency in IoMT networks and applications enor-mous multimedia-supported routing is gaining ground inthe research arena in the WMSN area in routing protocolsalgorithms and techniques based on network architecturesand application requirements [11] Nevertheless accordingto Ahmad et al [12] the enormous resource-constrainedheterogeneous environment of the IoT challenges its ex-pansion and deployment is is because most existing IoTapps comprise overlaid deployments of wireless sensor andactuator networks in which apps cannot interact with eachother or share and reuse the few available resources Inaddition to that ecient sensing and propagation of info

C

Users

WMSN

Image sensor

CM

CM

Audio sensor

Scalar sensor

CC

Internet

Laptops

Laptop

Wireless gateways

Storage

Sink

Scalar sensor

Video sensor

Figure 1 Architecture of the WMSN (a) single-tier at and homogeneous (b) single-tier clustered and heterogeneous (c) multitierheterogeneous

Journal of Computer Networks and Communications 3

and swift response to changes in the physical world arechallenging IoT requirements [12]

14 What Has Been Done In wireless communicationsystems routing is one issue that is quite a challenge yetthere are very few surveys available especially regardingWMSN protocols [1] and more so the MAC layer appli-cation layer and cross-layer protocols although some sur-veys have been published on WSNs To the best of ourknowledge no survey combines all these aspects -is makesour survey important to add to the existing literature -eauthors in [1] discuss a number of WMSN routing tech-niques and the properties and shortcomings of theseHowever they do not discuss any application layer tech-niques Muzakkari et al [13] survey some recent WSNcontention-based scheduling-based and hybrid MACprotocols whereby they focus on the underlying principlesadvantages limitations and their applications But much asthey only focus on MAC protocols they still leave out anumber of them in their survey Yigitel et al [14] carry out asurvey on QoS-aware MAC protocols for WSNs -ey re-view the QoS challenges and views for WSNs study the QoSmechanisms categorize the state-of-the-art QoS-awareMAC protocols and also talk about the advantages anddrawbacks of the same But being a 2011 release a number ofkey protocols developed since then to date are not presented

therein Abbas and Kure [15] review various methods forQoS provisioning at the levels of routing MAC layer andcross-layer including the schemes for admission control andscheduling for QoS provisioning as well as the problems andchallenges involved But they leave out the protocols from2010 to date Shatnawi [16] and Karagiannis et al [17] reviewthe application layer protocols for the Internet of things Intheir survey they address a number of application layerprotocols which are employed for IoT for affirming a re-liable tie among objects and things -ey evaluate thereviewed protocols in terms of architecture communicationmodel security and QoS as well as the weaknesses andstrengths per reviewed protocol However they only con-centrate on application layer protocols and still leave outsome critical protocols In a recent survey Bernard et al [18]discuss a number of cross-layer QoS strategies for IoT Butbecause of space constraint we leave out some criticalprotocols and besides this paper does not talk about ap-plication layer and MAC layer protocols AlAmri andAbdullah [19] carry out a survey on cross-layer QoS pro-tocols for WMSNs in which they state that the cross-layerarchitecture is a novel idea that brings together a number oflayers for enabling integration and exchange of informationin between them with higher efficiency compared to thetraditional layered model -ey discuss and compare theavailable cross-layer WMSN protocols that cross the uses ofadjacent or nonadjacent layers But they concentrate on

Database

MMS user databases

MMS relayserver

Home location register

External servers(eg e-mail and fax)

MMS value-added servicesservers (premium content

e-mails web )

Foreign MMSC(MMS relay server)

MM

S us

er ag

ents

Internet

Figure 2 Example of the multimedia service architecture in the context of IoT

4 Journal of Computer Networks and Communications

cross-layer protocols and leave out the MAC layer andapplication layer protocols and also some cross-layer pro-tocols are left out -e surveys that have been reviewed arecompared in Table 1

-emajor contributions that make our survey importantare outlined below

15 Importance of the Survey

(i) We discuss and draw a comparison between the IoTand the IoMT by discussing the differences betweenthe two

(ii) We further discuss the newer paradigm of WMSNsand compareWSNs andWMSNs in our discussion

(iii) We make a comprehensive survey of the recentstate-of-the-art routing protocols focusing on theMAC layer the application layer and the cross-layer paradigm

(iv) We discuss the WMSN protocols as well as theWMSN-enabled protocols that have been de-veloped in the recent studies by several scholars

(v) In some of the reviewed protocols we talk about therelative advantages and disadvantages of some ofthe routing approaches for readers to be in positionto comprehend different techniques and so choosethe most appropriate technique depending on theuser requirements

(vi) In all these we cover the challenges and oppor-tunities existing for the discussed categories

(vii) We consider the possible future research trendsbefore concluding this paper

16 Challenges

(i) Since WSNs and WMSNs are usually distributed andad hoc networks their nodes are powered by batteriesmaking energy optimization a challenge that needs tobe addressed when designing their routing protocols

(ii) Considering WSNs their routing protocols do notput into consideration multimedia applicationswhich need a lot of bandwidth and processingenergy and must be transmitted in real time withutmost fidelity Extra challenges in routing areimposed in WMSNs in deployment of heteroge-neous sensors since there could be audio video andstill pictures in addition to scalar data duringtransmission Besides the heterogeneous datathereby handled all have different QoS re-quirements and with many different businessneeds different services with varying requirementsgive rise to a big challenge for the routing design

(iii) WMSNs have also got nonrechargeable batteriesplacing a limitation on the energy of the nodes yetmultimedia applications consume a lot of energyQuite a number of complications and limitingfactors exist practically despite the fact that sensorscan be self-powering as enabled by energy

harvesting Predictive video encoding using MPEG-4 and other standards is also an energy-consumingprocess and so causes enormous degradation inrouting efficiency complicating the design ofWMSN protocols even more

(iv) -e heterogeneous nature of the network is anotherchallenge since there is a difference in the categoriesof nodes needed for communication so as to enablethe facilitation of effective data collection andprocessing as well as efficient transmission It is noteasy to have a uniform communication protocolplatform because of different functionalities unlikethe case of the conventional WSNs compared to theWMSNs

(v) -ere must be a trade-off between the energy effi-ciency and the multimedia QoS when deciding onthe route taken Data aggregation and compressionrouting protocols may be applicable for energy-saving though they can lead to intolerable delays inWMSNs Furthermore there might be networkcongestion due to the many-to-many and mutualinterference in wireless routes and the scarceWMSN resources In case a given node gets so manyhigh-rate streams it lowers the performance of thenetwork and raises the possibility of node failuredue to energy drainage

-is papermainly discusses QoS strategies forWSNs andWMSNs in the context of IoT from the MAC layer appli-cation layer and cross-layer perspective -is paper presentsa state-of-the-art survey on routing in WSNs and WMSNs-e rest of this paper is organized as follows In Section 2 wediscuss the QoS strategies at the MAC layer In Section 3 wediscuss the QoS strategies at the application layer Section 4discusses the cross-layer QoS strategies and in Section 5 wegive some future research directions before concluding thispaper in Section 6

2 QoS Strategies at the MAC Layer

-e medium access control (MAC) protocol is meant toregulate admittance to the shared medium as well astransmission reliability and efficiency through error cor-rection in wireless transmissions among others It is alsoresponsible for performance of framing addressing andflow control -e MAC protocol design affects energy effi-ciency too -is is majorly categorized into contention-freeprotocols contention-based protocols and a hybrid of these

21 MAC Protocols forWSNs Hybrid MAC protocols thesecombine contention-free and contention-based MAC pro-tocols on whose advantages they capitalize Some of these arediscussed below

211 S-MAC (Sensor MAC) -is is a MAC protocoldesigned for WSNs -e main components of S-MAC in-clude (i) periodic listen and sleep (ii) collision and over-hearing avoidance and (iii) message passing Energy

Journal of Computer Networks and Communications 5

consumption and self-configuration are controlled usingthese three techniques Nodes periodically go into the sleepmode to control the energy consumed in listening to idlechannels and virtual clusters are formed by neighbor nodesfor sleep schedule autosynchronization -e protocol em-ploys in-channel signaling when the radio is in the sleepmode and message passing is applied to decrease contentionlatency for store-and-forwardWSN applications during datatransmission [22] For adjacent nodes to solve their clockdrift S-MAC requires periodic synchronization To reducesynchronization errors exchanged timestamps are relativenot absolute

Advantages

(i) S-MAC has a longer listening period than the clockerror

(ii) It has more capacity to conserve energy than IEEE80211

(iii) It can do trade-off between energy and latencydepending on traffic conditions [22]

212 B-MAC (Berkeley MAC) -is is a carrier-sense MACprotocol designed for WSNs featuring a simplistic scalableimplementation that does network variations and is effectivein channel valuation [23] -e protocol has an interface withsome flexibility leading to very low-power usage and con-trols collisions through collision avoidance mechanismsleading to effective use of the frequency band It uses asampling scheme that is adjustable to minimize duty cyclingand idle listening so as to reduce power wastage Nodes arescheduled with their duty cycle each within which they sendthe preamble to the channel in case of availability of data fortransmission In different duty-cycle schedules channelusage is checked by the destination nodes that is if thepreamble found is long it will remain powered on until thedestination address is obtained and in case it is the target itwill either await senderrsquos data or go to sleep-e protocol haslower latency than synchronous MAC protocols but has aproblem of overhearing where the neighbor node getslengthy preambles though not the intended receiver hencewasting a lot of power in doing so [24] It outperforms anumber of protocols via reconfiguration feedback andbidirectional interfaces for upper layer services It can run at

low duty cycles and its apps cannot experience synchro-nization overhead [23]

Advantages

(i) -e protocol features a simplistic scalableimplementation that does network variations andis effective in channel valuation

(ii) It has an interface with flexibility leading to verylow-power usage

(iii) It controls collisions through collision avoidancemechanisms leading to effective use of the fre-quency band

(iv) It uses an adjustable sampling scheme to minimizeduty cycling and idle listening reducing powerwastage

Disadvantages

(i) It has a problem of overhearing where the neighbornode gets lengthy preambles though not theintended receiver hence wasting a lot of power indoing so

213 T-MAC -is is a contention-based MAC protocol forWSNs that dynamically with little difficulty via fine-grainedtimeouts adapts a listensleep duty cycle It uses this dutycycle by ending its active part so as to vary load in time andposition thereby decreasing the energy wastage on idlelistening [25] When starting an active period it uses a shortlistening window thereby improving the use of energy byS-MAC -rough adaptive duty cycling T-MAC savespower but at a cost of less throughput and enhanced delayBoth S-MAC and T-MAC under similar conditions performin the same way though in variable workloads S-MAC uses 5times more power than T-MAC But T-MAC is complex andnot scalable If its active window is decreased its snoopingcapability on nearby traffic and adaptation to variableconditions of the network is reduced [23] All nodes in thisprotocol are configured to wake up periodically listen toadjacent nodes and go back into the sleep mode till the nextframe as more messages come into the queue Collisionavoidance and reliable communication are ensured throughRequest-to-Send (RTS) Clear-to-Send (CTS) and Ac-knowledgment (ACK) messages among nodes [25]

Table 1 Comparison of this survey with existing surveys

Author(s) Year CharacteristicsBhandary et al [1] 2016 Routing strategies in WMSNs properties and limitationsMuzakkari et al [13] 2018 WSN contention-based scheduling-based and hybridMAC protocolsAhmad et al [12] 2011 QoS-aware MAC protocols for WSNsMuzakkari et al [13] 2010 QoS provisioning at the levels of routing MAC layer and cross-layerShatnawi [14] 2016 IoT application layer protocolsKaragiannis et al [15] 2015 Application layer protocols for the Internet of thingsBernard et al [16] 2019 Cross-layer QoS strategies for the IoT in WMSNsAlAmri and Abdullah [17] 2017 Cross-layer QoS protocols for WMSNsGuo et al [20] 2012 Cross-layer and multipath-based video transmission for WMSNs

Ksentini et al [21] 2006 Toward an improvement of H264 video transmission over IEEE80211e through a cross-layer architecture for WMSNs

6 Journal of Computer Networks and Communications

T-MAC design the major forms of energy consump-tion in T-MAC are through (i) idle listening (ii) col-lisions (iii) protocol overhead and (iv) overhearing Incontrast in the active period the node remains lis-tening and sending data If no activation event occursfor a certain time t then the active period ends -eactivation event might be data reception on a radiocommunication sensing on a radio and end oftransmission for the data packet of the node or ACKknowing that the neighbor has finished transmitting

Clustering and synchronization on waking up a nodebegins by listening to the channel If after a certain timethere is nothing heard it chooses a schedule fortransmitting Synchronization (SYNC) packetsappended with the time of the next frame startup Ifduring this time it hears another nodersquos SYNC packetit follows its schedule subsequently transmitting itsown SYNC Following both schedules enables it to havean activation incident at the beginning of both frames

RTS and t choice if a node does not get a CTS messageafter sending an RTS then either (i) there is a collisionpreventing the destination node from hearing the RTS(ii) RTS or CTS is overheard and destination node barredfrom responding or (iii) the destination node is in thesleep mode So if after time t there is no answer thesource node switches to the sleepmode It should only doso if none of its neighbors is still communicating A newinterval t can be triggered by a neighborrsquos RTS or CTSreception A node may be out of range to hear theinitiating RTS-e length t of the interval should be longenough to get the CTS packetrsquos beginning [25] and t isthus given by the comparison in the following equation

tgtC + R + T (1)

where C contention interval length RRTS packet lengthand T turn-around time

Advantages

(i) It saves power through adaptive duty cycling

Disadvantages

(i) It is unscalable and more complex(ii) It offers less throughput and enhanced delay

214 CU-MAC -is protocol supports IoT standards re-quiring request-response transmissions It employs a multi-channel mechanism for constant bidirectional transmissionof packets at low duty cycle to resolve hidden-terminal issuesAccording to Danmanee et al [24] CU-MAC is an SI-MACprotocol that is aimed at improving bandwidth of the channelby initiating connections and transmitting data throughdifferent channels using the multichannel approach

Protocol design the protocol has the following threeproperties that make it perform more efficiently withWSNs in the IoT

(i) It enables multichannel transmission using 9useable channels in IEEE 802154

(ii) It can enable linked data transmission using oneconnection which reduces additional overheadand delay

(iii) It can support continuous transmission of packets-is is enabled depending on how the bufferoverflow protection policy is set in the buffer stack-is stack has 2 parts (a) bidirectional buffer slotsfor enabling sensor nodes to support linked datatransmission and (b) normal buffer slots fornormal data queuing Buffer stack policies are setas follows

(a) A packet is only dropped in the buffer when itexpires

(b) Unless the packet is from a target node to which areceiver is sending the data the receiver deniespacket reception when buffer slots fill up

(c) -e bidirectional buffer slot is used for bi-directional buffer transfer

CU-MAC can reduce power wastage by choosing toemploy only a single channel for transmitting both controlpackets and data in case there is only one sender in the sameregion of coverage It can further enable multiple trans-missions through another usable channel if there are othersending nodes in this area [24] -e protocol has two phases(i) advertising phase in which a source node announces toneighbors before transmission and (ii) data transfer phasewhich commences when the source node receives a Ready-to-Receive Information packet from the destination node

Advantages

(i) It has a higher packet delivery ratio compared toother MAC protocols

(ii) It performs well in high traffic situations since itsduty-cycle is lower

(iii) It reduces power wastage by employing only asingle channel for transmitting both controlpackets and data

(iv) It can enable multiple transmissions through an-other usable channel if there are other sendingnodes

215 R-MAC -is is a reservation-based MAC protocoldesigned majorly with an aim to be fair and efficient in termsof energy use To eliminate collision of data packets R-MAC[26] schedules control and data packet transmissions at thesource and destination nodes rather than using RTSCTSmessage exchanges Channel utilization in R-MAC is en-hanced by an ARQ method called burst-based acknowl-edgment which together with other scheduling algorithmsaddresses the exposed terminal problem besides enhancingthe network throughput

R-MAC design all nodes in this protocol intermittentlyfunction in listen and sleep states so as to avoid wastingenergy in overhearing and idle modes -e time takento listensleep is the same for all nodes each of which

Journal of Computer Networks and Communications 7

chooses a schedule for itself rendering central sched-uling and synchronization irrelevant So in the absenceof traffic in the locality the node simply does periodiclistening and sleeping and to eliminate collisionstransmissions are distributively synchronized using areservation-based model in case there is a sender totransmit data R-MAC differs from S-MAC in that ittransmits data to a number of nodes using the sleepmode [24] -ere are 3 stages in this protocol (i) la-tency detection (ii) period announcement and (iii)periodic operation of which (i) and (ii) are for nodesynchronization and (iii) is for listensleep procedures

(i) Latency detection phase here all nodes broadcast aNeighbor Discovery control packet (ND) at an ar-bitrarily chosen time since they are powered onWhen the ND is received from the neighbor nodethe receiver notes its arrival time and chooses arandom time to send an acknowledgment (ACK-ND)mdashthe same as the receivedND in sizemdashinwhichthe duration from ND arrival to ACK-ND trans-mission I2 is specified and the time from ND packettransmission to ACK-ND arrival I1 is computed

(a) Propagation latency between the 2 nodesL (I1 minus I2)2

(b) -us propagation latency can be defined as ldquotheinterval from the time the first node sends the firstbit of a packet to the time the second node receivesthe last bit of the packetrdquo [24]

(ii) Period announcement phase here nodes choosetheir own random listensleep time and start timeand broadcast these and when packets are re-ceived the node turns the received schedule intoits own All nodes will record their neighborsrsquoschedules in relation to theirs -ese two phasesmake a number of rounds to ensure all nodes areinformed about their neighboring nodes

(iii) Periodic operation phase here there is datatransmission [26] and nodes intermittently go intothe sleep mode and wake up One period is a listensleep cycle and nodes possess similar periods Twoparts make a period T0 listen window TL and sleepwindow TS as shown in the following equation

T0 TL + TS (2)

In this protocol the nodes make use of control packetsfor communication-ese include REV (reservation packet)ACKREV (acknowledgment packet for REV) and DATAACK-DATA (acknowledgment packet for data) messageexchanges all of which are similar in size

Advantages

(i) It is fair and efficient in terms of energy use(ii) It eliminates collision of data packets

216 AR-MAC (Adaptive-Reliable Media Access Control)-is is a TDMA-based MAC protocol for wireless body areanetworks (WBANs) designed to reduce energy consumption

as proposed by Rahim et al [27] -e protocol allocates aguaranteed time slot (GTS) to all nodes for transmissiondepending on their needs It utilizes periodic sleep andwakeup to minimize overhearing and idle listening based onthe needs of the node It has a central node (CN) having bigbatteries with higher computational power and single ordouble transceivers For double transceivers the sum of thetime frame TFrame is assigned for node transmission But inAR-MAC a single transceiver CN is considered with TFramehaving 3 subdivisions contention-free period (CFP) forsensor connection contention access period (CAP) foremergencies and time TMS to communicate with themonitoring station (MS)

Channel selection the CN first scans for redundantchannels from which it chooses one for transmissionand broadcasts its address and other information to thenodes -e destination nodes search for channels byscanning the radio frequency (RF) channels If free itswitches to the next and if busy it will wait for time TCPlistening to packets and again switch to the subsequentchannel if it does not get the channel packets When thenode finally succeeds in getting a channel packettransmission commences and an ACK packet is sent tothe CNTime slot assignment a Time Slot Request (TSR) packetis sent to the CN by the sensor node after it has selectedthe RF channel -e TSR packet has data rate and timeslot info for the node Static-size time slots with staticguard-band time TGB are proposed in [28] -isprotocol employs the adaptive algorithm of time slot(TS) and GB time and the CN allocates TS whilesending Time Slot Request Reply (TSRR) based on thenode traffic pattern -e size of the time slots variesdepending on the node requirements Based on thetransmissionmodel the data packet transmission ACKpacket reception and some amount of delay can betolerated by allocated TS To prevent interference be-tween adjacent time slots resulting from the node andCN clock drift TGB is inserted between them and itsvalue is given as follows

TGB1

FxTS1100

TGBn

FxTSn

100

TGBnn+1

F

100X12

TSn + TSn+11113858 1113859

(3)

where FGB factor depending on the mean driftvalue Nodes go to sleep after slots have been suc-cessfully assigned and wake up to transmit in assignedslots thus saving energy wasted in idle listening sinceeven allocated slots are collision freeSynchronization for periodic synchronization TDMAneeds additional energy [29] It needs a lot of energy tosynchronize nodes after several cycles To reduce en-ergy wastage and control collisions AR-MAC employs

8 Journal of Computer Networks and Communications

a new synchronization scheme When a data packetarrives the current and expected packet arrival timeswith allowable delay (D) are compared by the CN and adrift value (DV) is computed based on the differenceΔT -e DV is sent to the node in the ACK packet forchanging the slot in subsequent transmissions thoughthe value is based on allowable delay and F In caseΔTgtD the CN will send the DV in the SYNC-ACKpacket for subsequent SYNC to nodes or else a mereACK packet is sent by the CN for the received packetFor any forthcoming data transmissions the node willchange its wake-up time plan based on the DV Withsuch a sync approach a sleeping node will not lose syncfor all cycles N Acceptable delay (D) is given as

D min TS1 TSn( 1113857XF

100 (4)

where F is the guard-band factor

ΔT expected arrival time minus current arrival time

DV 0 if |ΔT|ltD

ΔT if |ΔT|gtD1113896

(5)

Frame format the AR-MAC protocol employs twoforms of packets namely (i) data packets in which anode transmits in its assigned time slot and employsCAP for emergency data and (ii) control packets thatinclude channel packet Time Slot Request (TSR)packet Time Slot Request Reply (TSRR) packet Syn-chronization-Acknowledgment (SYNC-ACK) packetData Request (DR) packet and Acknowledgment(ACK) packet [27]Energy consumption if N is the number of cyclesenergy consumption for these is measured as follows

ETotal 1113944N

k1ESleepk

+ 1113944N

k1EActivek

ESleep TSleep x ISleep x V

TSleep Tframe minus Tactive

(6)

where ISleep current from the voltage source V whenthe node is asleep If Esw is the switching energy Etransthe transmission energy Erec the receiving energy andETout the time-out energy then

EActive 2Esw + Etrans + Erec + ETout (7)

Advantages

(i) It reduces energy consumption(ii) It minimizes overhearing and idle listening based

on the needs of the node via periodic sleep andwakeup

217 ER-MAC It is a combination of TDMA and CSMAmethods making it flexible and adaptable to topological andtraffic changes and hence applicable in rapid response

situations inWSNs Collision-free slots can be planned usingTDMA [30] ER-MAC is designed in such a way that thefollowing are met

(a) It has a higher ratio of packet delivery and low delayunder lower energy consumption levels out-performing Z-MAC

(b) It keeps two priority queues in order to distinguishhigh from least priority packets

(c) It permits contention in TDMA slots so as to handleenormous traffic volumes

(d) It has a harmonized slot shape and the nodes areable to locally change their plans enabling themconnect to or exit the network with easeProtocol design it discovers the topology in such away that it uses the simple flooding mechanism tobuild the tree as initiated by the base station (BS) -etopology discovery aims at neighbor discovery andchange tracking in addition to setting up a routingtree A TOPOLOGY_DISCOVERY message com-prising hop_count new_parent_id and old_-parent_id is generated by the BS A node broadcasts itto discover if it has any potential children and aresponse to the parent In case of the desire to changethe parent the former parent is notified For all nodesthe hop count to the BS parent ID children and one-hop neighbor are registered with the BS at this stageTDMA slot assignment when a BS sends a SYNCmessage it switches to the TDMAmode which whenreceived by the child uses the broadcast slot for itschildrenrsquos synchronization Neighboring nodescannot share a slot since slots are assigned andschedules are exchanged accordingly Slot assign-ment begins with a leaf node in a bottom-up modesuch that the flow of messages to the BS may besmooth Before getting the child-node schedules anonleaf node will not transmit its data via a unicastslot descendantrsquos data via many unicast slots andchild synchronization via a broadcast slot Childnodes send SCHEDULE_NOTIFICATION mes-sages to the BS to end this phase

Local time synchronization is done via parent-childbroadcast sync since they both need to operate the same clockso that when one is transmitting the other is ready to receivePacket prioritization happens in such a way that a highpriority queue is first emptied before a lower priority queue isallowed to transmit In case a queue is filled up the shortestslack packet is dropped and then fairness over source isconsidered as queue modification is done to enable the BSbalance info from different nodes Energy conservation is bysender turning off the radio in absence of any info to transmitand the receiver will switch to the sleep mode in case nopackets are received after a certain timeout [30]

Advantages

(i) It is flexible and adaptable to topological and trafficchanges

Journal of Computer Networks and Communications 9

(ii) It is applicable in rapid response situations inWSNs

(iii) It has a higher ratio of packet delivery and lowdelay under lower energy consumption levels

218 RI-MAC (Receiver-Initiated MAC) -is is an asyn-chronous duty-cycle protocol that tries to reduce timeoccupied by the communicating nodes on the medium toagree on time for data exchange-e sending node waits forthe receiving node to signify the beginning of the com-munication session through a short beacon frame that itsends -e channel is only occupied by the beacon and datatransmission reducing channel occupancy and thus en-abling data exchange for other nodes too -is implies animprovement in channel utilization that subsequently leadsto increment in throughput packet delivery ratio andpower efficiency over a wide-ranging traffic load [31] Beinga receiver-initiated model it ensures substantial reductionin overhearing and lowers the probability of colliding andthe cost of recovery compared to B-MAC [23] or X-MAC[32] RI-MAC is further capable of using a beacon withACK and Ready-to-Receive packets though it has aproblem if a sending node with Ready packets needs to havethe radio on till a ready-to-receive node is awake [13] whichthen sends it a beacon and transmission begins immediatelyand later is acknowledged by another beacon Nodes in-termittently wake up to listen to any possible frames meantfor them depending on their schedules A node announcesits readiness to receive a frame through a broadcast beaconon waking up when it switches on the radio When a dataframe has been received the receiver will spend some moretime known as ldquodwell timerdquo in the active state so as toreceive queued packets which is the time in RI-MAC thatdepends on the number of contending senders If there areseveral contending senders that might not have beenpredicted there is a challenge in trying to reduce the re-ceiverrsquos active time to optimize power efficiency and reducethe cost of collision detection and lost data recovery even inthe case of hidden senders For this protocol this can beaddressed by a beacon frame from the receiving node forcoordinating data frame transmissions of contendingsending nodes-e protocol also beats B-MAC and X-MACin minimizing the cost of collision detection and datarecovery through coordination of the receiversrsquo dataframes -is is because the receiving node is aware of themaximum delay prior to the arrival of the frame-e abilityto broadcast is supported by RI-MAC when it sends data asunicast to each senderrsquos neighboring node or continuouslytransmits the frame in a back-to-back mode for as long asthe sleep interval [31]

Advantages

(i) It improves channel utilization leading to in-crement in throughput packet delivery ratio andpower efficiency over a wide-ranging traffic load

(ii) It ensures substantial reduction in overhearing andlowers the probability of colliding and the cost ofrecovery

219 SRI-MAC (Synchronous Receiver-Initiated MAC)-is is a synchronous duty-cycle protocol that implementsthe method of receiver-initiated data transmission with anaim of enhancing network lifetime through energy wastageavoidance via collision overhearing and idle listening To dothis and also minimize duty cycle it uses a sequence ofadaptive beacons and RTSCTS packets In this protocolsending nodes do not go to sleep till they have got CTSpackets from receiving nodes to begin transmitting exceptfor nonsending nodes that return into the sleep mode rightaway as if the channel was idle [33] Since sending nodeshave separate schedules it reduces the contention proba-bility leading to more energy conservation A time frame isdivided into the information period the allocation periodand the communication period which are discussed asfollows

Information period at frame initiation a beacon is sentby the receiving node declaring its return from sleepand readiness to accept data packets in case the channelis not busy -e beacon is a small packet alerting otherlistening nodes of the transmitterrsquos readiness to get arequest for channel It consists of the receiverrsquos ID andthe duration allocation period (DAP) that relies onnumber of the receiving nodersquos neighbors [13]Allocation period here the receiving node stays awake forsome little time after broadcasting its beacon to confirm ifthere is any node having anRTS packet to transmitWhenthe beacon has been received an RTS packet (with 3fieldsmdashsender ID receiver ID and data size) is sent by thesender prompting the receiver to send a CTS for SYNC asthere is neighbor information exchangeCommunication period SRI-MAC is capable of or-ganizing and scheduling transmissions by means ofsensor node IDs -e receiver assigns order andtransmitting time for every sender such that they switchoff their radios till their time to transmit is dueHowever the receiving node remains in the wake-upmode to get all sending nodersquos data SRI-MAC canenable sensors have lengthy sleeping times duringtransmission such that energy can be saved as long as itis neither sending nor receiving data -e protocolrsquoscommunication period is premised on the TDMAtechnique which conserves energy since the radiorsquosduty cycle is lowered and collision is eliminated [33]

Advantages

(i) Since it employs the TDMA principle it eliminatescollisions while maintaining energy conservationtoo

(ii) Noncommunicating nodes switch off their radioand get into the sleep mode which eliminatesoverhearing

(iii) Since receiving nodes broadcast SYNC signals toall nodes in the form of CTS packets overemittingis avoided

(iv) It decreases duty cycling as sleep time increasesand lowers idle listening

10 Journal of Computer Networks and Communications

2110 H-MAC -is protocol is built on the power savingmechanism (PSM) of IEEE 80211 and slotted Aloha pro-tocols to enhance performance using multiple slots Timehere comprises large frames each having an active and asleeping part [34] If any node has got data to send it will askfor slots from the receiver node during active time and latertransmit the packets in the prenegotiated slots Nodes gointo the sleep mode during sleep-time slots if they do nothave data to transmit

Latency in IEEE 80211 if N is the number of hops nthe value for the current hop tcsn the back-off time andttx the transmission delay the total latency is calculatedusing the following equation

D(N) 1113944N

n1tcsn + ttx1113872 1113873 (8)

and the average latency is given by the followingequation

E[D(N)] N tcsn + ttx1113872 1113873 (9)

-roughput in H-MAC is given by the followingequation

Th npnm

tact + Cts (10)

where np is the number of transmitted packets nm is themaximum number of nodes connected in a time frameC is the number of slots with the same length tact is thelistening time and ts is the sleep delayAdvantages

(i) It enhances performance using multiple slots

2111 Z-MAC (ZebraMAC) -is rides on the strengths andoffsets weaknesses of TDMA and CSMA It yields very goodchannel usage and low delay under low contention and lowerscollision among two-hop neighbors at a reduced cost Itsperformance is strong via error synchronization failure inassigning slots and channel conditions that change with timeIn the set-up phase the protocol operates through certainsteps including neighbor discovery slot assignment localframe exchange and global time synchronization whichunless the topology of the network changes significantly donot run again When the data are being transmitted there isimprovement in network throughput and energy efficiencywhich covers up for the initial upfront operational costs [35]

Disadvantages

(i) Initial slot assignment is quite a cumbersome taskhere since it is done at the beginning yet this maynot be at the same time in reality It further as-sumes static links until setup happens again whichtoo is impractical

(ii) It also suffers from the hidden-terminal problemsince slot owners have a small contention windowand priority for slots allocated to them due to theECN messages

(iii) Synchronizing time also poses a challenge in thisprotocol due to a large clock drift between nodes

(iv) Source and destination nodes too are not wellcoordinated in Z-MAC [36]

2112 Q-MAC (QoS-Aware Media Access Control) -isprotocol offers high QoS with improved energy efficiencyand accesses the wireless channel using MACAW as theunderlying protocol [37] Traffic from different nodes isprioritized for QoS satisfaction depending on how criticalthe data are It mainly has two scheduling algorithms inWSNs (i) intranode and (ii) internode scheduling -eintranode scheduling scheme adopts a first-in first-out-(FIFO-) based queuing algorithm based on application andMAC layer abstraction for data packet classification In-ternode scheduling reduces idle listening and collision on thechannel for improved energy consumption

Advantages

(i) It gives a high QoS with improved energy efficiency

2113 WiseMAC (Wireless Sensor MAC) WiseMAC isbased on the Aloha protocol and employs preamble sam-pling in achieving low-power transmissions in infrastructuresensor networks by occasionally sampling the channel tocheck for activity [3238] -e protocol employs the samemethod as B-MAC except that the sender shortens theextended preamblersquos length by scheduling its transmissionsaccordingly after learning the receiverrsquos awake schedules-is is done by the receiver node putting its subsequentawake time on the ACK packet such that when the senderneeds to send to it again it starts the preamble shortly beforeit wakes up so as to avoid energy wastage in transmitting thepreamble [32] -e data frame is repeated instead of theextended preamble for less traffic transmissions that have alengthier preamble than the data frame in WiseMAC Afterprocessing the frame a node that is not the target recipientwill go back into the sleep mode while the recipient staysawake till the session ends thereby sending an ACK packetFor an idle channel the power consumption is rather lowwith this method Nevertheless there is a problem of longwake-up preamble limiting throughput and causing anenormous power consumption overhead in reception [38]All nodes overhearing a transmission bear the overheadtogether with the recipient node -e protocol can addressissues concerning low-power transmission but has nomechanism of nodes adapting to varying traffic patterns[32] To compensate for drift between clock at the accesspoint and on the sensor node there is a need to compute thewake-up preamble duration -e drift varies directly withtime since previous resynchronization and is given by thefollowing equation

TP min 4θl TW( 1113857 (11)

where θ is the time-base quartz frequency tolerance and l isthe interval between two transmissions Increase in trafficvolume naturally reduces overhearing

Journal of Computer Networks and Communications 11

Radiomodel for protocols with lower power transitiondelays between transceiver states and their powerconsumption are modeled Some of the states includeDOZE where the transceiver neither transmits norreceives but is prepared to power on to the receivetransmit state very fast RXwhere the transceiver listensto the channel receives data or demodulates data fromthe noisyidle channel and TX which is the trans-mission state by the transceiver Let TS be the set-uptime needed to turn the transceiver from the DOZE tothe RX or TX state TT the turn-around time needed toswitch between RX and TX states and B the trans-ceiverrsquos bit rate PZ PR and PT are the values of powerconsumed in the DOZE RX and TX statesrespectively

1113954PR PR minus PZ

1113954PT PT minus PZ(12)

where 1113954PR and 1113954PT are the changes in power as a result ofbeing in RX and TX states respectivelyTraffic model let N be the number of sensor nodesconnected to an access point (AP) and λ be the global rateat which the downlink Poisson traffic arrives at the APfrom the fixed network Traffic toward each sensor node isthe same and is given by λN Average packet interarrivaltime Lwith which data packets arrive at a node is given byL Nλ If TC and TD are the control packets and datapackets respectively assuming a small traffic volume theglobal interarrival 1λ far exceeds the summation ofdurations for the data packet and control packet and theturn-around time Mathematically it is given as

1λ≫TD + TC + TT (13)

-e average power consumed by WiseMAC andtransmission delay incurred are given by equations (14)and (15) [38]

PW PZ +1113954PR TS +(1B)( 1113857

TW+

1113954PR X + TD + TT( 1113857

L

+ 1113954PR(N minus 1)Y

L

(14)

where

X 2θL 1 minus eminus TD4θL( )1113874 1113875

Y T2D + 12TDθL

2TW1 minus e

minus TW4θL( )1113874 1113875

DW TD +TW

21 minus e

minus TW4θL( )1113874 1113875

+ 2θL 2 minus eminus TD4θL( ) minus e

minus TW4θL( )1113874 1113875

(15)

where PW and DW are the power consumed byWiseMAC and the transmission delay for WiseMACrespectivelyAdvantages

(i) It achieves low-power transmissions in in-frastructure sensor networks through preamblesampling while checking for activity

(ii) It reduces overhearing due to increased trafficvolume

Disadvantages

(i) It has a problem of long wake-up preamble limitingthroughput and causing enormous power con-sumption overhead

2114 X-MAC It is a simple minimal transmission powerMAC protocol that uses a short preamble with decouplingfor the sending and receiving node sleep plans inWSNs-isleads to a significant reduction in energy usage at sendingand receiving nodes and decrease in per-hop latency andgives flexibility in adapting to bursty and periodic sensordata sources [32]

Protocol design X-MACrsquos design is aimed at increasingenergy efficiency reducing data latency and increasingthroughput -is enables application in different kindsof packetized and bit stream digitized radios as well assimplified low-overhead distributed application -edesign is as such in order to address the low-powerlistening problems including overhearing excessivepreamble and incompatibility with packetizing radiosas in [32]Asynchronous duty cycling a node with information totransmit will send the packets after sending a preamblewhereas the remaining nodes will preserve their sleepschedules that are not synchronized -e receiver willsample the link on waking up and stays awake if it detectsa preamble and if it is the target it will return to sleep if itis not the target after getting the whole preambleEmbedding the target ID in the preamble to avoidoverhearing the lengthy preamble is divided into manyshortened packets embedded with the receiver nodersquosID so as to eliminate overhearing One lengthy pre-amble is constituted by the shortened preamble packetstream Any node on receiving the short preamble afterwaking up will examine the target ID on the packet Itgoes back into the sleepmode right away resuming dutycycling if it is not the targeted recipient or stay awakefor successive packets if it is the one -is approachchecks energy wastage due to network density It canalso be used on several types of radiosReducing excessive preamble using strobing it ispossible to enable low-power transmission and alsosave energy by reducing time for preamble trans-mission through the use of enhanced preambles inaddition to preamble sampling X-MAC does not sendan endless preamble packet stream but puts slight

12 Journal of Computer Networks and Communications

pauses in between packets to allow the sending node topause and listen to the channel and allow the receivingnode to resend an early ACK packet during that pause-is makes the sending node begin sending datapackets and the receiving node cut excessive preamblesand thus decrease per-hop latency and energy wastageon unnecessary waiting and transmissionPacketizing radios low-power listening is limited incapability for supporting packetizing radios butX-MACrsquos short strobed preambles can support all typesof digital radiosAdaptation to traffic load while a number of WSNapplications are reliable in traffic production they stillmust adapt to changing traffic Nodes are going to haveseparate sleep schedules depending on traffic loadConsidering systems that have time-fluctuating trafficall predetermined static schedules are suboptimalOptimality Buettner et al [32] model the expectationof energy as follows

Es (preamble energy + energy perACK listen)

lowast(expected preamble minus listen iterations)

+(energy to send parket)

Es PTxSp + PRxSal1113872 1113873 Rl + Rs( 1113857

Rl minus Sp+ SdPTx

(16)

where Es expected energy to send a packet PTx andPRx are the transmission and receiver power re-spectively Sp Sal and Sd are the duration of thesenderrsquos preamble ACK listen and data transmissionperiods respectively and Rl and Rs are the receiverrsquoslisten and sleep periods respectively Expected energyto receive a packet is given as

Er (listen cycle energy + sleep cycle energy)

lowast(expected iterations for preamble arrival)

+(energy to sendACK) +(energy to receive packet)

Er PsRs + PRxRl( 1113857

1 minus 1 minus Pd(t)( 1113857 Rl + Rs( 1113857+ RaPTx + RdPRx

(17)

And the one-hop expected latency is given as

Lat (preamble duration + ACK listen)

lowast(expected number of iterations)

+ duration to send packet

Lat Sp + Sal1113872 1113873 Rl + Rs( 1113857

Rl minus Sp

(18)

Advantages

(i) It increases energy efficiency(ii) It reduces data latency(iii) It increases throughput

2115 A-MAC (Advanced MAC) A-MAC is a TDMA-based protocol whose design is aimed at enhancing thelifetime of the network for low rate and reliability of datatransmission [39] To assign time slots it does not depend ona central controller like most protocols but it gathers in-formation from its neighbors so as to allocate the slotsthrough a distributed algorithm Energy is controlled byscheduling power when there is not a single event Its designcomprises many frames with each having many time slotswhich at the beginning have got a beacon transmitted tosynchronize and exchange information with neighborsNodes to participate in the subsequent session are selectedby the controlled node so they do not go into the sleep modewith others [13] -is protocol operates under four time slotdivisions namely initial wait discover and active states Anode in the initial state begins by listening to the medium forthe beacon packet from other nodes for network synchro-nization which when received makes it change the timerthrough subtraction of beacon transmission time from re-ception time To avoid synchronization problems due todrift and allow for continuity the node has to get thestrongest beacon signal It then enters the wait state afterchoosing a number of waiting frames so as to reduce theprobability that nodes get into the discover state simulta-neously In case the synchronization beacon is lost while stillin the wait state it will return to the initial state for anotherone On expiry of the waiting counter it enters the discoverstate and collects information from neighbors by listening totheir beacon messages On successful selection of a time slotit enters the active state where it endlessly transmits beaconsat the initiation of the slot A node goes to sleep when abeacon has been transmitted and there are no more data fortransmission or in case the neighborrsquos beacon received in-dicates no data packets are coming in [39]

Advantages

(i) It can enable multicasting(ii) It enhances the lifetime of the network(iii) It offers reliable data transmission

2116 L-MAC (Lightweight MAC) Based on TDMA thisprotocol is aimed at minimizing the number of transceiverswitches adapting the nodal sleep interval to data trafficamount and making implementation less complex [40] Apacket is transmitted comprising a control and a data partControl packets have a fixed size and the time slot con-trollerrsquos ID -ey show the distance from the transmittingnode for simplified routing to the network gateway as well asaddressing target node and reporting data unit length andinternode synchronization Adjacent nodes are more in-terested in getting control packets from nearby nodes -enodersquos transceivers are switched off until the subsequenttime slot unless the node is also addressed or the message isomnicast If so it listens to the data unit which may not fillthe remaining time slot in totality-e transceivers on eitherside are turned off when the transfer of packets is successfulEnergy wastage through idle listening in uncontrolled timeslots is avoided via a short time-out interval Network nodes

Journal of Computer Networks and Communications 13

are able to communicate without colliding-is saves energyand increases the lifetime of the network A node in L-MACcan only transmit one message per frame [13] -e networkis set up in such a way that all nodes are not synchronized onbeing switched on-e gateway takes control of a timeslot soas to have them synchronized and its immediate neighborgets the control messages thereby synchronizing their timerswith the gatewayrsquos Neighboring nodes learn of multiplegatewaysrsquo time slots in their vicinity just after one frame-ejust synchronized nodes randomly choose time slots forcontrolling as long as they are not occupied -e nodes thuspreserve a table for the neighboring nodes which makes atimeslot reusable after a minimum of 3 hops and preventsmessage collision All nodes keep their slots before thebattery is down or learn of an impending collision All nodescan track their hop distances to a selected gateway node andthe information is communicated in control messages [40]

Advantages

(i) It minimizes the number of transceiver switches(ii) It adapts the nodal sleep interval to data traffic

amount(ii) It makes implementation less complex(iii) It offers improved network lifetime(iv) Energy wastage through idle listening in un-

controlled time slots is avoided via a short time-out interval

Disadvantages

(i) L-MAC is not very good for mobile SNs since slotsare calculated only once

(ii) It suffers idle listening because nodes keep listeningto slotsrsquo control parts so as to get data and enableother nodes to join the network all the time [13]

2117 C-MAC -is was designed for WSNs with 3 seg-ments namely (i) aggressive RTS having double channelcheck to assess the channel (ii) anycast to initialize the flowand (iii) convergent packet forwarding to stabilize the flowIt promises an improvement in energy efficiency lowerlatency and enhanced throughput while avoiding syn-chronization overhead though it operates only in low duty-cycling apps especially where there is less traffic But whenthere is more traffic the protocol employs anycast-basedpacket forwarding for waking the nodes up or finding aforwarder quickly and converges from route-suboptimal toroute-optimal unicast After this energy could be preservedby the nodes using a synchronized wake-up plan [13]

Advantages

(i) It improves energy efficiency(ii) It lowers latency(iii) It enhances throughput avoiding synchronization

overhead

Disadvantages

(i) It operates only in low duty-cycling applicationsespecially where there is less traffic

22 MAC Protocols for WMSNs

221 Diff-MAC -is is a CSMACA-based QoS-awarehybrid-priority-based MAC protocol for WMSNs It targetsenhancing the channel utilization while making use of ef-ficient service differentiation to do coordination for mediumaccess of each traffic class while giving fair quick datatransfer It offers QoS using the following features [14]

(i) It balances energy consumption and delay throughsensor node duty-cycling adaptation depending onthe dominant traffic class

(ii) It has a feature for fragmentation and messagepassing that breaks lengthy video frames intosmaller packets that are transmitted as a burst thusreducing retransmission costs if there are any MACfailures

(iii) -ere is fair data delivery amongst the nodes andtraffic classes respectively resulting from intranodeand intraqueue prioritization reducing unbearableperformance

(iv) -e number of collisions can be reduced and packetlatencies are kept within limits by adjusting the sizeof the congestion window in this protocol as per thetraffic needsAdvantages

(i) Diff-MAC is very suitable for WMSNs since itadapts quickly to changes in networkconditions

(ii) It is a fair protocol(iii) It is scalable

Disadvantages

(i) It is quite difficult to monitor the statistics ofthe network

(ii) Adapting dynamically to the network condi-tions is not an easy task

(iii) It exhibits some degree of latency in packetdelivery

-e implementation of Diff-MAC on Imote2 indicatesthat resource requirements can be met on the sensorhardware Results of extensive simulation runs showed thatDiff-MAC outperforms its competitors

222 Cluster-Based On-Demand Multichannel MAC Pro-tocol for WMSN (COM-MAC) -is is an energy-efficientprotocol in WMSNs with high throughput and reliable datatransmission It achieves energy efficiency by schedulingmultichannel media access to reduce channel contentionand remove collision idle listening and overhearing [41]

Design it has a three-phased operation with the (i)request phase (ii) scheduling phase and (iii) datatransmission phaseRequest phase here all nodes have two protocols eachfor sending requests to the cluster head (CH) ie thecontention-based and contention-free TDMAFDMA

14 Journal of Computer Networks and Communications

protocols -e contention-based protocol allocates achannel per node for sending the request message andfree channels are useable as control channels Forcontention-free protocols a control slot is allocatedwhen a network is deployed Here a slot per channel isallocated to a different node enabling requests to be sentwithout interfering with transmissions from othernodes -e contention-based protocol operates undertwo steps (i) Control channel assignment phase achannel is assigned per node for transmitting a request(REQ) message A channel is assigned to multiplenodes and nodes are evenly distributed to availablechannels for congestion control (ii) Request trans-mission phase nodes with data send a REQmessage viathe assigned control channel to notify the CH whichresponds with an ACK message -e contention-freeTDMAFDMA protocol too has two phases includingthe control slot assignment and request transmissionphases For unreliable channel conditions and lowtraffic load the contention-based protocol is fit for usebecause of less collisions and congestion whereas thecontention-free protocol is preferable in heavy trafficload and reliable channel applicationsScheduling phase from the above phase a schedule isgenerated by the CH for coordination and this enablesnodes to transmit and broadcast their data via controlchannels-e schedule for broadcasting only comprisesdata for nodes assigned to a channel -e schedule isrebroadcast for transmission reliability enhancementbut for energy efficiency the nodersquos transceiver isswitched off when all the schedule is completely de-livered till it is the time for transmitting again-e timeslot and data communication radio channel are allincluded in the schedule generated for each nodeData transmission phase on getting the schedule thesensor node sends its data as per the time slot and thechannel assigned -e time slot is divided into datatransmission andACK sections ACK supports link layererror control Using implicit selective repeat ARQ theCH sends an ACK message for all received packets Ifthere is no ACK the packet is assumed lost and hence aretransmission during the subsequent interval is done-is enhances reliability of transmissions though excessdelay is possible here which is also intolerable for real-time critical apps like video or audio -is gives rise to ahybrid MAC protocol to exploit the unused spectrumduring data transmission sessions [41]Advantages

(i) It achieves energy efficiency by scheduling mul-tichannel media access to reduce channel con-tention and remove collision idle listening andoverhearing

(ii) It maximizes network throughput by dynamicallyallocating time slots and channels for nodes byexecution of a traffic-adaptive QoS-aware sched-uling algorithm depending on needs for QoS andtraffic data information

(iii) It increases transmission reliability by in-corporating spectrum-aware ARQ to avoid spec-trum wastage

223 EQoSA -is protocol is developed for video andimage transfer across WMSNs in a hybrid mode withprovision of QoS Concerning the active sensor nodes andtheir traffic loads it employs dynamic session sizes andchanges bitmap-assisted (BMA) MACrsquos fixed session sizeNodes declare any data available for transmission duringcontention time -is forces the cluster head to make aschedule of how slots have been assigned It communicatesthis schedule to the nodes giving each node a given numberof slots per session and so enabling the busty traffic to beaccommodated [14] -e protocol does not perform wellwith DiffServe mechanisms but does well with schemessupporting busty traffic loads and strong cluster heads arenecessary in WMSNs that can assign and broadcast slots

224 QoS-Supported Energy-Efficient MAC (QE-MAC)Based on the IEEE 80211e standard this protocol ensuresfairness and has low latency and jitter and better efficiencyon energy usage than other QoS-aware MAC protocols forWMSNs It operates under a two-phased routine

-e first phase introduces the innovating prioritymechanism in IEEE 80211e QoS and several data types areassigned priorities Its coordination function is a combi-nation of distributed contention-based channel allocationand centralized polling-based channel access mechanisms

-e second phase preserves the nodesrsquo energy usingdynamic duty cycling -ere is an exchange of RTSCTSpackets to supplement CSMACA by muting nodes near thesenderreceiver during packet duration [42]

Advantages

(i) It ensures fairness(ii) It has low latency and jitter and better efficiency on

energy usage than other QoS-awareMAC protocolsfor WMSNs

225 Black-Burst (BB) Contention A contention-basedMAC scheme for enhancing instantaneous QoS access tocarrier-sense WSNs is proposed in [42] In this schemenodes contend for the channel until it is free Real-timepackets are given higher priority over non-real-time packetsand if a node is transmitting real-time packets it is prior-itized over other nodes [15] Voice and video are some of thereal-time apps in consideration here which periodicallyaccess the common radio channel when transmitting -eseneed constrained end-to-end delay of packets at the MAClayer -is protocol majorly does the following [42]

(i) Prioritization of real-time traffic(ii) Implementation of the round-robin mechanism in

real-time nodes(iii) Bounded access delays among real-time packets

Journal of Computer Networks and Communications 15

(iv) Implementation of real-time communication withvarying needs of bandwidth good for multimediatraffic

Amechanism for dynamically assigning packet prioritiesbased on their deadlines and the traversed hops is proposedin [43] -e ReAllocative Priority (ReAP) mechanism ismeant to provide QoS at the MAC layer to video traffic -isscheme introduces adaptive TXOP by modifying the TXOPdynamically based on the queue length [15] Other protocolsin this category include dynamic duty cycle and adaptivecontention window-based MAC protocol frame sharing(FRASH) MAC protocol and real-time independentchannels (RICH) MAC protocol

Advantages

(i) It can implement real-time communication withvarying needs of bandwidth

23 Challenges and Opportunities(a) One of the challenges in satellite transmission is

cross-talk leading to interference in different fre-quency bands It is necessary to increase the rise andfall times as well as protect sensitive nodes as wecircumvent floating ones if we are to regulate cross-talk as in satellite communication therefore Wefurther need to investigate the possibility of usingdifferential signaling -is constitutes an opportu-nity for further research

(b) Another challenge is that the hidden and exposedterminal problems are still noticeable despite the useof RTSCTS solutions in MAC since collisions arestill possible from transmissions of separate nodesAt the MAC layer ACKs might block efficientchannel usage by prohibiting exposed terminalsfrom reusing the channel and may be eliminatedleaving no proof of sender data packet receptionUsing a directional antenna is one of the proposedtechniques of high-speed WSNs like IEEE 80211-is constitutes an opportunity for further study intothese issues for better solutions though

(c) Another challenge is that since nodes in contention-based MAC protocols are mobile there are manycollisions by the packets and schedule in-consistencies in schedule-based protocols in the two-hop neighborhood information when nodes areentering or leaving [44] Deng et al [45] propose amobility-based clustering (MBC) protocol for mo-bile nodes to enhance the packet delivery rate -isprotocol in comparison with CBR has 25 chancesof decreasing collision-induced packet losses and50 chances in comparison with low-energyadaptive clustering hierarchy (LEACH)

(d) -e process of analyzing and evaluating routingperformance for WMSNs is quite complex becauseof the bursty multimedia traffic Because of the wayencoding scheme frames are structured as well as thechanges in between the scenes that occur naturally

the compressed video frames display some substantialburstiness on several time scales -ere will also befrequent route recalculation that is as well energy in-tensive for nodes frequently capturing and transmittingenergy -e solution here could as well possibly be theuse of the MBC protocol Route recalculation tooshould be dynamic to cater for the mobile nodes whichmight be in and out of the range at different instants oftime Recently the use of smart dumpsters for effectivewaste management in smart cities has also been pro-posed as a possible solution [46]

(e) For WSNs their routing protocols do not look atmultimedia applications which require enormousbandwidth and processing energy and should betransmitted in real time with utmost fidelity Extrachallenges in routing are imposed in WMSNs indeployment of heterogeneous sensors since therecould be audio video and still pictures in addition toscalar data during transmission Heterogeneous datahandled also have different QoS requirements andwith many different business needs different ser-vices with varying requirements give rise to a bigchallenge for the routing design -ere is a need todesign protocols that can support WMSNs such asEQoSA [14] for this purpose

Table 2 summarizes the reviewed MAC protocols forWSNs and the ones designed for suitability in WMSNs

3 QoS Strategies at the Application Layer

We discuss many protocols used to solve different needs ofcommunication between machines

31 ConstrainedApplication Protocol (CoAP) It is a requestresponse protocol designed by the Internet Engineering TaskForce (IETF) for constrained devices such as those havinginsufficient RAM or CPU and WPANs with low-powerconstraints resulting in poor transfer of packets andenormous overhead -e protocol was designed for ma-chine-to-machine (M2M) apps and system automation toreduce bandwidth requirements increase packet transferdecrease the overhead and make work simpler leading tolightweight implementation HTTP commands like GETPOST PUT and DELETE are used in a client-server ar-chitecture Using the publishsubscribe architecture CoAPcan support multiple users and hence yield better perfor-mance via multicasting enabling asynchronous messageexchange -is protocol can support both multicasting andunicasting via UDP and hence is a reliable and simpleprotocol -is is done via the messaging sublayer (for re-liability based on UDP) and the requestresponse sublayer(for communication and interaction) [16]

It uses the following messages

(i) Confirmable message this guarantees reliabilityusing ACK messages sent by the receiver node-ese could be synchronous or asynchronous ifthere is a need for more computational time

16 Journal of Computer Networks and Communications

(ii) Nonconfirmable message here no ACK messagesare sent back

(iii) Acknowledgment (ACK) message this confirmsreceipt of a confirmable message

(iv) Reset message this confirms receipt of an un-processed message in case any critical parts nec-essary for interpreting the message are missing

(v) Piggybacked response the receiver directly re-sponds while getting the ACK message

(vi) Separate response the reply by the receiver is dis-crete from the ACK message

CoAP has no in-built security features It is secured bythe datagram transport layer security (DTLS) that runs ontop of UDP though it was not designed for the IoT -ismakes its suitability debatable since DTLS offers no supportfor multicasting It also needs more packets for handshakingthereby increasing the traffic volume and computationalresources and so shortening the lifespan of mobile devicessince they use batteries [17]

Advantages

(i) It can support multiple users and hence yieldbetter performance via multicasting

Table 2 Comparison of MAC layer protocols used in QoS provisioning

Protocol Features QoS parameter Comments Access Priorityassignment Energy eff Designed for

S-MAC[22] Collision avoidance Energy latency Requires periodic

synchronization CSMA Hybrid Medium WSNs

B-MAC[23] Collision avoidance Energy delay

throughput Flexible interface CSMA Dynamic High WSNs

T-MAC[25] Collision avoidance Energy Complex nonscalable CSMA Dynamic High WSNs

C-MAC[13]

Dynamic channelallocation Energy Multichannel MAC CSMA Dynamic Medium WSNs

CU-MAC[24]

Multichannelmechanism

-roughputlatency High packet delivery ratio TDMA Dynamic mdash WSNs

A-MAC[39]

Per flow serviceguarantees Delay m-ary tree model

dynamic prioritiesTDMACSMA Dynamic Medium WSNs

ER-MAC[30]

Emergencyresponse

Energy delaydelivery ratio

Adapts to traffic andtopology changes

TDMACSMA Hybrid Higher than Z-

MAC WSNs

RI-MAC[31] Collision detection Delay packet

delivery ratioHigh packet delivery ratio

and throughput CSMA Dynamic Medium WSNs

SRI-MAC[33] Collision avoidance Energy delay Good data transfer and

energy consumption TDMA Dynamic High WSNs

H-MAC[34] Slotted sleep time -roughput

latencyData transmission only in

sleep timeCSMAAloha Dynamic Medium WSNs

Z-MAC[36] Collision avoidance Energy delay

throughputVery good channel usage

and error syncCSMATDMA Dynamic Medium WSNs

Q-MAC[37]

Intrainternodescheduling Energy High QoS improved

energy efficiency CSMA Hybrid High WSNs

X-MAC[32]

Overhearingavoidance

Energy delaythroughput

Reduces excessivepreamble by strobing CSMA Adaptive

static Medium WSNs

L-MAC[40] Collision avoidance Network

lifetime energyReduces the number oftransceiver switches TDMA Static Medium WSNs

WiseMAC[38]

No downlinkchannel collision Power delay Significantly low-power

consumptionAlohaCSMA Static Higher than

ZigbeeInfrastructure

WSNsAR-MAC[27] Collision avoidance Energy delay Uses data packets and

control packets TDMA Static Higher thanIEEE 802154 WBANs

BB-MAC[42]

Real-time trafficprioritization Delay Employs round robin in

real-time nodesCSMATDMA Dynamic High WMSNs

COM-MAC [41] Collision avoidance Packet delay

throughputBalances reliability and

retransmissionTDMAFDMA Dynamic High WMSNs

EQoSA[14]

Busty traffic loadaccommodation Energy Provides QoS support for

video and imagesTDMACDMA mdash High WMSNs

QE-MAC[42] Collision avoidance Energy jitter

latencyRuns central polling-like

channel access CSMA Dynamic High WMSNs

R-MAC[26] Collision avoidance Energy delay Reservation based mdash Static High Underwater

SNs

Journal of Computer Networks and Communications 17

(ii) It enables asynchronous message exchange(iii) It is a reliable and simple protocol

32 Message Queue Telemetry Transport (MQTT)Developed by IBM and later improved by OASIS with anintention of decreasing bandwidth MQTT is a simple open-source publishsubscribe protocol analogous to the client-server version It is appropriate for lightweight IoT appli-cations and M2M transmissions in constrained situationslike limited power computation capability and memory orbandwidth -e network bandwidth and the computationalresource usage are decreasing since clients are not updatedmaking publishsubscribe protocols outperform requestresponse protocols via IoT requirements [16] -is protocolcan run over TCPIP and so has the following advantages

(i) Delivering of packets with reliable guarantee(ii) Enabling multicasting(iii) Remote device session establishment(iv) Reduction of transport overhead and protocol ex-

change to minimize network traffic(v) Provision of the notification mechanism when

anomalies occur [47]

MQTTrsquos design targets economical use of bandwidthand battery so it has a major application in FacebookMessenger MQTT ensures QoS and reliability of messagedelivery by providing the following three QoS levels

(1) Fire and forget a message is sent once depending onthe networkrsquos best effort and there is no need forACK (lowest QoS level)

(2) Delivered at least once a message is sent at least onceand an ACK message is needed to avoid duplicates

(3) Delivered exactly once by a four-way handshake itensures a message is sent exactly one time (highestQoS level)-e architecture it has three key components (i)publishersmdashlightweight nodes sending data to thebroker before returning to sleep (ii) broker whichorganizes sensory data in topics and sends them tointerested subscribers and (iii) subscribersmdashIoTapps interested in data sent by sensors through thebroker as shown in Figure 3

MQTT is a simple lightweight protocol that is betterthan CoAP via low and high packet loss probability interms of throughput and exhibits lower latency in lowsampling rates [48] It supports a number of imple-mentations for embedded devices and mobile applicationsand offers web-socket support too Nevertheless it hassome shortcomings

(i) It is only employed in simple data types(ii) It exhibits high latency in high sampling rates(iii) It is unsuitable for real-time apps in such cases(iv) It does not provide automatic discovery andor

sufficient security at the protocol level [49]

33 Secure Message Queue Telemetry Transport (SMQTT)-is is an extension of MQTT that is based on lightweightattribute-based encryption using elliptic curve cryptogra-phy -e protocol has three entities as shown in Figure 4(i) publisher that publishes information under a certaintopic (ii) subscriber that obtains the same topic in-formation via a broker and (iii) public key generator(PKG) or broker that acts as the trusted third party -eprotocol has four phases namely (a) set-up phase (b)encrypt phase (c) publish phase and (d) decrypt phase

(a) Set-up phase publisher and subscriber provide aunique ID with attributes for registration with thePKG -is generates a secret master key set andpublic parameters that it publishes with set U A1A2 An For ciphertext policy-attribute-basedencryption (CP-ABE) the PKG gives a key set todevices with attributes [50]

(b) Encrypt phase based on the access tree the publishercreates an access policy An access tree is sent by thepublisher to the PKG for generating the key-gen-eration policy in the key policy- (KP-) based ABEalgorithm -e publishers know beforehand thetopics and subscribers to access as well as the rulesfor access Subscribers are issued with private keysfor all communication sessions Publishers in theCP-ABE algorithm come up with access trees andrules where the payload is encrypted by the pub-lisher who also issues more info needed for de-cryption and the rules With the help of public keycryptography data are encrypted giving rise to ci-phertext on the basis of KPCP-ABE and the receivergets SUBACK from the broker [51]

(c) Publish phase ciphertext is embedded in theSPublish command as payload and topic name isdefined in the header by the publisher -e brokerreceives the SPublish packet and replies withPUBACK prompting the publisher to respond withthe PUBREL packet -e broker then broadcasts totopic subscribers and deletes the data forwarding thePUBCOMP packet to the sender node

(d) Decrypt phase the encrypted information isdecrypted by the subscriber with the help of a secretkey Considering KP-ABE on verification that itsatisfies the access policy the subscriber invokes thePKG for a corresponding key which it sends afterverifying the request With CP-ABE subscribers usethe private key for decryption after authenticationthereby enabling offline interaction [51]Advantages

(i) -e information being encrypted makes it moresecure

34 MQTT for Sensor Networks (MQTT-SN) -is is a ver-sion of MQTT designed to operate under environments oflow bandwidth high link failures and short message lengthIt is a publishsubscribe (pubsub) protocol designed for

18 Journal of Computer Networks and Communications

UDP-based WSNs to visualize communication for low-power devices [5152] It comprises a publisher device actingas the MQTT-SN client to forward communications to agateway that also changes the MQTT-SN message to theMQTTmessage before delivering it to the broker that sendsit to MQTT-SN subscribers [51]

It has a gateway for changing protocols from MQTT toMQTT-SN (Figure 5) [52] -is protocol supports nodes in

the sleeping mode using an offline keep-alive procedure thatcan be used by battery-operated devices to get into thesleeping mode To support sleeping clients MQTT-SN has anew offline keep-alive procedure with which battery-oper-ated devices can go to the sleeping state in which theirmessages are buffered at the server until they are awake InMQTT-SN clients with no preconfigured gateway addresscan figure out the real address of their operational gateway

Publisher

Subscriber

Subscriber

Broker

Publisher

4G

Figure 3 Architecture of the MQTT protocol

Register by mentioning URI Register by mentioning URI

Access tree

Set-up

Publisher ReceiverPKG(broker)

Key management Key management

Encrypt data using public parametersand random key This is done using

CP-ABE Publish it as

Subscribe (topics)SUBACK

S-Publish (TopicEncrypt (data))S-Publish (TopicEncrypt (data))

Encryptand publish

PUBACK

PUBACK

PUBRELPUBRELPUBCOMP

PUBCOMP

Figure 4 SMQTT protocol

Journal of Computer Networks and Communications 19

within the network through a discovery procedure Simul-taneously we may have several gateways in a single wirelessnetwork that can be cooperative via load sharing

MQTT-SN architecture it is composed of MQTT-SN clients gateways and forwarders (Figure 5)rough the MQTT-SN gateway (GW) the MQTT-SN protocol is used by the MQTT-SN clients toconnect to the MQTT server Integration betweenthe MQTT-SN GW and the MQTT-SN server mightbe possible or not Translating between the MQTT-SN and the MQTT for a stand-alone GW is doneusing the MQTT-SN protocol If a GW is detachedfrom the network it can be accessed by the MQTT-SN clients through a forwarder is just encapsu-lates the received MQTT-SN frames that are on thewireless side before forwarding them to the GW un-altered It then decapsulates the received frames tooresending them to the client as sent by the GW [52]ere are 2 types of gateways (a) transparent gatewayand (b) aggregating gateway as shown in Figure 6

(i) Transparent gateway this launches and sustains adirect MQTT connection between the MQTT-SNclient and the MQTT server which is entirely re-served for end-to-end client-server messaging enumber of servers and that of MQTT-SN clientsconnected to the GW are the same is transparent

GW translates syntax between these protocols and iseasier to implement compared to the aggregatingGW But the MQTT server must support dierentsessions for all connected clients Some MQTTservers are congured to enable few parallel sessions

(ii) Aggregating gateway this maintains a single MQTTconnection to the server on behalf of all clients Allexchanged messages from the MQTT-SN client to theaggregating GW are directed to the GW which sortsthe information to forward to the server e aggre-gating GW decreases the number of MQTT sessionssimultaneously supported by the server in WSNs withmany battery-operated sensors and actuators (SAs)having limited storage and processing capabilities It ishowever more complex to implement compared to atransparent GWe general format of the MQTT-SNprotocol message has two or four message headeroctets and n octets of variable header that is optionaldepending on the message in question MQTT-SN has256 message types (MsgType) with those from0X1Eminus 0XFD 0X19 and 0XFF reserved for futureusage [51]

Advantages

(i) e protocol can operate under environmentsof low bandwidth high link failures and shortmessage length

MQTT-SN gateway

MQTT-SN

MQTT

MQTT-SN

MQTT-SN

MQTT-SN

MQTT-SN client

MQTT-SN clients

MQTT broker

MQTT-SN client

MQTT-SN client

MQTT-SN gateway

MQTT-SN forwarder

Figure 5 MQTT-SN architecture

20 Journal of Computer Networks and Communications

(ii) It is a pubsub protocol designed for UDP-basedWSNs to visualize communication for low-power devices

(iii) It can be optimal for low-cost battery-operatednodes having low processing power and lessstorage

(iv) It supports nodes in the sleeping mode using theobrvbarine keep-alive procedure used by battery-operated devices

35e Extensible Messaging and Presence Protocol (XMPP)e standardization of this protocol by the IETF made itpopular for usage over the Internet especially in instantmessaging and chatting platforms It is however susceptibleto spamming attacks and its failure to have worldwidesupport has made Google too withdraw its support of thestandard [17] e protocol is mainly useable in the client-server architecture but can also be applicable in client-clientXMPP communication [53] Among the application layerprotocols XMPP is the only one that supports both asyn-chronous (publishsubscribe) and synchronous (requestresponse) services while running over TCP but can still runover other protocols

Given the simplicity of its implementation and ease ofusability in XMPP instant messaging (IM) is suitable forXMPP and is implemented using the message stanzaltmessagegt ese stanzas are usable in one-to-one ormultiuser sessions or chatting or in error reporting andsending headlines Attribute type of the stanza denes themessage type e attribute type could be a chat group chatheadline normal or error An instant messaging network inXMPP can support many XMPP clients and servers with agateway congured to enable other IM network users toconnect to other XMPP users e XMPP server mainlycontrols sessions and XML streams to clients and servers inaddition to routing XML stanzas in the streams [53] eTCP ports 5269 and 5222 have been respectively registeredfor the XMPP server-to-server connections and XMPP clientconnections by the Internet Assigned Numbers Authority(IANA) XMPP enables specication of XMPP ExtensionProtocols (XEP) to enhance its usefulness [17]

XMPP security clients connect to the server usingtransport layer security (TLS) and simple authentication andsecurity layer (SASL) protocols which are at the same timeused by the servers for interdomain communications Fordata condentiality and integrity TLS will encrypt thestreams of XML for all sessions To conrm the identity of anyclient that needs to access the server the SASL protocol doesthe authentication of the XML stream For any communi-cating session with a server to start the client needs to resolvethe serverrsquos DNS hostname For protection of the server of theclient from third-party attacks the server maintains the au-thentication details secret ie IP address and access methodand only original connections of the server are required

In server-to-server communications the SASL protocolis employed for authentication and condentiality Server-to-server communications may as well be deactivated from aserver by an administrator depending on the organizationalpolicy To guard against domain-spoong attacks serverdialback can be a good solution since servers support itsecuring more the XML stanzas Nevertheless it cannot beused to authenticate secure or encrypt in-server streamsand the server identication resulting therein is not strongenough to be relied on It further cannot secure against DNSpoisoning attacks and IP session hijacking for remote do-mains [53] High-security domains can use TLS and SASL

Advantages

(i) It is popular for usage over the Internet especially ininstant messaging and chatting platforms

Disadvantages

(i) It is susceptible to spamming attacks(ii) It has failed to have worldwide support making

Google withdraw support of the standard(iii) It is not secure against DNS poisoning attacks and

IP session hijacking for remote domains

36 Advanced Message Queuing Protocol (AMQP) isprotocol can be utilized by a number of platforms formessage exchange since it is open source and written inmany languages It is a publishsubscribe model depending

MQTT-SNclients Broker

MQTT

(a)

MQTT-SNclients Broker

(b)

Figure 6 (a) Transparent gateway and (b) aggregating gateway

Journal of Computer Networks and Communications 21

on a reliable and efficient messaging queue with capacity forinstant messaging or e-mail for different applications es-pecially in IoT environments It differs from others in thesense that message types can be specified and have theirsource traceable and how to do trade-off between perfor-mance security and reliability For reliability of messagetransmission AMQPmakes use of some delivery guaranteeslike at most once at least once and exactly once It can alsohowever utilize the TCP transport layer for guaranteeingreliability -e AMQPrsquos publishsubscribe method has twocomponents namely (i) exchange queue meant for routingof messages to an appropriate order in the queue and (ii)message queue that acts as storage pending their delivery toreceivers [16] In this protocol model the broker can makerouting decisions contrary to othermessaging systems whereapps using the queue have in them embedded the decision-making logic for recipients or senders -e developer mayneed to check each of the affected apps so as to change thelogic for routing and delivery of the messages [54]

Advantages

(i) It is convenient(ii) It has capacity for interoperability between dif-

ferent vendors(iii) It can use open standards to enable connection

between business partners(iv) It can enable innovations based on AMQP foun-

dations [55](v) It can enable supporting of mission-critical ap-

plications in e-commerce(vi) It can guarantee reliable delivery of messages

Disadvantages

(i) It is not very appropriate for real-timeapplications

(ii) It may not offer automatic discovery(iii) Interoperability is not guaranteed(iv) It lacks open-source libraries for constrained

gadgets [49]

37 Data Distribution Service (DDS) A product of theObject Management Group (OMG) DDS is a pubsubprotocol meant for M2M communications with two sub-layers -ese are the data-centric pubsub sublayer re-sponsible for connecting anonymous data publishers tosubscribers and the data-local reconstruction sublayer that isoptional and integrates DDS into the application layer [56]-e DDS architecture connects system components asshown in Figure 7 Publishers and subscribers are discon-nected with respect to (i) time (ii) space (iii) flow andbehavior (iv) platforms and (v) programming languages[57] A DDS data publisher generates topics and publisher-subscriber coupling is represented as topic name data typeschema and publishers and subscribersrsquo QoS attributesDelivering of data is done by the publisher layer -e datawriter and publishers together make a decision about anynecessary alterations for subsequent forwarding to thesubscribers who need to send data to IoT applications Data

readers are responsible for delivering topics to the sub-scribers after reading through them [56]

Some of the major components of the DDS architectureinclude the following [57]

(i) Topic this is a logical path specifying publicationand subscriptionrsquos data type between DataWritersand DataReaders For a successful session to begintopic names types and DataWriters and Data-Readersrsquo QoS should correspond

(ii) Domain this is a logical setting for transmissionthat can be utilized for separating and optimizingnetwork transmission in a setup of similar appli-cations If DDS apps share a domain ID then theycan exchange data among themselves

(iii) DataWriter and DataReader these are terminalobjects for writing and reading messages re-spectively to and from a universal data space

(iv) Participant this is an object that stands for apublisher or subscriber of the DDS app in a do-main acting like a container for other objects

Advantages

(i) It is highly reliable and offers QoS support(ii) It is highly scalable(iii) It is appropriate for real-time apps(iv) It is fault tolerant(v) It has high level of interoperability(vi) It does not need brokers that may act as

bottlenecks(vii) It offers automatic discovery(viii) It has a user-defined data structure for topic(ix) It supports publisher and subscriber decoupling

Disadvantages

(i) It is originally developed for only stand-aloneLANs

(ii) A lot of memory is needed(iii) It has no open-source libraries to support con-

strained devices(iv) It is hard to design and configure compared to

other protocols [49]

38 RESTful Services REpresentational State Transfer (REST)-is is an architectural style based on a set of principles thatdefine and address networked resources It offers web servicesthat enable transmission from the HTTP device to the devicewithin the IoT architecture and enables clients to reach theserver [16]-is protocol utilizes HTTP commands to backupthe requestresponse messaging model HTTP is a popularand secure protocol in the WWW using TLSSSL [16] ARESTful application has the following characteristics

(i) State and functionality are separated into distrib-uted resources

(ii) All resources are distinctively named by the use ofunique HTTP commands (GET POST PUT orDELETE over the Internet)

22 Journal of Computer Networks and Communications

(iii) It is a clientserver stateless layered protocol thatsupports caching REST uses the same methods asHTTP for all requests or responses with POST andGET for creating and retrieving resources PUT forupdating and changing the state of resources andDELETE for removing resourcesAdvantages of REST over arbitrary web services(SOAP) in the IoT context

(i) It reduces overhead(ii) It reduces parsing complexity(iii) It is stateless(iii) It integrates more tightly with HTTP [58](iv) It is easy to implement and learnis together

with its natural t over HTTP makes it apreferred method for Web 20 applications toshowcase their data and secure M2M sessionsin the IoT

(v) Applications that support RESTful web ser-vices exhibit better performance than the re-source-constrained WSN nodes [59]

(vi) REST is important for the IoT since it supportscommercial M2M cloud platforms

(vii) Its implementation in smart phones and tabletapps is also easy since it needs only an HTTPlibrary that is accessible to all operating systemplatforms [56]

Disadvantages

(i) Although it enjoys wide usage in commercialM2M platforms it is not likely to dominatebecause it is hard to implement

(ii) Its HTTP usage makes it incompatible withconstrained M2M devices leaving its usage toend applications

39 WebSockets It is a web-based protocol operating on asingle TCP channel and providing full-duplex transmissions

It permits secure duplex communication between a clientexecuting unreliable code in a restricted environment and adistant node communicating from that code while using theorigin-based security model employed by web browsers overTCP [60] e protocol is not a requestresponse model andis not a publishsubscribe model So unlike other protocolsa web-socket session is established by the client initiating ahandshake to the server to launch the session is onceestablished a full-duplex client-to-server connection beginsasynchronously and runs until terminated by either party[56] It gives a model for browser-based apps utilizing bi-directional communication with no need of many HTTPconnections to servers

WebSocket comprises two parts ie (i) the handshakehaving a clientrsquos message and the serverrsquos handshake responseand (ii) the data transfer HTTP headers should be repeated byapplications in all client requests and server responses arisingfrom the endless polling which might increase the commu-nication overhead subject to the application [61] WebSocketcan be used to come up with scalable real-time web appli-cations since it can enable full-duplex transmission via onesocket across the Web Its security can be ensured with thehelp of TLSSSL is protocol is suitable for resource-con-strained devices and IoTapplications will not have a good userexperience with it since it operates a clientserver architectureNevertheless it can outcompete all TCP-based protocolsbecause of the following advantages

(i) It is t for real-time transmission(ii) It has improved security(iii) It minimizes transmission overhead(iv) It can provide ecient messaging systems using

WebSocket ApplicationMessaging Protocol (WAMP)[17]

310 Streaming (Simple) Text-Oriented Messaging Protocol(STOMP) is is a TCP-based protocol designed for textmessage-oriented middleware (MOM) with the use of

Topic

Datareader

Datareader

Datareader

Datawriter

Datawriter

Datawriter

Subscriber PublisherPublisher

Data bus (DDS domain)

Subscriber

Topic Topic

Figure 7 DDS architecture

Journal of Computer Networks and Communications 23

HTTP-like commands for interoperability among platformslanguages and brokers Data communication takes placebetween the client and the broker in line with commands(like CONNECT DISCONNECT ACK NACK SUB-SCRIBE UNSUBSCRIBE SEND BEGIN COMMIT orABORT) then headers in the form ltkeygt ltvaluegt (one perline) then a blank line and then body content terminatingin a null character Server-client communication takes placeover a MESSAGE RECEIPT or ERROR frame with headerand body content format being analogous MOM productssupporting STOMP include Apache ActiveMQ or FuseMessage Broker HornetQ Open Message Queue(OpenMQ) RabbitMQ and syslog-ng through its STOMPdestination plugin [62] In addition to these other appli-cation layer protocols include simple media control protocol(SMCP) lightweight local automation protocol (LLAP)simple sensor interface (SSI) lightweight M2M (LWM2M)XMPP-IOT and simple object access protocol (SOAP)

Advantages

(i) It is lightweight for constrained networks(ii) It has flexibility to choose quality of services with

the given functionality(iii) It is standardized by OASIS Technical Committee(iv) It is easy and quick to implement

Challenges and opportunities

Here security threats are imminent because of the open-endedness of the application layer [63] -ese could includethe following

(i) -ere is poor or lack of a security design for anapplicationrsquos function Coming up with a gooddesign to address this could be a good ideaopportunity

(ii) Some of the programs might have backdoorsevading otherwise secure controls to allow un-authorized access to network resources We requiremechanisms to patch up such loopholes

(iii) Applications that lack or have weak authenticationmechanisms can be easily targeted by unlicensedusers to abuse the network Such mechanisms needto be strengthened further

(iv) Excessively complicated access control rules maybecause of their misunderstanding or poor writinggive unauthorized access allowing dubious activityrather than protecting the network they purport toprotect -ese should be simple and clear to un-derstand by all users

(v) In booting workstations without disks and managingnetwork devices we usually use the TFTP protocolalthough access does not require authenticating theuser by usernamepassword -is makes an intruderable to gain easy access to configaccess info as longas he can guess the filenames -is needs to be fixed

A few steps can be used to secure the application layerTo protect network data transmitted across the network

applications need to strengthen their encryption and au-thentication mechanisms Apps too need to enable strictcontrols of privileges for data access using good mechanismsto balance between usability and effectiveness [63] Appli-cations handling sensitive data need detailed logging andaudit capability as well as testing and reviewing

Table 3 summarizes the application layer protocols dis-cussed with their weaknesses and differences among them

4 Cross-Layer QoS Strategies

41 Service-Differentiated Real-Time Communication Scheme(SDRCS) -is is an event-driven routing protocol that routesreal-time traffic by featuring a cross-layer packet-forwardingdesign in which it integrates the real-time routing functionalitywith a new priority-based MAC scheme -e protocol ap-proximates distributed packet traversal speed for traffic clas-sification and admission control on the basis of this design Itfurther localizes decision-making by prioritizing packet for-warding to maximize the speed of packets [64]

Advantages

(i) It improves bandwidth utilization bandwidthdegradation due to unschedulable packets can beavoided

(ii) It offers a lightweight packet schedulability estima-tion mechanism Here it uses received signalstrength and admission control and then earlymissed-deadline packet-dropping policies are made

(iii) It delivers high end-to-end throughput via highersource data rates alongside strict end-to-end latencyneeds

(iv) Nodes which meet real-time requirements of lesstraffic load and better channel quality are givenhighest priority when it comes to packet forwarding

-e SDRCS operates via 5 mechanisms namely (1) RSS-(Rich Site SummaryReally Simple Syndication-) basedgrouping (2) admission control (3) prioritized queuing(4) real-time MAC and (5) dynamic forwarding -isprotocol has a queuing policy that employs per-hopdeadline-based queues and the nodes schedule theirpackets using FIFO priority-based queues [1] -e prin-cipal features of this protocol are mentioned as follows

(i) It is event driven and is easily adaptable to anynetwork changes

(ii) It does not need additional hardware for locali-zation or multichannel transmission

(iii) -e cross-layer packet-forwarding strategy can beused when transmitting multimedia traffic

(iv) It is able to efficiently circumvent any voids therein

Early deadline miss (EDM) policy drops all unscheduledpackets -e SDRCS circumvents wrong packet dropsdue to EDM since it adjusts to network changes easily

42 Sensor Fuzzy-Based Image Transport (SUIT) -is pro-tocol regulates congestion using a scheme based on fuzzy

24 Journal of Computer Networks and Communications

logic It transmits packets with reduced quality to allowablelevels if there is congestion enabling it to transmit lower-quality packetsframes without being dropped-e numberof packets delivered per second thus rises and while thisenhances video streaming apps it is more appropriatelymade for those that quickly transmit non-real-time videosuch as in video surveillance SUIT can stream JPEG pic-tures other than the traditional predictive video codingmethods that are not suitable for sensor devices because oftheir constraints of energy memory and computationalspeed Progressive JPEG-PJPEG performs better via packettransmission rate and delay though it degrades the imagequality because of congestion [65] -is protocol employscross-layer methods for interlayer data exchange though itstill employs some layering making it still possible topreserve all advantages of the layered design since it hasapplication and transport layers but has noMAC or routinglayer

For congestion detection and control SUIT bases on thepercentage number of in-bound and out-bound frames perwindow number of contenders and buffer occupancy of thenext-hop node Since buffer management has an effect oneffectiveness and QoS inWMSNs it is useful in the transportlayer so the protocol prioritizes packets for better QoS SUITreorders packets by tracking the sequence numbers of thedelivered packets since they all have a source ID framenumber and sequence number that it uses to detect anymissing packets -e protocol however does not providepacket reliability [65]

Advantages

(i) -e protocol prioritizes packets for better QoS(ii) It can transmit lower-quality packetsframes

without being dropped

Disadvantages

(i) It does not provide packet reliability

43 Network Layer QoS Support Enforced by a Cross-LayerController (NLQS) -is scheme permits packet-level servicedifferentiation as a function of throughput end-to-endpacket error rate and delay [66] (see Figure 8) -is en-hances network layer QoS It comprises a cross-layer controlunit (XLCU) for configuring and controlling networkingfunctionalities at physical MAC and network layers [18]-is depends upon the unified logic which affects choices forapplication layer needs and status of functional blocks whichdo implementation of networking functions [66]

Advantages

(i) Cross-layer interactions can be controlled withoutweakening the upgradability modularity andsimplicity of designing the system

44 Cross-Layer Signaling Shortcuts (CLASS) CLASS has ahigh level of effectiveness flexibility and comprehensive-ness According to [67] it has the following features

(i) Direct signaling among remote layers leading tofaster transmission

(ii) Lightweighted internal message format standard-ized protocols are not suitable for use in internalsignaling since they are not usually lightweight forinstance transmitting against faults in the networkCLASS requires only three fields namely destina-tion address event type and event contents

(iii) -e setup of its exterior messages is standardconsidering external signaling ICMP and TCPIPare suitable for general messages and short notifi-cations respectively

(iv) Message control protocol should ensure optimizedand organized exchange of dense simultaneousmessages across layers for high efficiency andavoidance of possible conflicts -e protocol appliesto many cross-layer signaling situations

Table 3 Differences and problems with application layer protocols

Protocol Architecture Transport QoS options Security Weaknesses

CoAP Requestresponse UDP DTLS No in-built security features DTLS does not supportmulticast

MQTT Publishsubscribe TCP TLSSSL Insufficient security at the protocol level sincepayload values are not encrypted

SMQTT Publishsubscribe TCP CPKP-ABE Key revocation and group pubsub for distributedSMQTT are still a challenge

MQTT-SN Publishsubscribe UDP TLSSSL SSLTLS suffers from attacks like BEAST CRIMERC4 and Heartbleed

XMPP RequestresponsePublishsubscribe TCP X TLSSSL Susceptible to spamming attacks and lacks worldwide

supportAMQP Publishsubscribe TCP TLSSSL Inappropriate for real-time applications

DDS Publishsubscribe TCPUDP TLSSSL Memory intensive and no open-source libraries forconstrained devices

REST Requestresponse HTTP X HTTPS Hard to implement Uses HTTP so incompatiblewith constrained apps

WebSockets PublishsubscribeClientserver TCP X TLSSSL Bad user experience for IoT apps since it runs the

clientserver architectureSTOMP Clientserver TCP HTTP -e broker can act as a bottleneck

Journal of Computer Networks and Communications 25

Advantages

(i) e protocol is scalable and has a very lowpropagation delay

(ii) It is very ecient and exible since it uses directsignaling between any two layers

(iii) It is very comprehensive and eective

45 Cross-Layer and Multipath-Based Video Transmission(CMVT) CMVT is a hybrid of application and networklayers At the application layer it carries out the encoding ofvideo streams to video data frames (I- P- and B-frames) bymeans of the MPEG-4 encoding format At the networklayer route discovery and data transfer take place Twoalgorithms greedy forwarding and rollback are used todiscover a number of paths from the source to the sink nodevia the route discovery method [20] A specied node icalculates the evaluation of its adjacent node j using thefollowing equation

fij (1 minus α)d2(j D) minus d2min(i)d2max(i) minus d

2min(i)

+ αeinit(j) minus eres(j)

einit(j)

(19)

where fij evaluation value for the node i to j d2(j D) isthe distance from the node j to the destination node D

d2min(i) and d2max(i) are the minimum and maximum dis-

tances of neighbors of the node i to D respectively einit(j) isthe initial energy of the node j eres(j) is the current residualenergy of the node j and α is the energy coecient given asfollows

α emax(i) minus emin(i)

emax(i) (20)

where emax(i) and emin(i) are respectively the maximumand minimum energy left of all node irsquos neighbors

e network layer further transmits video streams withCMVT doing status evaluation choosing an appropriatecommunication path for any packets and the QoS guaranteelevel for the path i is calculated as

fi (1 minus ω)hisum hi

+ ωnisum ni

(21)

where fi is the evaluation value for the path i hi are the hopsfor the path i ni is the summation of packets sent via isum ni isthe summation of packets sent by sources and ω is theenergy consumption factor [18]

Advantages

(i) e CMVT protocol is very superior in mediatransmission particularly large-scale WMSNs

To network peers

Flow requirements

-

Atimes

Admissioncontrol

Capacityassignment

Next-hopselection

Coding rate

Data rate

QoScontracts

Cross-layer controlunit (XLCU)

Localization subsystem

Synchronizationsubsystem

Networkcoordinationunit (NCU)

UWB modulator

Channel coder

QoS scheduler

Georouting

Source coder

Rate control

Figure 8 Cross-layer controller architecture

26 Journal of Computer Networks and Communications

46 Cross-Layer Cooperative MAC (CoopMAC)CoopMAC [68] is based on the IEEE 80211 DCF approach-ere is a reduction in interference of neighboring cellsleading to uniform coverage in densely deployed networks[68] A connection is established by two-way authenticationbetween the sender and the receiver When a Clear-to-Send(CTS) packet and short interframe space (SIFS) are obtainedby the sender packets are transmitted right to the receiverwith no valuable cooperation If available the two nodes findout whether they can exchange data -is is done by the useof a helper identification signal to establish how feasiblecooperative communication is It is established in thepresence of a signal and if it is not available direct com-munication is activated For distributed systems helper-initiated cooperation is preferable because of the presence ofthe RTSCTS packets [69]

Advantages

(i) CoopMAC has an advantage over others due to thespatial diversity amongst the three nodes [70]

(ii) It has the potential to achieve significantthroughput and delay with no complexity in thedesign of the system

47 Cooperative MAC Protocol for Multihop Networks (M-CMAC) Just as CoopMAC M-CMAC is made in such away that high data rate stations help low data rate stations toforward traffic for broadcasting [71] -ere are also helperschosen in such a way that 2 fast-hop transmissions replace aslow hop transmission -e helper via whom there existsleast delay from the sender to the receiver is considered bestbecause of theminimum two-hop transmission rate and so isthe chosen neighbor to the duo Here the working as-sumption is that all nodes have their position coordinatesknown and thus the Euclidean distances between all pairscan be calculated and converted to the data rate for that linkAll nodes have cooperative tables (CT) of potential helperswith destination and helper MAC address Euclidean dis-tance and total distance through the helper Any sender withdata to transmit checks out for the existence of any helper inthe CT for the destination and if available forwards RTS forchannel reservation for single-hop duration M-CMAC hasgot an RTS format having a shape with five fields similar tothe one given as follows

Framecontrol Duration Source

addressDestination

addressHelperaddress

-e destination nodersquos address is kept by the source in thepart for the helper address that is in turn saved as the desti-nation address Nodes inspect helper and destination addressfields on receipt of RTS whereby the node acts as the othernodersquos helper in case the helper address field is not the sameand the helper transmits CTS back to the source if at all it wantsto forward data On delivery of the CTS packet the sourcetransmits to the helper which in turn sends the packets to thedestination that also sends ACK to the helper on receipt of thepacket In M-CMAC there exists a higher level of channelreuse (parallel transmissions) due to the increased number of

nodes therein that subsequently raises availability of helpers fordata forwarding -is results in increased throughput incomparison with CoopMAC and IEEE 80211 DCF [71]

Advantages

(i) It promises a higher throughput compared toCoopMAC and IEEE 80211 DCF

48 Cluster-Based Cooperative Routing (CBCR) Protocol-e CBCR protocol has a multihop data-forwarding func-tion realized at the link layer with cooperative links that useM-CMAC -is protocol encompasses two stages namelythe routing relay selection phase and data forwarding phase

(i) Routing relay selection phase all nodes announcetheir presence to their neighbors by broadcastingperiodical beacon messages that carry the MAC ad-dress of the node Each of these constructs a relaytable that contains all neighboring nodes with whichit is able to communicate -e node further broad-casts its neighbor list in case of any change to itsentries since the previous broadcast -e MAC ad-dresses of nodes next to the node X are contained inits relay tablersquos column one and the neighbor nodersquosrow has MAC addresses for neighbors of the adjacentnode Based on its relay table each node in-dependently chooses routing relays Choosing a nodeas a relay node depends on the number of nodes itconnectsmdashit should be the highest number [71]

(ii) Data forwarding phase a node with packets totransmit needs to first verify if the receiver is in thesame cluster and if that receiver has a helper to whichthe packet is thence forwarded but in whose absencethe packet is delivered right to the receiver In casethe intended receiver is in a different cluster therelay table will be checked to see whether that re-ceiver is reachable via other routing relays If it isreachable packets will be sent to the routing relaydirectly or via the helper if at all the relay has one Incase of destination unreachable via relays packetsare multicast to all relays by the node [71]Advantages

(i) It enables multicasting

49 MAC-PHY Cross-Layer Protocol A cooperative cross-layer standard for cooperation at the physical layer in next-generation WMSNs is developed in [70] It provides acomplete MAC layer algorithm that gives a supportive shellfor the PHY-MAC layer Like CoopMAC its scheme de-pends on the middle node for communication betweennodes-eMAC layer protocol has been changed to regulateinformation exchanges at the physical layer -e receivernode receives duplicate packets from source and helpernodes to decode the data [70]

410 MAC-Centric Approach -is cross-layer protocoltargets multimedia applications by using the MPEG-4

Journal of Computer Networks and Communications 27

scheme [21] It is characterized by 4 access categories(AC3ndashAC0) according to their priority in transmission It ismeant for supporting different QoS needs in upcomingvideo applications and enabling MAC layer differentiationfor H264 partitioning [18] Because of low bandwidth delayand other QoS challenges that cause inefficiency in trans-mittingmultimedia data inWMSNs a number of algorithmsdepending on IEEE 80211e have been proposed to supporttransmission of quality videos [72] Selecting an AC dependson the measures for QoS such as the loss rate and delay-us AC3 having the highest priority is mapped to theparameter set concept since the stream is sensitive to bit lossespecially in video transmission [21]

411 Adaptive Cross-Layer Forward Error Correction(ACFEC) In the ACFEC [73] model data packets are ex-changed among nodes through the access point (AP) whichoperates the infrastructure mode in which it adds FEC to videodata [72] -ese data are dealt by encapsulation using astreaming server to the receiver through thewireless AP as RTPpackets An adaptive FEC controller senses the category ofpackets out of the RTP header and recovers the header ofpackets from UDP -e encoder generates some error-cor-recting packets whose number is determined by the sourcepacket number of the block Multimedia transmissions aremonitored by the controller using MAC failure data and itscounter is increased by one if there is a failed transmission-econtroller uses the failure counter to change the packet numberproduced after transmitting a block [18] In case there are lostpackets it changes the redundancy rates and produces extrapackets to replace the lost ones and satisfy the requirements ofthe receiving node [72] -e FEC packet number is enhancedor reduced to satisfy the receiverrsquos requirements and stoppacket losses-is is done through accurate detection of packetlosses and adjustment of redundancy rates If all video datapackets are well received there will not be generation of FECpackets [73]

Advantages

(i) It promises a good QoS through packet loss re-duction and redundancy rate adjustment

412 Balanced Cross-Layer Fuzzy Logic (BCFL) DesignRouting Algorithm in WSNs A new fuzzy logic-basedrouting algorithm (BCFL) [74] was designed using disper-sion of the cross-layer parameter as the fuzzy logic inferencesystem input Each cross-layer parameter has a dynamicweight depending on the value of dispersion -e designcomes with some innovations as follows (i) for fuzzy logicinference system input the absolute parameter value issubstituted by parameter dispersion thereby significantlyreducing algorithmic complexity (ii) dispersion does notchange with the order of magnitude according to the dis-persion formula and (iii) the weight of the parameter de-pends on the size of its dispersion and the two are inverselyproportional to each other -is enables BCFL to have somenobility unlike other algorithms as per the following dis-tinguishing properties making it advantageous

(i) It has got simple if-then rules which remain con-stant even when the constraints increase

(ii) It is capable of dealing with many constraints withno increment in complexity

(iii) It has capacity to yield a more balanced solutionthan other algorithms

(iv) It is easily adaptable to changes in network con-ditions and topology even when the changes arefrequent like in underwater WSNs -e algorithm isuseable in choosing the CH in cluster-based routingprotocols [74]

413 Minimum Hop Disjoint Multipath Routing Algorithmwith Time Slice Load-Balancing Congestion Control SchemeMHDMwTS is a two-phased routing protocol comprisingpath build-up and path acknowledgment phases It consists ofmultiple sources each of which has three build-up disjointpaths namely primary alternate and back-up paths Onactivation the source node requests to build up a route to thenearest hop neighbor which is the path build-up phase [19] Inthis phase step one has source activation in which therequested node for path building adds its number and time-stamp and sends to the least hop-count neighboring node-isgoes on until the least time latency sink having the informationneeded to construct the primary route is reached [75] In steptwo there is path extraction when the new package fromanother path arrives whereby the extracted path is comparedwith the primary path If the node is shared the package will berejected or an alternative path will be searched for to get abackup through comparison of the preceding two paths Phasetwo is path acknowledgment where we have the third step inwhich the sink returns the ACK packet to the sender with pathinformation having nodes and their time information after ithas computed it using the timestamp [1819]

Advantages

(i) -e protocol minimizes end-to-end latency(ii) It enhances congestion control

414 Cross-Layer Optimal Design (CLOD) Authors in [76]come up with a CLOD for scheduling at the data link layerrouting at the network layer and controlling congestion atthe transport layer with an assumption of fixed link capacity-rough congestion control energy efficiency is improvedTransport layer congestion at the nodes is minimized bycompressed sensing (CS) in which transmitted bits are re-duced whereas optimally allocating resources decreasescongestion on the links at the data link layer CLODpromises minimized computational complexity and betterperformance in light traffic scenarios [77] Generating andstoring CLOD is more efficient compared to Gaussianrandom matrices It prolongs network lifetime and savesenergy

Advantages

(i) It does congestion control subsequently increasingenergy efficiency

28 Journal of Computer Networks and Communications

(ii) It reduces computational complexity(iii) It improves performance in light traffic scenarios

415 Challenges and Opportunities of the Cross-Layer Design(CLD)

(i) Different layers in a real network assume differentfunctions andor services A layer only communi-cates with its neighboring layers -e layered modeldegrades system performance because of many

features in wireless transmission QoS parameters inthe layers of WMSNs are shared by routing pro-tocols to optimize the performance Neverthelessthere is a need to design cross-layer models forincreased efficiency in routing [1]

(ii) -e physical layer plays a highly vital role in CLDRate adaptation and channel allocation take place atthe physical layer via signal processing to enhanceQoS End-to-end performance in wireless media isaffected by the changes therein given the effect on

Table 4 A summary of some of the reviewed cross-layer models

Protocol Layers Aims Comments QoS parameters

SDRCS MACPHY Routing real-time traffic Transmits multimedia traffic viacross-layer packet forwarding

-roughputlatency

SUIT Applicationtransport Transmission of non-real-time video Uses fuzzy logic to regulate

congestionTransmission rate

delay

CLASS Any twoDesign serves as a framework fordifferent implementations ofdifferent application scenarios

Scalable very efficient flexible andhas very low propagation delay

Propagation delayjitter

NLQS PHYMACnetwork

Permits packet-level servicedifferentiation as a function of

throughput packet error rate anddelay

Network layer QoS is enhanced -roughput delaypacket error rate

CMVT Applicationnetwork

Encodes video streams to video dataframes via MPEG-4 encoding anddoes route discovery and data

transfer

Greedy forwarding and rollback areused to find source-to-sink paths Energy

CoopMAC PHYMAC Offers spatial diversity among thethree nodes

Helper ID signal finds out howfeasible cooperative communication

is

-roughput ratedelay

M-CMAC PHYMAC Increases end-to-end throughputand packet delivery ratio

Euclidean distances between nodesare calculated and converted to the

data rate for a link

-roughputpacket delivery

ratio

CBCR PHYMACMinimizes control overhead andtime consumed in establishing thecooperative paths than M-CMAC

Energy consumption is moreuniformly distributed in a network

enhancing network lifetime

-roughputpacket deliveryratio energy

MAC-PHY MACPHYMeant to enable PHY layer

cooperation and maximize gains ofcooperation at the MAC layer

Leverages both spatial diversity andcoding gain -roughput delay

MAC-centric MACAPPL

Meant to support QoS needs in newvideo apps and enable MAC layer

differentiation for H264partitioning

Targets multimedia applications byusing the MPEG-4 scheme

Delay packet lossrate

ACFEC MACnetwork(UDP)

Meant to enhance the quality ofvideo streaming over 80211WLANs

and overcome packet losses

Adjusts redundancy rates toovercome channel fluctuations and

detect and reduce packet lossPacket loss rate

MHDMwTS mdash

Meant to provide reliable datatransfer with the multipath routingand load-balancing congestioncontrol method in WMSNs

More reliable than basic routingschemes for transport multimedia

data

Latency packagetransmit rate

BCFL mdash

Introduces dispersion into fuzzylogic-based routing and sets every

cross-layer parameter with adynamic weight

Can be used to select a CH in cluster-based routing protocols andproposes a dispersion formula

Node utilitydispersion

CLODDatalinknetworktransport

Prolongs network lifetime andachieves congestion control anddesigned for lightly loaded WSNs

Assumes fixed link capacity andintegrates compressed sensing

technology

-roughputaverage energy CS

error ratio

Journal of Computer Networks and Communications 29

the purposefulness of the protocols at the networklayer CLD offers solutions to conserving powerminimizing energy and controlling both the flowand congestion in the network as well as fault tol-erance so it is an opportunity for designers con-sidering other layers We further desire to developCLDs that shall make an impact on network oper-ation and get close attention [18]

(iii) CLD interface standardization the architecturemust deliver module functionality although thereexist some queries amongst modules about possibleinterfaces predicted by the necessity to share in-formation among remote protocol layers Technicalchallenges include developing designing andstandardizing of cross-layer interfaces and algo-rithms that satisfy the cross-layer optimization re-quirements among protocol layers

Table 4 summarizes these cross-layer models with a fewcomments

5 Future Research Direction

51 At theMACLayer In WMSNs achieving duty cycling istricky since video traffic is volatile Listening to adjacentnodes wastes energy We need to switch states from awake tosleep -is requires further research

Self-organizing networks and green communicationalthough most CRWSN energy-saving algorithms aredesigned to reduce node transmission power energyconsumption is inevitable during the operational mode ofBSs because of internal processing Scholars are trying withpolicy-makers to encourage a shift to greenWSNs to reduceoperational costs and carbon footprints With self-orga-nizing CRWSNs for automatic switch-off when idle energyis saved which calls for cognitive switching algorithms Sointegrating cognitive switching with radio resource allo-cation in CRWSNs is a promising research area [78]

52 At the Application Layer Considering a WMSN ar-chitecture end users access the network via the BS to andfrom which they send their sensed data We desire to getguaranteed QoS and energy efficiency by designing a newarchitecture in the form of hardware and software whichneeds to be investigated further for a fix of the same [1]

QoS and energy efficiency are very crucial in WMSNsand mostly real-time apps that require guaranteed band-width and throughput in their network lifetime Mostprotocols ignore base station (BS) mobility and WMSNnodes Traffic management telemedicine and battlefieldsurveillance apps require mobile nodes or BSs so designingdynamic routing protocols to be adaptable in these cir-cumstances is necessary

53 Cross-Layer Security of data transmitted on WMSNseg in military surveillance and e-commerce is key withQoS and energy efficiency WMSNs are vulnerable to attacks

like worms and sinkhole Moreover their computationpower is not that high complicating implementation ofstrong secure protocols -is needs further attention

For fairness and priority issues in CRWSNs nodes areassigned diverse priorities depending on importance andurgency of their data For better network performance thereis a need to develop fair resource allocation mechanismswith priority-based fairness in sensors-ere is still potentialin this research area [78]

We must develop multiobjective adaptive protocols tooptimize QoS metrics involved in routing for best trans-mission results to get a trade-off between different opti-mization metrics We need to design standardized cross-layer algorithms which are able to meet cross-layer opti-mization standards

6 Conclusion

We have reviewed the different QoS strategies for WSNs inthe context of IoT from the MAC layer and application layeras well as the cross-layer paradigm For the MAC layer wehave reviewed protocols for WSNs such as the hybrid of thecontention-free and contention-based MAC protocols andthose for WMSNs We further reviewed a number of ap-plication layer protocols including many machine-to-ma-chine requestresponse and publishsubscribe protocols Forsystem optimization cross-layer QoS strategies are veryimportant in wireless communication We have reviewed anumber of cross-layer strategies For all categories reviewedchallenges and opportunities are discussed Finally somepossible future directions are discussed for research andapplication showing a promising potential for future re-search in this relatively new area especially as more WMSNapplications emerge recently

Conflicts of Interest

-e authors hereby declare no conflicts of interest regardingthe publication of this work

Acknowledgments

-is research was funded by the National Natural ScienceFoundation of China (Nos 61602398 61672447 and61711540306)

References

[1] V Bhandary A Malik and S Kumar ldquoRouting in wirelessmultimedia sensor networks a survey of existing protocolsand open research issuesrdquo Journal of Engineering vol 2016Article ID 9608757 27 pages 2016

[2] I F Akyildiz T Melodia and K R Chowdhury ldquoWirelessmultimedia sensor networks applications and testbedsrdquoProceedings of the IEEE vol 96 no 10 pp 1588ndash1605 2008

[3] M Iqbal M Naeem A Ahmed M Awais A Anpalagan andA Ahmad ldquoSwarm intelligence based resource managementfor cooperative cognitive radio network in smart hospitalsrdquoWireless Personal Communications vol 98 no 1 pp 571ndash5922018

30 Journal of Computer Networks and Communications

[4] S A Alvi B Afzal G A Shah L Atzori and W MahmoodldquoInternet of multimedia things vision and challengesrdquo AdHoc Networks vol 33 pp 87ndash111 2015

[5] Q F Hassan A U R Khan and S A Madani Internet ofings Challenges Advances and Applications CRC PressTaylor amp Francis Group A Chapman amp Hall Group BocaRaton FL USA 2018

[6] Q Wang Y Zhao W Wang et al ldquoMultimedia IoT systemsand applicationsrdquo in Proceedings of the 2017 Global Internet ofings Summit (GIoTS) vol 2 Geneva Switzerland June 2017

[7] A Aslam and E Curry ldquoTowards a generalized approach fordeep neural network based event processing for the internet ofmultimedia thingsrdquo IEEE Access vol 6 pp 25573ndash25587 2018

[8] M Jridi T Chapel V Dorez L Bougeant and A Le BotlanldquoSoC-based edge computing gateway in the context of theinternet of multimedia things experimental platformrdquoJournal of Low Power Electronics and Applications vol 8no 1 p 1 2018

[9] H Noura A Chehab L Sleem andM N Rapha ldquoOne roundcipher algorithm for multimedia IoT devicesrdquo MultimediaTools and Applications vol 77 no 14 pp 18383ndash18413 2018

[10] T Balan D Robu and F Sandu ldquoMultihoming for mobileinternet of multimedia thingsrdquo Mobile Information Systemsvol 2017 Article ID 6965028 16 pages 2017

[11] H Shen and G Bai ldquoRouting in wireless multimedia sensornetworks a survey and challenges aheadrdquo Journal of Networkand Computer Applications vol 71 pp 30ndash49 2016

[12] A Ahmad M H Rehmani H Tembine O A Mohammedand A Jamalipour ldquoIEEE access special section Editorial optimization for emerging wireless Networks IoT 5G andsmart grid communication networksrdquo IEEE Access vol 5pp 2096ndash2100 2017

[13] B A Muzakkari M A Mohamed M F A Kadir andZ Mohamad ldquoRecent advances in energy efficient-QoS awareMAC protocols for wireless sensor networksrdquo InternationalJournal of Advanced Computer Research vol 8 no 38pp 212ndash228 2018

[14] M A Yigitel O D Incel and C Ersoy ldquoQoS-aware MACprotocols for wireless sensor networks a surveyrdquo ComputerNetworks vol 55 no 8 pp 1982ndash2004 2011

[15] A M Abbas and O Kure ldquoQuality of Service in mobile ad hocnetworks a surveyrdquo International Journal of Ad Hoc andUbiquitous Computing vol 6 no 2 p 75 2010

[16] M Q Shatnawi ldquoApplication layer protocols for the internetof things a surveyrdquo in Proceedings of the InternationalConference on Engineering amp MIS (ICEMIS) Agadir Mo-rocco September 2016

[17] V Karagiannis P Chatzimisios F Vazquez-gallego andJ Alonso-zarate ldquoA survey on application layer protocols forthe internet of things research motivationrdquo InternationalJournal of Advanced Research in Computer Science vol 8no 3 pp 1ndash10 2015

[18] M S Bernard T Pei Z Li and K Li ldquoQoS strategies forwireless multimedia sensor networks in the context of IoTrdquoLecture Notes of the Institute for Computer Sciences SocialInformatics and Telecommunications Engineering vol 4pp 228ndash253 2019

[19] A AlAmri and M Abdullah ldquoCross-layer quality of serviceprotocols for wireless multimedia sensor networksrdquo in Pro-ceedings of the International Conference on CommunicationManagement and Information Technology pp 649ndash658Cosenza Italy April 2017

[20] J Guo L Sun and R Wang ldquoA cross-layer and multipathbased video transmission scheme for wireless multimedia

sensor networksrdquo Journal of Networks vol 7 no 9pp 1334ndash1340 2012

[21] A Ksentini A Naimi and M Gueroui ldquoToward an im-provement of H264 video transmission over IEEE 80211ethrough a cross-layer architecturerdquo IEEE CommunicationsMagazine vol 44 no 1 pp 107ndash114 2006

[22] W Ye J Heidemann and D Estrin ldquoAn energy-efficientMAC protocol for wireless sensor networksrdquo in Proceedings ofthe 21st International Annual Joint Conference of the IEEEComputer and Communications Societies (INFOCOM 2002)New York NY USA June 2002

[23] J Polastre J Hill and D Culler ldquoVersatile low power mediaaccess for wireless sensor networks categories and subjectdescriptorsrdquo in Proceedings of the Second InternationalConference on Embedded Networked Sensor Systems Sen-Sysrsquo04 pp 95ndash107 Baltimore MA USA November 2004

[24] T Danmanee K N Nakorn and K Rojviboonchai ldquoCU-MAC a duty-cycle MAC protocol for internet of things inwireless sensor networksrdquo ECTI Transactions on ElectricalEngineering Electronics and Communications vol 16 no 2pp 30ndash43 2018

[25] T Van Dam and K Langendoen ldquoAn adaptive energy-efficientMAC protocol for wireless sensor networksrdquo in Proceedings ofthe 1st International Conference on Embedded NetworkedSensor Systems Los Angeles CA USA November 2003

[26] P Xie and J Cui ldquoR-MAC an energy-efficient MAC protocolfor underwater sensor networksrdquo in International Conferenceon Wireless Algorithms Systems and Applications (WASA2007) Chicago IL USA August 2007

[27] A Rahim N Javaid M Aslam U Qasim and Z A KhanldquoAdaptive-reliable medium access control protocol forwireless body area networksrdquo in Proceedings of the AnnualIEEE Communications Society Conference on Sensor Meshand Ad Hoc Communications and Networks (SECON) SeoulKorea June 2012

[28] H S Kwak S Ullah D H Kwak and C H Lee ldquoA powerefficient MAC protocol for wireless body area networksrdquoJournal of the Korean Institute of Intelligent Systems vol 8no 6 pp 131ndash140 2009

[29] C Omeni A J Burdett and M Park ldquoEnergy efficientmedium access protocol for wireless medical body area sensornetworksrdquo in Proceedings of the 4th IEEEEMBS InternationalSummer School and Symposium on Medical Devices andBiosensors pp 29ndash32 Cambridge UK August 2007

[30] L Sitanayah C J Sreenan and K N Brown ldquoER-MAC ahybrid MAC protocol for emergency response wireless sensornetworksrdquo in Proceedings of the Fourth International Con-ference on Sensor Technologies and Applications pp 244ndash249Mestre Italy July 2010

[31] Y Sun and D B Johnson ldquoRI-MAC a receiver-initiatedasynchronous duty cycle MAC protocol for dynamic trafficloads in wireless sensor networksrdquo in Proceedings of the 6thACM conference on Embedded network sensorsystemsmdashSenSysrsquo08 pp 1ndash14 Raleigh NC USA November2008

[32] M Buettner G V Yee E Anderson and R Han ldquoX-MAC ashort preamble MAC protocol for duty-cycled wireless sensornetworksrdquo in Proceedings of the 4th International Conferenceon Embedded Networked Sensor Systems SenSys 2006pp 307ndash320 Boulder Colorado USA October 2006

[33] S Boulfekhar and M Benmohammed ldquoSynchronous receiverinitiated MAC protocol for long-lived sensor networksrdquoComputers amp Electrical Engineering vol 40 no 2 pp 504ndash516 2014

Journal of Computer Networks and Communications 31

[34] S Mehta and K S Kwak ldquoH-MAC a hybrid Mac protocol forwireless sensor networksrdquo International Journal of ComputerNetworks amp Communications vol 2 no 2 pp 108ndash117 2010

[35] I Rhee A Warrier M Aia J Min and M L Sichitiu ldquoZ-MAC a hybrid MAC for wireless sensor networksrdquo IEEEACM Transactions on Networking vol 16 no 3 pp 511ndash5242008

[36] S Arshad A Al-sadi and A Barnawi ldquoZ-MAC performanceevaluation and enhancementsrdquo Procedia Computer Sciencevol 21 pp 485ndash490 2013

[37] Y Liu I Elhanany and H Qi ldquoAn energy-efficient QoS-aware media access control protocol for wireless sensornetworksrdquo in Proceedings of the IEEE International Confer-ence on Mobile ad hoc and Sensor Systems pp 9ndash11 Wash-ington DC USA December 2005

[38] J Decotignie ldquoWiseMAC an ultra low power MAC protocolfor the downlink of infrastructure wireless sensor networksrdquoin Proceedings of the Ninth International Symposium onComputers And Communications pp 244ndash251 AlexandriaEgypt June 2004

[39] R A Rashid ldquoDevelopment of energy aware TDMA-basedMAC protocol for wireless sensor network systemrdquo EuropeanJournal of Scientific Research vol 30 no 4 pp 571ndash578 2009

[40] H P Van Hoesel ldquoA lightweight medium access protocol(LMAC) for wireless sensor networksrdquo in Proceedings of theInternational Workshop on Networked Sensing Systems Bal-timore MD USA November 2004

[41] C Li P Wang H Chen andM Guizani ldquoA cluster based on-demand multi-channel MAC protocol for wireless multi-media sensor networksrdquo in Proceedings of the IEEE In-ternational Conference on Communications pp 2371ndash2376Beijing China May 2008

[42] J L Sobrinho and A S Krishnakumar ldquoQuality-of-service inad hoc carrier sense multiple access wireless networksrdquo IEEEJournal on Selected Areas in Communications vol 17 no 8pp 1353ndash1368 1999

[43] J P John ldquoProviding MAC QoS for multimedia traffic in80211e based multi-hop ad hoc wireless networksrdquo ComputerNetworks vol 51 no 1 pp 153ndash176 2007

[44] S Lofty and N Padmavati ldquoA survey on mobility basedprotocols in WSNsrdquo in Proceedings of the InternationalConference Communication and Manufacturing pp 12ndash15Hangzhou China July 2014

[45] S Deng L Shen and J Li ldquoMobility-based clustering pro-tocol for wireless sensor networks with mobile nodesrdquo IETWireless Sensor Systems vol 1 no 1 pp 39ndash47 2011

[46] M Khalil A Khalid F U Khan and A Shabbir ldquoA review ofrouting protocol selection for wireless sensor networks insmart citiesrdquo in Proceedings of the 24th Asia-Pacific Confer-ence on Communications (APCC) Ningbo China November2014

[47] O Message Q Telemetry and T Mqtt Committee Specifi-cation Draft 02 April 2014

[48] M Collina M Bartolucci A Vanelli-coralli andG E Corazza ldquoInternet of things application layer protocolanalysis over error and delay prone linksrdquo in Proceedings ofthe 7th Advanced Satellite Multimedia Systems Conference andthe 13th Signal Processing for Space CommunicationsWorkshop Livorno Italy September 2014

[49] A Talaminos-barroso M A Estudillo-valderrama L M RoaJ Reina-tosina and F Ortega-ruiz ldquoA machine-to-machineprotocol benchmark for e health applicationsmdashuse case re-spiratory rehabilitationrdquo Computer Methods and Programs inBiomedicine vol 129 pp 1ndash11 2016

[50] V Goyal O Pandey A Sahai and B Waters ldquoAttribute-based encryption for fine-grained access control of encrypteddatardquo in Proceedings of the 13th ACM Conference on Com-puter and Communications Security pp 89ndash98 AlexandriaVA USA October 2006

[51] M Singh M A Rajan V L Shivraj and P BalamuralidharldquoSecure MQTT for internet of things (IoT)rdquo in Proceedings ofthe Fifth International Conference on Communication Systemsand Network Technologies Gwalior India April 2015

[52] A Stanford-Clark and H L Truong MQTT for Sensor Net-works (MQTT-SN) Protocol Specification httpmqttorgnewwp-contentuploads200906MQTT-SN_spec_v12pdf 2013

[53] RFC ldquoExtensible messaging and presence protocol (XMPP)rdquoin Proceedings of the IETF pp 1ndash9 San Diego CA USAAugust 2005

[54] S Vinoski and I Technologies ldquoQueuing protocolrdquo IEEEInternet Computing vol 10 no 6 pp 87ndash89 2006

[55] J L Fernandes I C Lopes J J P C Rodrigues and S UllahldquoPerformance evaluation of restful web services and AMQPprotocolrdquo in Proceedings of the Fifth International Conferenceon Ubiquitous and Future Networks (ICUFN) pp 2ndash7 DaNang Vietnam July 2013

[56] M Asim ldquoA survey on application layer protocols for internetof things (IoT)rdquo International Journal of Advanced Researchin Computer Science vol 8 no 3 pp 996ndash1000 2017

[57] K An A Gokhale D Schmidt S Tambe P Pazandak andG Pardo-castellote ldquoContent-based filtering discovery pro-tocol (CFDP) scalable and efficient OMG DDS discoveryprotocolrdquo in Proceedings of the 8th ACM InternationalConference on Distributed Event-Based Systems New YorkNY USA July 2014

[58] E Web What is a web service pp 52ndash57 2010[59] M Laine RESTful Web Services for the Internet of ings

pp 2ndash4 2012 httpmediaTkk[60] IETF -e WebSocket protocol pp 1ndash71 2011[61] V Pimentel ldquoCommunicating and displaying real-time data

with websocketrdquo IEEE Internet Computing vol 16 no 4pp 45ndash53 2012

[62] Wikipedia 2019 httpsenwikipediaorgwikiStreaming_Text_Oriented_Messaging_Protocol2019

[63] K Raghavendra and S Nireshwalya ldquoApplication layer se-curity issues and its solutionsrdquo IJCSET vol 2 no 6pp 1266ndash1269 2012

[64] Y Xue B Ramamurthy andM C Vuran ldquoSDRCS a service-differentiated real-time communication scheme for eventsensing in wireless sensor networksrdquo Computer Networksvol 55 no 15 pp 3287ndash3302 2011

[65] C Sonmez S Isik M Y Donmez O D Incel and C ErsoyldquoSUIT a cross layer image transport protocol with fuzzy logicbased congestion control for wireless multimedia sensornetworksrdquo in Proceedings of the 5th International Conferenceon New Technologies Mobility and Security (NTMS) pp 1ndash6IEEE Istanbul Turkey May 2012

[66] T Melodia and I Akyildiz ldquoCross-layer QoS-aware com-munication for ultra wide band wireless multimedia sensornetworksrdquo IEEE Journal on Selected Areas in Communica-tions vol 28 no 5 pp 653ndash663 2010

[67] Q Wang and M A Abu-Rgheff ldquoCross-layer signalling fornext-generation wireless systemsrdquo in Proceedings of the 2003IEEE Wireless Communications and Networking 2003WCNC 2003 vol 2 pp 1084ndash1089 IEEE New Orleans LAUSA March 2003

[68] P Liu Z Tao S Narayanan T Korakis and S PanwarldquoCoopMAC a cooperative MAC for wireless LANsrdquo IEEE

32 Journal of Computer Networks and Communications

Journal on Selected Areas in Communications vol 25 no 2pp 340ndash354 2007

[69] H Shan H T Cheng and W Zhuang ldquoCross-layer co-operative MAC protocol in distributed wireless networksrdquoIEEE Transactions onWireless Communications vol 10 no 8pp 2603ndash2615 2011

[70] F Liu T Korakis Z Tao and S Panwar ldquoAMAC-PHY cross-layer protocol for wireless Ad-Hoc networksrdquo in Proceedingsof the 2008 IEEE Wireless Communications and NetworkingConference pp 1792ndash1797 IEEE Las Vegas NV USA March2008

[71] L Jacob and H R Shamna ldquoEfficient cooperative MAC androuting in wireless networksrdquo Transactions on Networks andCommunications vol 3 no 5 2015

[72] S Rao and K Shama ldquoCross layer protocols for multimediatransmission in wireless networksrdquo International Journal ofComputer Science amp Engineering Survey vol 3 no 3pp 15ndash28 2012

[73] L Han S Park S Kang and H Peter ldquoAn adaptive cross-layer FEC mechanism for video transmission over 80211WLANsrdquo in Proceedings of the 2009 International Conferenceon Internet pp 209ndash215 Las Vegas NV USA December2009

[74] N Li J Martınez and V H Dıaz ldquo-e balanced cross-layerdesign routing algorithm in wireless sensor networks usingfuzzy logicrdquo Sensors vol 15 no 8 pp 19541ndash19559 2015

[75] G Sun J Qi Z Zang and Q Xu ldquoA reliable multipathrouting algorithm with related congestion control scheme inwireless multimedia sensor networksrdquo in Proceedings of the2011 3rd International Conference on Computer Research andDevelopment vol 4 pp 229ndash233 IEEE Shanghai ChinaMarch 2011

[76] M Li Y Jing and C Li ldquoA robust and efficient cross-layeroptimal design in wireless sensor networksrdquoWireless PersonalCommunications vol 72 no 4 pp 1889ndash1902 2013

[77] J Yan M Zhou and Z Ding ldquoRecent advances in energy-efficient routing protocols for wireless sensor networks areviewrdquo IEEE Access vol 4 pp 5673ndash5686 2016

[78] A Ahmad S Ahmad M H Rehmani and N U Hassan ldquoAsurvey on radio resource allocation in cognitive radio sensornetworksrdquo IEEE Communications Surveys amp Tutorials vol 17no 2 pp 888ndash917 2015

Journal of Computer Networks and Communications 33

International Journal of

AerospaceEngineeringHindawiwwwhindawicom Volume 2018

RoboticsJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Active and Passive Electronic Components

VLSI Design

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Shock and Vibration

Hindawiwwwhindawicom Volume 2018

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawiwwwhindawicom

Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Control Scienceand Engineering

Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Journal ofEngineeringVolume 2018

SensorsJournal of

Hindawiwwwhindawicom Volume 2018

International Journal of

RotatingMachinery

Hindawiwwwhindawicom Volume 2018

Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Navigation and Observation

International Journal of

Hindawi

wwwhindawicom Volume 2018

Advances in

Multimedia

Submit your manuscripts atwwwhindawicom

Page 4: ReviewArticle - downloads.hindawi.com · ReviewArticle QoSStrategiesforWirelessMultimediaSensorNetworksinthe ContextofIoTattheMACLayer,ApplicationLayer,and Cross-LayerAlgorithms MuwongeSsajjabbiBernard

and swift response to changes in the physical world arechallenging IoT requirements [12]

14 What Has Been Done In wireless communicationsystems routing is one issue that is quite a challenge yetthere are very few surveys available especially regardingWMSN protocols [1] and more so the MAC layer appli-cation layer and cross-layer protocols although some sur-veys have been published on WSNs To the best of ourknowledge no survey combines all these aspects -is makesour survey important to add to the existing literature -eauthors in [1] discuss a number of WMSN routing tech-niques and the properties and shortcomings of theseHowever they do not discuss any application layer tech-niques Muzakkari et al [13] survey some recent WSNcontention-based scheduling-based and hybrid MACprotocols whereby they focus on the underlying principlesadvantages limitations and their applications But much asthey only focus on MAC protocols they still leave out anumber of them in their survey Yigitel et al [14] carry out asurvey on QoS-aware MAC protocols for WSNs -ey re-view the QoS challenges and views for WSNs study the QoSmechanisms categorize the state-of-the-art QoS-awareMAC protocols and also talk about the advantages anddrawbacks of the same But being a 2011 release a number ofkey protocols developed since then to date are not presented

therein Abbas and Kure [15] review various methods forQoS provisioning at the levels of routing MAC layer andcross-layer including the schemes for admission control andscheduling for QoS provisioning as well as the problems andchallenges involved But they leave out the protocols from2010 to date Shatnawi [16] and Karagiannis et al [17] reviewthe application layer protocols for the Internet of things Intheir survey they address a number of application layerprotocols which are employed for IoT for affirming a re-liable tie among objects and things -ey evaluate thereviewed protocols in terms of architecture communicationmodel security and QoS as well as the weaknesses andstrengths per reviewed protocol However they only con-centrate on application layer protocols and still leave outsome critical protocols In a recent survey Bernard et al [18]discuss a number of cross-layer QoS strategies for IoT Butbecause of space constraint we leave out some criticalprotocols and besides this paper does not talk about ap-plication layer and MAC layer protocols AlAmri andAbdullah [19] carry out a survey on cross-layer QoS pro-tocols for WMSNs in which they state that the cross-layerarchitecture is a novel idea that brings together a number oflayers for enabling integration and exchange of informationin between them with higher efficiency compared to thetraditional layered model -ey discuss and compare theavailable cross-layer WMSN protocols that cross the uses ofadjacent or nonadjacent layers But they concentrate on

Database

MMS user databases

MMS relayserver

Home location register

External servers(eg e-mail and fax)

MMS value-added servicesservers (premium content

e-mails web )

Foreign MMSC(MMS relay server)

MM

S us

er ag

ents

Internet

Figure 2 Example of the multimedia service architecture in the context of IoT

4 Journal of Computer Networks and Communications

cross-layer protocols and leave out the MAC layer andapplication layer protocols and also some cross-layer pro-tocols are left out -e surveys that have been reviewed arecompared in Table 1

-emajor contributions that make our survey importantare outlined below

15 Importance of the Survey

(i) We discuss and draw a comparison between the IoTand the IoMT by discussing the differences betweenthe two

(ii) We further discuss the newer paradigm of WMSNsand compareWSNs andWMSNs in our discussion

(iii) We make a comprehensive survey of the recentstate-of-the-art routing protocols focusing on theMAC layer the application layer and the cross-layer paradigm

(iv) We discuss the WMSN protocols as well as theWMSN-enabled protocols that have been de-veloped in the recent studies by several scholars

(v) In some of the reviewed protocols we talk about therelative advantages and disadvantages of some ofthe routing approaches for readers to be in positionto comprehend different techniques and so choosethe most appropriate technique depending on theuser requirements

(vi) In all these we cover the challenges and oppor-tunities existing for the discussed categories

(vii) We consider the possible future research trendsbefore concluding this paper

16 Challenges

(i) Since WSNs and WMSNs are usually distributed andad hoc networks their nodes are powered by batteriesmaking energy optimization a challenge that needs tobe addressed when designing their routing protocols

(ii) Considering WSNs their routing protocols do notput into consideration multimedia applicationswhich need a lot of bandwidth and processingenergy and must be transmitted in real time withutmost fidelity Extra challenges in routing areimposed in WMSNs in deployment of heteroge-neous sensors since there could be audio video andstill pictures in addition to scalar data duringtransmission Besides the heterogeneous datathereby handled all have different QoS re-quirements and with many different businessneeds different services with varying requirementsgive rise to a big challenge for the routing design

(iii) WMSNs have also got nonrechargeable batteriesplacing a limitation on the energy of the nodes yetmultimedia applications consume a lot of energyQuite a number of complications and limitingfactors exist practically despite the fact that sensorscan be self-powering as enabled by energy

harvesting Predictive video encoding using MPEG-4 and other standards is also an energy-consumingprocess and so causes enormous degradation inrouting efficiency complicating the design ofWMSN protocols even more

(iv) -e heterogeneous nature of the network is anotherchallenge since there is a difference in the categoriesof nodes needed for communication so as to enablethe facilitation of effective data collection andprocessing as well as efficient transmission It is noteasy to have a uniform communication protocolplatform because of different functionalities unlikethe case of the conventional WSNs compared to theWMSNs

(v) -ere must be a trade-off between the energy effi-ciency and the multimedia QoS when deciding onthe route taken Data aggregation and compressionrouting protocols may be applicable for energy-saving though they can lead to intolerable delays inWMSNs Furthermore there might be networkcongestion due to the many-to-many and mutualinterference in wireless routes and the scarceWMSN resources In case a given node gets so manyhigh-rate streams it lowers the performance of thenetwork and raises the possibility of node failuredue to energy drainage

-is papermainly discusses QoS strategies forWSNs andWMSNs in the context of IoT from the MAC layer appli-cation layer and cross-layer perspective -is paper presentsa state-of-the-art survey on routing in WSNs and WMSNs-e rest of this paper is organized as follows In Section 2 wediscuss the QoS strategies at the MAC layer In Section 3 wediscuss the QoS strategies at the application layer Section 4discusses the cross-layer QoS strategies and in Section 5 wegive some future research directions before concluding thispaper in Section 6

2 QoS Strategies at the MAC Layer

-e medium access control (MAC) protocol is meant toregulate admittance to the shared medium as well astransmission reliability and efficiency through error cor-rection in wireless transmissions among others It is alsoresponsible for performance of framing addressing andflow control -e MAC protocol design affects energy effi-ciency too -is is majorly categorized into contention-freeprotocols contention-based protocols and a hybrid of these

21 MAC Protocols forWSNs Hybrid MAC protocols thesecombine contention-free and contention-based MAC pro-tocols on whose advantages they capitalize Some of these arediscussed below

211 S-MAC (Sensor MAC) -is is a MAC protocoldesigned for WSNs -e main components of S-MAC in-clude (i) periodic listen and sleep (ii) collision and over-hearing avoidance and (iii) message passing Energy

Journal of Computer Networks and Communications 5

consumption and self-configuration are controlled usingthese three techniques Nodes periodically go into the sleepmode to control the energy consumed in listening to idlechannels and virtual clusters are formed by neighbor nodesfor sleep schedule autosynchronization -e protocol em-ploys in-channel signaling when the radio is in the sleepmode and message passing is applied to decrease contentionlatency for store-and-forwardWSN applications during datatransmission [22] For adjacent nodes to solve their clockdrift S-MAC requires periodic synchronization To reducesynchronization errors exchanged timestamps are relativenot absolute

Advantages

(i) S-MAC has a longer listening period than the clockerror

(ii) It has more capacity to conserve energy than IEEE80211

(iii) It can do trade-off between energy and latencydepending on traffic conditions [22]

212 B-MAC (Berkeley MAC) -is is a carrier-sense MACprotocol designed for WSNs featuring a simplistic scalableimplementation that does network variations and is effectivein channel valuation [23] -e protocol has an interface withsome flexibility leading to very low-power usage and con-trols collisions through collision avoidance mechanismsleading to effective use of the frequency band It uses asampling scheme that is adjustable to minimize duty cyclingand idle listening so as to reduce power wastage Nodes arescheduled with their duty cycle each within which they sendthe preamble to the channel in case of availability of data fortransmission In different duty-cycle schedules channelusage is checked by the destination nodes that is if thepreamble found is long it will remain powered on until thedestination address is obtained and in case it is the target itwill either await senderrsquos data or go to sleep-e protocol haslower latency than synchronous MAC protocols but has aproblem of overhearing where the neighbor node getslengthy preambles though not the intended receiver hencewasting a lot of power in doing so [24] It outperforms anumber of protocols via reconfiguration feedback andbidirectional interfaces for upper layer services It can run at

low duty cycles and its apps cannot experience synchro-nization overhead [23]

Advantages

(i) -e protocol features a simplistic scalableimplementation that does network variations andis effective in channel valuation

(ii) It has an interface with flexibility leading to verylow-power usage

(iii) It controls collisions through collision avoidancemechanisms leading to effective use of the fre-quency band

(iv) It uses an adjustable sampling scheme to minimizeduty cycling and idle listening reducing powerwastage

Disadvantages

(i) It has a problem of overhearing where the neighbornode gets lengthy preambles though not theintended receiver hence wasting a lot of power indoing so

213 T-MAC -is is a contention-based MAC protocol forWSNs that dynamically with little difficulty via fine-grainedtimeouts adapts a listensleep duty cycle It uses this dutycycle by ending its active part so as to vary load in time andposition thereby decreasing the energy wastage on idlelistening [25] When starting an active period it uses a shortlistening window thereby improving the use of energy byS-MAC -rough adaptive duty cycling T-MAC savespower but at a cost of less throughput and enhanced delayBoth S-MAC and T-MAC under similar conditions performin the same way though in variable workloads S-MAC uses 5times more power than T-MAC But T-MAC is complex andnot scalable If its active window is decreased its snoopingcapability on nearby traffic and adaptation to variableconditions of the network is reduced [23] All nodes in thisprotocol are configured to wake up periodically listen toadjacent nodes and go back into the sleep mode till the nextframe as more messages come into the queue Collisionavoidance and reliable communication are ensured throughRequest-to-Send (RTS) Clear-to-Send (CTS) and Ac-knowledgment (ACK) messages among nodes [25]

Table 1 Comparison of this survey with existing surveys

Author(s) Year CharacteristicsBhandary et al [1] 2016 Routing strategies in WMSNs properties and limitationsMuzakkari et al [13] 2018 WSN contention-based scheduling-based and hybridMAC protocolsAhmad et al [12] 2011 QoS-aware MAC protocols for WSNsMuzakkari et al [13] 2010 QoS provisioning at the levels of routing MAC layer and cross-layerShatnawi [14] 2016 IoT application layer protocolsKaragiannis et al [15] 2015 Application layer protocols for the Internet of thingsBernard et al [16] 2019 Cross-layer QoS strategies for the IoT in WMSNsAlAmri and Abdullah [17] 2017 Cross-layer QoS protocols for WMSNsGuo et al [20] 2012 Cross-layer and multipath-based video transmission for WMSNs

Ksentini et al [21] 2006 Toward an improvement of H264 video transmission over IEEE80211e through a cross-layer architecture for WMSNs

6 Journal of Computer Networks and Communications

T-MAC design the major forms of energy consump-tion in T-MAC are through (i) idle listening (ii) col-lisions (iii) protocol overhead and (iv) overhearing Incontrast in the active period the node remains lis-tening and sending data If no activation event occursfor a certain time t then the active period ends -eactivation event might be data reception on a radiocommunication sensing on a radio and end oftransmission for the data packet of the node or ACKknowing that the neighbor has finished transmitting

Clustering and synchronization on waking up a nodebegins by listening to the channel If after a certain timethere is nothing heard it chooses a schedule fortransmitting Synchronization (SYNC) packetsappended with the time of the next frame startup Ifduring this time it hears another nodersquos SYNC packetit follows its schedule subsequently transmitting itsown SYNC Following both schedules enables it to havean activation incident at the beginning of both frames

RTS and t choice if a node does not get a CTS messageafter sending an RTS then either (i) there is a collisionpreventing the destination node from hearing the RTS(ii) RTS or CTS is overheard and destination node barredfrom responding or (iii) the destination node is in thesleep mode So if after time t there is no answer thesource node switches to the sleepmode It should only doso if none of its neighbors is still communicating A newinterval t can be triggered by a neighborrsquos RTS or CTSreception A node may be out of range to hear theinitiating RTS-e length t of the interval should be longenough to get the CTS packetrsquos beginning [25] and t isthus given by the comparison in the following equation

tgtC + R + T (1)

where C contention interval length RRTS packet lengthand T turn-around time

Advantages

(i) It saves power through adaptive duty cycling

Disadvantages

(i) It is unscalable and more complex(ii) It offers less throughput and enhanced delay

214 CU-MAC -is protocol supports IoT standards re-quiring request-response transmissions It employs a multi-channel mechanism for constant bidirectional transmissionof packets at low duty cycle to resolve hidden-terminal issuesAccording to Danmanee et al [24] CU-MAC is an SI-MACprotocol that is aimed at improving bandwidth of the channelby initiating connections and transmitting data throughdifferent channels using the multichannel approach

Protocol design the protocol has the following threeproperties that make it perform more efficiently withWSNs in the IoT

(i) It enables multichannel transmission using 9useable channels in IEEE 802154

(ii) It can enable linked data transmission using oneconnection which reduces additional overheadand delay

(iii) It can support continuous transmission of packets-is is enabled depending on how the bufferoverflow protection policy is set in the buffer stack-is stack has 2 parts (a) bidirectional buffer slotsfor enabling sensor nodes to support linked datatransmission and (b) normal buffer slots fornormal data queuing Buffer stack policies are setas follows

(a) A packet is only dropped in the buffer when itexpires

(b) Unless the packet is from a target node to which areceiver is sending the data the receiver deniespacket reception when buffer slots fill up

(c) -e bidirectional buffer slot is used for bi-directional buffer transfer

CU-MAC can reduce power wastage by choosing toemploy only a single channel for transmitting both controlpackets and data in case there is only one sender in the sameregion of coverage It can further enable multiple trans-missions through another usable channel if there are othersending nodes in this area [24] -e protocol has two phases(i) advertising phase in which a source node announces toneighbors before transmission and (ii) data transfer phasewhich commences when the source node receives a Ready-to-Receive Information packet from the destination node

Advantages

(i) It has a higher packet delivery ratio compared toother MAC protocols

(ii) It performs well in high traffic situations since itsduty-cycle is lower

(iii) It reduces power wastage by employing only asingle channel for transmitting both controlpackets and data

(iv) It can enable multiple transmissions through an-other usable channel if there are other sendingnodes

215 R-MAC -is is a reservation-based MAC protocoldesigned majorly with an aim to be fair and efficient in termsof energy use To eliminate collision of data packets R-MAC[26] schedules control and data packet transmissions at thesource and destination nodes rather than using RTSCTSmessage exchanges Channel utilization in R-MAC is en-hanced by an ARQ method called burst-based acknowl-edgment which together with other scheduling algorithmsaddresses the exposed terminal problem besides enhancingthe network throughput

R-MAC design all nodes in this protocol intermittentlyfunction in listen and sleep states so as to avoid wastingenergy in overhearing and idle modes -e time takento listensleep is the same for all nodes each of which

Journal of Computer Networks and Communications 7

chooses a schedule for itself rendering central sched-uling and synchronization irrelevant So in the absenceof traffic in the locality the node simply does periodiclistening and sleeping and to eliminate collisionstransmissions are distributively synchronized using areservation-based model in case there is a sender totransmit data R-MAC differs from S-MAC in that ittransmits data to a number of nodes using the sleepmode [24] -ere are 3 stages in this protocol (i) la-tency detection (ii) period announcement and (iii)periodic operation of which (i) and (ii) are for nodesynchronization and (iii) is for listensleep procedures

(i) Latency detection phase here all nodes broadcast aNeighbor Discovery control packet (ND) at an ar-bitrarily chosen time since they are powered onWhen the ND is received from the neighbor nodethe receiver notes its arrival time and chooses arandom time to send an acknowledgment (ACK-ND)mdashthe same as the receivedND in sizemdashinwhichthe duration from ND arrival to ACK-ND trans-mission I2 is specified and the time from ND packettransmission to ACK-ND arrival I1 is computed

(a) Propagation latency between the 2 nodesL (I1 minus I2)2

(b) -us propagation latency can be defined as ldquotheinterval from the time the first node sends the firstbit of a packet to the time the second node receivesthe last bit of the packetrdquo [24]

(ii) Period announcement phase here nodes choosetheir own random listensleep time and start timeand broadcast these and when packets are re-ceived the node turns the received schedule intoits own All nodes will record their neighborsrsquoschedules in relation to theirs -ese two phasesmake a number of rounds to ensure all nodes areinformed about their neighboring nodes

(iii) Periodic operation phase here there is datatransmission [26] and nodes intermittently go intothe sleep mode and wake up One period is a listensleep cycle and nodes possess similar periods Twoparts make a period T0 listen window TL and sleepwindow TS as shown in the following equation

T0 TL + TS (2)

In this protocol the nodes make use of control packetsfor communication-ese include REV (reservation packet)ACKREV (acknowledgment packet for REV) and DATAACK-DATA (acknowledgment packet for data) messageexchanges all of which are similar in size

Advantages

(i) It is fair and efficient in terms of energy use(ii) It eliminates collision of data packets

216 AR-MAC (Adaptive-Reliable Media Access Control)-is is a TDMA-based MAC protocol for wireless body areanetworks (WBANs) designed to reduce energy consumption

as proposed by Rahim et al [27] -e protocol allocates aguaranteed time slot (GTS) to all nodes for transmissiondepending on their needs It utilizes periodic sleep andwakeup to minimize overhearing and idle listening based onthe needs of the node It has a central node (CN) having bigbatteries with higher computational power and single ordouble transceivers For double transceivers the sum of thetime frame TFrame is assigned for node transmission But inAR-MAC a single transceiver CN is considered with TFramehaving 3 subdivisions contention-free period (CFP) forsensor connection contention access period (CAP) foremergencies and time TMS to communicate with themonitoring station (MS)

Channel selection the CN first scans for redundantchannels from which it chooses one for transmissionand broadcasts its address and other information to thenodes -e destination nodes search for channels byscanning the radio frequency (RF) channels If free itswitches to the next and if busy it will wait for time TCPlistening to packets and again switch to the subsequentchannel if it does not get the channel packets When thenode finally succeeds in getting a channel packettransmission commences and an ACK packet is sent tothe CNTime slot assignment a Time Slot Request (TSR) packetis sent to the CN by the sensor node after it has selectedthe RF channel -e TSR packet has data rate and timeslot info for the node Static-size time slots with staticguard-band time TGB are proposed in [28] -isprotocol employs the adaptive algorithm of time slot(TS) and GB time and the CN allocates TS whilesending Time Slot Request Reply (TSRR) based on thenode traffic pattern -e size of the time slots variesdepending on the node requirements Based on thetransmissionmodel the data packet transmission ACKpacket reception and some amount of delay can betolerated by allocated TS To prevent interference be-tween adjacent time slots resulting from the node andCN clock drift TGB is inserted between them and itsvalue is given as follows

TGB1

FxTS1100

TGBn

FxTSn

100

TGBnn+1

F

100X12

TSn + TSn+11113858 1113859

(3)

where FGB factor depending on the mean driftvalue Nodes go to sleep after slots have been suc-cessfully assigned and wake up to transmit in assignedslots thus saving energy wasted in idle listening sinceeven allocated slots are collision freeSynchronization for periodic synchronization TDMAneeds additional energy [29] It needs a lot of energy tosynchronize nodes after several cycles To reduce en-ergy wastage and control collisions AR-MAC employs

8 Journal of Computer Networks and Communications

a new synchronization scheme When a data packetarrives the current and expected packet arrival timeswith allowable delay (D) are compared by the CN and adrift value (DV) is computed based on the differenceΔT -e DV is sent to the node in the ACK packet forchanging the slot in subsequent transmissions thoughthe value is based on allowable delay and F In caseΔTgtD the CN will send the DV in the SYNC-ACKpacket for subsequent SYNC to nodes or else a mereACK packet is sent by the CN for the received packetFor any forthcoming data transmissions the node willchange its wake-up time plan based on the DV Withsuch a sync approach a sleeping node will not lose syncfor all cycles N Acceptable delay (D) is given as

D min TS1 TSn( 1113857XF

100 (4)

where F is the guard-band factor

ΔT expected arrival time minus current arrival time

DV 0 if |ΔT|ltD

ΔT if |ΔT|gtD1113896

(5)

Frame format the AR-MAC protocol employs twoforms of packets namely (i) data packets in which anode transmits in its assigned time slot and employsCAP for emergency data and (ii) control packets thatinclude channel packet Time Slot Request (TSR)packet Time Slot Request Reply (TSRR) packet Syn-chronization-Acknowledgment (SYNC-ACK) packetData Request (DR) packet and Acknowledgment(ACK) packet [27]Energy consumption if N is the number of cyclesenergy consumption for these is measured as follows

ETotal 1113944N

k1ESleepk

+ 1113944N

k1EActivek

ESleep TSleep x ISleep x V

TSleep Tframe minus Tactive

(6)

where ISleep current from the voltage source V whenthe node is asleep If Esw is the switching energy Etransthe transmission energy Erec the receiving energy andETout the time-out energy then

EActive 2Esw + Etrans + Erec + ETout (7)

Advantages

(i) It reduces energy consumption(ii) It minimizes overhearing and idle listening based

on the needs of the node via periodic sleep andwakeup

217 ER-MAC It is a combination of TDMA and CSMAmethods making it flexible and adaptable to topological andtraffic changes and hence applicable in rapid response

situations inWSNs Collision-free slots can be planned usingTDMA [30] ER-MAC is designed in such a way that thefollowing are met

(a) It has a higher ratio of packet delivery and low delayunder lower energy consumption levels out-performing Z-MAC

(b) It keeps two priority queues in order to distinguishhigh from least priority packets

(c) It permits contention in TDMA slots so as to handleenormous traffic volumes

(d) It has a harmonized slot shape and the nodes areable to locally change their plans enabling themconnect to or exit the network with easeProtocol design it discovers the topology in such away that it uses the simple flooding mechanism tobuild the tree as initiated by the base station (BS) -etopology discovery aims at neighbor discovery andchange tracking in addition to setting up a routingtree A TOPOLOGY_DISCOVERY message com-prising hop_count new_parent_id and old_-parent_id is generated by the BS A node broadcasts itto discover if it has any potential children and aresponse to the parent In case of the desire to changethe parent the former parent is notified For all nodesthe hop count to the BS parent ID children and one-hop neighbor are registered with the BS at this stageTDMA slot assignment when a BS sends a SYNCmessage it switches to the TDMAmode which whenreceived by the child uses the broadcast slot for itschildrenrsquos synchronization Neighboring nodescannot share a slot since slots are assigned andschedules are exchanged accordingly Slot assign-ment begins with a leaf node in a bottom-up modesuch that the flow of messages to the BS may besmooth Before getting the child-node schedules anonleaf node will not transmit its data via a unicastslot descendantrsquos data via many unicast slots andchild synchronization via a broadcast slot Childnodes send SCHEDULE_NOTIFICATION mes-sages to the BS to end this phase

Local time synchronization is done via parent-childbroadcast sync since they both need to operate the same clockso that when one is transmitting the other is ready to receivePacket prioritization happens in such a way that a highpriority queue is first emptied before a lower priority queue isallowed to transmit In case a queue is filled up the shortestslack packet is dropped and then fairness over source isconsidered as queue modification is done to enable the BSbalance info from different nodes Energy conservation is bysender turning off the radio in absence of any info to transmitand the receiver will switch to the sleep mode in case nopackets are received after a certain timeout [30]

Advantages

(i) It is flexible and adaptable to topological and trafficchanges

Journal of Computer Networks and Communications 9

(ii) It is applicable in rapid response situations inWSNs

(iii) It has a higher ratio of packet delivery and lowdelay under lower energy consumption levels

218 RI-MAC (Receiver-Initiated MAC) -is is an asyn-chronous duty-cycle protocol that tries to reduce timeoccupied by the communicating nodes on the medium toagree on time for data exchange-e sending node waits forthe receiving node to signify the beginning of the com-munication session through a short beacon frame that itsends -e channel is only occupied by the beacon and datatransmission reducing channel occupancy and thus en-abling data exchange for other nodes too -is implies animprovement in channel utilization that subsequently leadsto increment in throughput packet delivery ratio andpower efficiency over a wide-ranging traffic load [31] Beinga receiver-initiated model it ensures substantial reductionin overhearing and lowers the probability of colliding andthe cost of recovery compared to B-MAC [23] or X-MAC[32] RI-MAC is further capable of using a beacon withACK and Ready-to-Receive packets though it has aproblem if a sending node with Ready packets needs to havethe radio on till a ready-to-receive node is awake [13] whichthen sends it a beacon and transmission begins immediatelyand later is acknowledged by another beacon Nodes in-termittently wake up to listen to any possible frames meantfor them depending on their schedules A node announcesits readiness to receive a frame through a broadcast beaconon waking up when it switches on the radio When a dataframe has been received the receiver will spend some moretime known as ldquodwell timerdquo in the active state so as toreceive queued packets which is the time in RI-MAC thatdepends on the number of contending senders If there areseveral contending senders that might not have beenpredicted there is a challenge in trying to reduce the re-ceiverrsquos active time to optimize power efficiency and reducethe cost of collision detection and lost data recovery even inthe case of hidden senders For this protocol this can beaddressed by a beacon frame from the receiving node forcoordinating data frame transmissions of contendingsending nodes-e protocol also beats B-MAC and X-MACin minimizing the cost of collision detection and datarecovery through coordination of the receiversrsquo dataframes -is is because the receiving node is aware of themaximum delay prior to the arrival of the frame-e abilityto broadcast is supported by RI-MAC when it sends data asunicast to each senderrsquos neighboring node or continuouslytransmits the frame in a back-to-back mode for as long asthe sleep interval [31]

Advantages

(i) It improves channel utilization leading to in-crement in throughput packet delivery ratio andpower efficiency over a wide-ranging traffic load

(ii) It ensures substantial reduction in overhearing andlowers the probability of colliding and the cost ofrecovery

219 SRI-MAC (Synchronous Receiver-Initiated MAC)-is is a synchronous duty-cycle protocol that implementsthe method of receiver-initiated data transmission with anaim of enhancing network lifetime through energy wastageavoidance via collision overhearing and idle listening To dothis and also minimize duty cycle it uses a sequence ofadaptive beacons and RTSCTS packets In this protocolsending nodes do not go to sleep till they have got CTSpackets from receiving nodes to begin transmitting exceptfor nonsending nodes that return into the sleep mode rightaway as if the channel was idle [33] Since sending nodeshave separate schedules it reduces the contention proba-bility leading to more energy conservation A time frame isdivided into the information period the allocation periodand the communication period which are discussed asfollows

Information period at frame initiation a beacon is sentby the receiving node declaring its return from sleepand readiness to accept data packets in case the channelis not busy -e beacon is a small packet alerting otherlistening nodes of the transmitterrsquos readiness to get arequest for channel It consists of the receiverrsquos ID andthe duration allocation period (DAP) that relies onnumber of the receiving nodersquos neighbors [13]Allocation period here the receiving node stays awake forsome little time after broadcasting its beacon to confirm ifthere is any node having anRTS packet to transmitWhenthe beacon has been received an RTS packet (with 3fieldsmdashsender ID receiver ID and data size) is sent by thesender prompting the receiver to send a CTS for SYNC asthere is neighbor information exchangeCommunication period SRI-MAC is capable of or-ganizing and scheduling transmissions by means ofsensor node IDs -e receiver assigns order andtransmitting time for every sender such that they switchoff their radios till their time to transmit is dueHowever the receiving node remains in the wake-upmode to get all sending nodersquos data SRI-MAC canenable sensors have lengthy sleeping times duringtransmission such that energy can be saved as long as itis neither sending nor receiving data -e protocolrsquoscommunication period is premised on the TDMAtechnique which conserves energy since the radiorsquosduty cycle is lowered and collision is eliminated [33]

Advantages

(i) Since it employs the TDMA principle it eliminatescollisions while maintaining energy conservationtoo

(ii) Noncommunicating nodes switch off their radioand get into the sleep mode which eliminatesoverhearing

(iii) Since receiving nodes broadcast SYNC signals toall nodes in the form of CTS packets overemittingis avoided

(iv) It decreases duty cycling as sleep time increasesand lowers idle listening

10 Journal of Computer Networks and Communications

2110 H-MAC -is protocol is built on the power savingmechanism (PSM) of IEEE 80211 and slotted Aloha pro-tocols to enhance performance using multiple slots Timehere comprises large frames each having an active and asleeping part [34] If any node has got data to send it will askfor slots from the receiver node during active time and latertransmit the packets in the prenegotiated slots Nodes gointo the sleep mode during sleep-time slots if they do nothave data to transmit

Latency in IEEE 80211 if N is the number of hops nthe value for the current hop tcsn the back-off time andttx the transmission delay the total latency is calculatedusing the following equation

D(N) 1113944N

n1tcsn + ttx1113872 1113873 (8)

and the average latency is given by the followingequation

E[D(N)] N tcsn + ttx1113872 1113873 (9)

-roughput in H-MAC is given by the followingequation

Th npnm

tact + Cts (10)

where np is the number of transmitted packets nm is themaximum number of nodes connected in a time frameC is the number of slots with the same length tact is thelistening time and ts is the sleep delayAdvantages

(i) It enhances performance using multiple slots

2111 Z-MAC (ZebraMAC) -is rides on the strengths andoffsets weaknesses of TDMA and CSMA It yields very goodchannel usage and low delay under low contention and lowerscollision among two-hop neighbors at a reduced cost Itsperformance is strong via error synchronization failure inassigning slots and channel conditions that change with timeIn the set-up phase the protocol operates through certainsteps including neighbor discovery slot assignment localframe exchange and global time synchronization whichunless the topology of the network changes significantly donot run again When the data are being transmitted there isimprovement in network throughput and energy efficiencywhich covers up for the initial upfront operational costs [35]

Disadvantages

(i) Initial slot assignment is quite a cumbersome taskhere since it is done at the beginning yet this maynot be at the same time in reality It further as-sumes static links until setup happens again whichtoo is impractical

(ii) It also suffers from the hidden-terminal problemsince slot owners have a small contention windowand priority for slots allocated to them due to theECN messages

(iii) Synchronizing time also poses a challenge in thisprotocol due to a large clock drift between nodes

(iv) Source and destination nodes too are not wellcoordinated in Z-MAC [36]

2112 Q-MAC (QoS-Aware Media Access Control) -isprotocol offers high QoS with improved energy efficiencyand accesses the wireless channel using MACAW as theunderlying protocol [37] Traffic from different nodes isprioritized for QoS satisfaction depending on how criticalthe data are It mainly has two scheduling algorithms inWSNs (i) intranode and (ii) internode scheduling -eintranode scheduling scheme adopts a first-in first-out-(FIFO-) based queuing algorithm based on application andMAC layer abstraction for data packet classification In-ternode scheduling reduces idle listening and collision on thechannel for improved energy consumption

Advantages

(i) It gives a high QoS with improved energy efficiency

2113 WiseMAC (Wireless Sensor MAC) WiseMAC isbased on the Aloha protocol and employs preamble sam-pling in achieving low-power transmissions in infrastructuresensor networks by occasionally sampling the channel tocheck for activity [3238] -e protocol employs the samemethod as B-MAC except that the sender shortens theextended preamblersquos length by scheduling its transmissionsaccordingly after learning the receiverrsquos awake schedules-is is done by the receiver node putting its subsequentawake time on the ACK packet such that when the senderneeds to send to it again it starts the preamble shortly beforeit wakes up so as to avoid energy wastage in transmitting thepreamble [32] -e data frame is repeated instead of theextended preamble for less traffic transmissions that have alengthier preamble than the data frame in WiseMAC Afterprocessing the frame a node that is not the target recipientwill go back into the sleep mode while the recipient staysawake till the session ends thereby sending an ACK packetFor an idle channel the power consumption is rather lowwith this method Nevertheless there is a problem of longwake-up preamble limiting throughput and causing anenormous power consumption overhead in reception [38]All nodes overhearing a transmission bear the overheadtogether with the recipient node -e protocol can addressissues concerning low-power transmission but has nomechanism of nodes adapting to varying traffic patterns[32] To compensate for drift between clock at the accesspoint and on the sensor node there is a need to compute thewake-up preamble duration -e drift varies directly withtime since previous resynchronization and is given by thefollowing equation

TP min 4θl TW( 1113857 (11)

where θ is the time-base quartz frequency tolerance and l isthe interval between two transmissions Increase in trafficvolume naturally reduces overhearing

Journal of Computer Networks and Communications 11

Radiomodel for protocols with lower power transitiondelays between transceiver states and their powerconsumption are modeled Some of the states includeDOZE where the transceiver neither transmits norreceives but is prepared to power on to the receivetransmit state very fast RXwhere the transceiver listensto the channel receives data or demodulates data fromthe noisyidle channel and TX which is the trans-mission state by the transceiver Let TS be the set-uptime needed to turn the transceiver from the DOZE tothe RX or TX state TT the turn-around time needed toswitch between RX and TX states and B the trans-ceiverrsquos bit rate PZ PR and PT are the values of powerconsumed in the DOZE RX and TX statesrespectively

1113954PR PR minus PZ

1113954PT PT minus PZ(12)

where 1113954PR and 1113954PT are the changes in power as a result ofbeing in RX and TX states respectivelyTraffic model let N be the number of sensor nodesconnected to an access point (AP) and λ be the global rateat which the downlink Poisson traffic arrives at the APfrom the fixed network Traffic toward each sensor node isthe same and is given by λN Average packet interarrivaltime Lwith which data packets arrive at a node is given byL Nλ If TC and TD are the control packets and datapackets respectively assuming a small traffic volume theglobal interarrival 1λ far exceeds the summation ofdurations for the data packet and control packet and theturn-around time Mathematically it is given as

1λ≫TD + TC + TT (13)

-e average power consumed by WiseMAC andtransmission delay incurred are given by equations (14)and (15) [38]

PW PZ +1113954PR TS +(1B)( 1113857

TW+

1113954PR X + TD + TT( 1113857

L

+ 1113954PR(N minus 1)Y

L

(14)

where

X 2θL 1 minus eminus TD4θL( )1113874 1113875

Y T2D + 12TDθL

2TW1 minus e

minus TW4θL( )1113874 1113875

DW TD +TW

21 minus e

minus TW4θL( )1113874 1113875

+ 2θL 2 minus eminus TD4θL( ) minus e

minus TW4θL( )1113874 1113875

(15)

where PW and DW are the power consumed byWiseMAC and the transmission delay for WiseMACrespectivelyAdvantages

(i) It achieves low-power transmissions in in-frastructure sensor networks through preamblesampling while checking for activity

(ii) It reduces overhearing due to increased trafficvolume

Disadvantages

(i) It has a problem of long wake-up preamble limitingthroughput and causing enormous power con-sumption overhead

2114 X-MAC It is a simple minimal transmission powerMAC protocol that uses a short preamble with decouplingfor the sending and receiving node sleep plans inWSNs-isleads to a significant reduction in energy usage at sendingand receiving nodes and decrease in per-hop latency andgives flexibility in adapting to bursty and periodic sensordata sources [32]

Protocol design X-MACrsquos design is aimed at increasingenergy efficiency reducing data latency and increasingthroughput -is enables application in different kindsof packetized and bit stream digitized radios as well assimplified low-overhead distributed application -edesign is as such in order to address the low-powerlistening problems including overhearing excessivepreamble and incompatibility with packetizing radiosas in [32]Asynchronous duty cycling a node with information totransmit will send the packets after sending a preamblewhereas the remaining nodes will preserve their sleepschedules that are not synchronized -e receiver willsample the link on waking up and stays awake if it detectsa preamble and if it is the target it will return to sleep if itis not the target after getting the whole preambleEmbedding the target ID in the preamble to avoidoverhearing the lengthy preamble is divided into manyshortened packets embedded with the receiver nodersquosID so as to eliminate overhearing One lengthy pre-amble is constituted by the shortened preamble packetstream Any node on receiving the short preamble afterwaking up will examine the target ID on the packet Itgoes back into the sleepmode right away resuming dutycycling if it is not the targeted recipient or stay awakefor successive packets if it is the one -is approachchecks energy wastage due to network density It canalso be used on several types of radiosReducing excessive preamble using strobing it ispossible to enable low-power transmission and alsosave energy by reducing time for preamble trans-mission through the use of enhanced preambles inaddition to preamble sampling X-MAC does not sendan endless preamble packet stream but puts slight

12 Journal of Computer Networks and Communications

pauses in between packets to allow the sending node topause and listen to the channel and allow the receivingnode to resend an early ACK packet during that pause-is makes the sending node begin sending datapackets and the receiving node cut excessive preamblesand thus decrease per-hop latency and energy wastageon unnecessary waiting and transmissionPacketizing radios low-power listening is limited incapability for supporting packetizing radios butX-MACrsquos short strobed preambles can support all typesof digital radiosAdaptation to traffic load while a number of WSNapplications are reliable in traffic production they stillmust adapt to changing traffic Nodes are going to haveseparate sleep schedules depending on traffic loadConsidering systems that have time-fluctuating trafficall predetermined static schedules are suboptimalOptimality Buettner et al [32] model the expectationof energy as follows

Es (preamble energy + energy perACK listen)

lowast(expected preamble minus listen iterations)

+(energy to send parket)

Es PTxSp + PRxSal1113872 1113873 Rl + Rs( 1113857

Rl minus Sp+ SdPTx

(16)

where Es expected energy to send a packet PTx andPRx are the transmission and receiver power re-spectively Sp Sal and Sd are the duration of thesenderrsquos preamble ACK listen and data transmissionperiods respectively and Rl and Rs are the receiverrsquoslisten and sleep periods respectively Expected energyto receive a packet is given as

Er (listen cycle energy + sleep cycle energy)

lowast(expected iterations for preamble arrival)

+(energy to sendACK) +(energy to receive packet)

Er PsRs + PRxRl( 1113857

1 minus 1 minus Pd(t)( 1113857 Rl + Rs( 1113857+ RaPTx + RdPRx

(17)

And the one-hop expected latency is given as

Lat (preamble duration + ACK listen)

lowast(expected number of iterations)

+ duration to send packet

Lat Sp + Sal1113872 1113873 Rl + Rs( 1113857

Rl minus Sp

(18)

Advantages

(i) It increases energy efficiency(ii) It reduces data latency(iii) It increases throughput

2115 A-MAC (Advanced MAC) A-MAC is a TDMA-based protocol whose design is aimed at enhancing thelifetime of the network for low rate and reliability of datatransmission [39] To assign time slots it does not depend ona central controller like most protocols but it gathers in-formation from its neighbors so as to allocate the slotsthrough a distributed algorithm Energy is controlled byscheduling power when there is not a single event Its designcomprises many frames with each having many time slotswhich at the beginning have got a beacon transmitted tosynchronize and exchange information with neighborsNodes to participate in the subsequent session are selectedby the controlled node so they do not go into the sleep modewith others [13] -is protocol operates under four time slotdivisions namely initial wait discover and active states Anode in the initial state begins by listening to the medium forthe beacon packet from other nodes for network synchro-nization which when received makes it change the timerthrough subtraction of beacon transmission time from re-ception time To avoid synchronization problems due todrift and allow for continuity the node has to get thestrongest beacon signal It then enters the wait state afterchoosing a number of waiting frames so as to reduce theprobability that nodes get into the discover state simulta-neously In case the synchronization beacon is lost while stillin the wait state it will return to the initial state for anotherone On expiry of the waiting counter it enters the discoverstate and collects information from neighbors by listening totheir beacon messages On successful selection of a time slotit enters the active state where it endlessly transmits beaconsat the initiation of the slot A node goes to sleep when abeacon has been transmitted and there are no more data fortransmission or in case the neighborrsquos beacon received in-dicates no data packets are coming in [39]

Advantages

(i) It can enable multicasting(ii) It enhances the lifetime of the network(iii) It offers reliable data transmission

2116 L-MAC (Lightweight MAC) Based on TDMA thisprotocol is aimed at minimizing the number of transceiverswitches adapting the nodal sleep interval to data trafficamount and making implementation less complex [40] Apacket is transmitted comprising a control and a data partControl packets have a fixed size and the time slot con-trollerrsquos ID -ey show the distance from the transmittingnode for simplified routing to the network gateway as well asaddressing target node and reporting data unit length andinternode synchronization Adjacent nodes are more in-terested in getting control packets from nearby nodes -enodersquos transceivers are switched off until the subsequenttime slot unless the node is also addressed or the message isomnicast If so it listens to the data unit which may not fillthe remaining time slot in totality-e transceivers on eitherside are turned off when the transfer of packets is successfulEnergy wastage through idle listening in uncontrolled timeslots is avoided via a short time-out interval Network nodes

Journal of Computer Networks and Communications 13

are able to communicate without colliding-is saves energyand increases the lifetime of the network A node in L-MACcan only transmit one message per frame [13] -e networkis set up in such a way that all nodes are not synchronized onbeing switched on-e gateway takes control of a timeslot soas to have them synchronized and its immediate neighborgets the control messages thereby synchronizing their timerswith the gatewayrsquos Neighboring nodes learn of multiplegatewaysrsquo time slots in their vicinity just after one frame-ejust synchronized nodes randomly choose time slots forcontrolling as long as they are not occupied -e nodes thuspreserve a table for the neighboring nodes which makes atimeslot reusable after a minimum of 3 hops and preventsmessage collision All nodes keep their slots before thebattery is down or learn of an impending collision All nodescan track their hop distances to a selected gateway node andthe information is communicated in control messages [40]

Advantages

(i) It minimizes the number of transceiver switches(ii) It adapts the nodal sleep interval to data traffic

amount(ii) It makes implementation less complex(iii) It offers improved network lifetime(iv) Energy wastage through idle listening in un-

controlled time slots is avoided via a short time-out interval

Disadvantages

(i) L-MAC is not very good for mobile SNs since slotsare calculated only once

(ii) It suffers idle listening because nodes keep listeningto slotsrsquo control parts so as to get data and enableother nodes to join the network all the time [13]

2117 C-MAC -is was designed for WSNs with 3 seg-ments namely (i) aggressive RTS having double channelcheck to assess the channel (ii) anycast to initialize the flowand (iii) convergent packet forwarding to stabilize the flowIt promises an improvement in energy efficiency lowerlatency and enhanced throughput while avoiding syn-chronization overhead though it operates only in low duty-cycling apps especially where there is less traffic But whenthere is more traffic the protocol employs anycast-basedpacket forwarding for waking the nodes up or finding aforwarder quickly and converges from route-suboptimal toroute-optimal unicast After this energy could be preservedby the nodes using a synchronized wake-up plan [13]

Advantages

(i) It improves energy efficiency(ii) It lowers latency(iii) It enhances throughput avoiding synchronization

overhead

Disadvantages

(i) It operates only in low duty-cycling applicationsespecially where there is less traffic

22 MAC Protocols for WMSNs

221 Diff-MAC -is is a CSMACA-based QoS-awarehybrid-priority-based MAC protocol for WMSNs It targetsenhancing the channel utilization while making use of ef-ficient service differentiation to do coordination for mediumaccess of each traffic class while giving fair quick datatransfer It offers QoS using the following features [14]

(i) It balances energy consumption and delay throughsensor node duty-cycling adaptation depending onthe dominant traffic class

(ii) It has a feature for fragmentation and messagepassing that breaks lengthy video frames intosmaller packets that are transmitted as a burst thusreducing retransmission costs if there are any MACfailures

(iii) -ere is fair data delivery amongst the nodes andtraffic classes respectively resulting from intranodeand intraqueue prioritization reducing unbearableperformance

(iv) -e number of collisions can be reduced and packetlatencies are kept within limits by adjusting the sizeof the congestion window in this protocol as per thetraffic needsAdvantages

(i) Diff-MAC is very suitable for WMSNs since itadapts quickly to changes in networkconditions

(ii) It is a fair protocol(iii) It is scalable

Disadvantages

(i) It is quite difficult to monitor the statistics ofthe network

(ii) Adapting dynamically to the network condi-tions is not an easy task

(iii) It exhibits some degree of latency in packetdelivery

-e implementation of Diff-MAC on Imote2 indicatesthat resource requirements can be met on the sensorhardware Results of extensive simulation runs showed thatDiff-MAC outperforms its competitors

222 Cluster-Based On-Demand Multichannel MAC Pro-tocol for WMSN (COM-MAC) -is is an energy-efficientprotocol in WMSNs with high throughput and reliable datatransmission It achieves energy efficiency by schedulingmultichannel media access to reduce channel contentionand remove collision idle listening and overhearing [41]

Design it has a three-phased operation with the (i)request phase (ii) scheduling phase and (iii) datatransmission phaseRequest phase here all nodes have two protocols eachfor sending requests to the cluster head (CH) ie thecontention-based and contention-free TDMAFDMA

14 Journal of Computer Networks and Communications

protocols -e contention-based protocol allocates achannel per node for sending the request message andfree channels are useable as control channels Forcontention-free protocols a control slot is allocatedwhen a network is deployed Here a slot per channel isallocated to a different node enabling requests to be sentwithout interfering with transmissions from othernodes -e contention-based protocol operates undertwo steps (i) Control channel assignment phase achannel is assigned per node for transmitting a request(REQ) message A channel is assigned to multiplenodes and nodes are evenly distributed to availablechannels for congestion control (ii) Request trans-mission phase nodes with data send a REQmessage viathe assigned control channel to notify the CH whichresponds with an ACK message -e contention-freeTDMAFDMA protocol too has two phases includingthe control slot assignment and request transmissionphases For unreliable channel conditions and lowtraffic load the contention-based protocol is fit for usebecause of less collisions and congestion whereas thecontention-free protocol is preferable in heavy trafficload and reliable channel applicationsScheduling phase from the above phase a schedule isgenerated by the CH for coordination and this enablesnodes to transmit and broadcast their data via controlchannels-e schedule for broadcasting only comprisesdata for nodes assigned to a channel -e schedule isrebroadcast for transmission reliability enhancementbut for energy efficiency the nodersquos transceiver isswitched off when all the schedule is completely de-livered till it is the time for transmitting again-e timeslot and data communication radio channel are allincluded in the schedule generated for each nodeData transmission phase on getting the schedule thesensor node sends its data as per the time slot and thechannel assigned -e time slot is divided into datatransmission andACK sections ACK supports link layererror control Using implicit selective repeat ARQ theCH sends an ACK message for all received packets Ifthere is no ACK the packet is assumed lost and hence aretransmission during the subsequent interval is done-is enhances reliability of transmissions though excessdelay is possible here which is also intolerable for real-time critical apps like video or audio -is gives rise to ahybrid MAC protocol to exploit the unused spectrumduring data transmission sessions [41]Advantages

(i) It achieves energy efficiency by scheduling mul-tichannel media access to reduce channel con-tention and remove collision idle listening andoverhearing

(ii) It maximizes network throughput by dynamicallyallocating time slots and channels for nodes byexecution of a traffic-adaptive QoS-aware sched-uling algorithm depending on needs for QoS andtraffic data information

(iii) It increases transmission reliability by in-corporating spectrum-aware ARQ to avoid spec-trum wastage

223 EQoSA -is protocol is developed for video andimage transfer across WMSNs in a hybrid mode withprovision of QoS Concerning the active sensor nodes andtheir traffic loads it employs dynamic session sizes andchanges bitmap-assisted (BMA) MACrsquos fixed session sizeNodes declare any data available for transmission duringcontention time -is forces the cluster head to make aschedule of how slots have been assigned It communicatesthis schedule to the nodes giving each node a given numberof slots per session and so enabling the busty traffic to beaccommodated [14] -e protocol does not perform wellwith DiffServe mechanisms but does well with schemessupporting busty traffic loads and strong cluster heads arenecessary in WMSNs that can assign and broadcast slots

224 QoS-Supported Energy-Efficient MAC (QE-MAC)Based on the IEEE 80211e standard this protocol ensuresfairness and has low latency and jitter and better efficiencyon energy usage than other QoS-aware MAC protocols forWMSNs It operates under a two-phased routine

-e first phase introduces the innovating prioritymechanism in IEEE 80211e QoS and several data types areassigned priorities Its coordination function is a combi-nation of distributed contention-based channel allocationand centralized polling-based channel access mechanisms

-e second phase preserves the nodesrsquo energy usingdynamic duty cycling -ere is an exchange of RTSCTSpackets to supplement CSMACA by muting nodes near thesenderreceiver during packet duration [42]

Advantages

(i) It ensures fairness(ii) It has low latency and jitter and better efficiency on

energy usage than other QoS-awareMAC protocolsfor WMSNs

225 Black-Burst (BB) Contention A contention-basedMAC scheme for enhancing instantaneous QoS access tocarrier-sense WSNs is proposed in [42] In this schemenodes contend for the channel until it is free Real-timepackets are given higher priority over non-real-time packetsand if a node is transmitting real-time packets it is prior-itized over other nodes [15] Voice and video are some of thereal-time apps in consideration here which periodicallyaccess the common radio channel when transmitting -eseneed constrained end-to-end delay of packets at the MAClayer -is protocol majorly does the following [42]

(i) Prioritization of real-time traffic(ii) Implementation of the round-robin mechanism in

real-time nodes(iii) Bounded access delays among real-time packets

Journal of Computer Networks and Communications 15

(iv) Implementation of real-time communication withvarying needs of bandwidth good for multimediatraffic

Amechanism for dynamically assigning packet prioritiesbased on their deadlines and the traversed hops is proposedin [43] -e ReAllocative Priority (ReAP) mechanism ismeant to provide QoS at the MAC layer to video traffic -isscheme introduces adaptive TXOP by modifying the TXOPdynamically based on the queue length [15] Other protocolsin this category include dynamic duty cycle and adaptivecontention window-based MAC protocol frame sharing(FRASH) MAC protocol and real-time independentchannels (RICH) MAC protocol

Advantages

(i) It can implement real-time communication withvarying needs of bandwidth

23 Challenges and Opportunities(a) One of the challenges in satellite transmission is

cross-talk leading to interference in different fre-quency bands It is necessary to increase the rise andfall times as well as protect sensitive nodes as wecircumvent floating ones if we are to regulate cross-talk as in satellite communication therefore Wefurther need to investigate the possibility of usingdifferential signaling -is constitutes an opportu-nity for further research

(b) Another challenge is that the hidden and exposedterminal problems are still noticeable despite the useof RTSCTS solutions in MAC since collisions arestill possible from transmissions of separate nodesAt the MAC layer ACKs might block efficientchannel usage by prohibiting exposed terminalsfrom reusing the channel and may be eliminatedleaving no proof of sender data packet receptionUsing a directional antenna is one of the proposedtechniques of high-speed WSNs like IEEE 80211-is constitutes an opportunity for further study intothese issues for better solutions though

(c) Another challenge is that since nodes in contention-based MAC protocols are mobile there are manycollisions by the packets and schedule in-consistencies in schedule-based protocols in the two-hop neighborhood information when nodes areentering or leaving [44] Deng et al [45] propose amobility-based clustering (MBC) protocol for mo-bile nodes to enhance the packet delivery rate -isprotocol in comparison with CBR has 25 chancesof decreasing collision-induced packet losses and50 chances in comparison with low-energyadaptive clustering hierarchy (LEACH)

(d) -e process of analyzing and evaluating routingperformance for WMSNs is quite complex becauseof the bursty multimedia traffic Because of the wayencoding scheme frames are structured as well as thechanges in between the scenes that occur naturally

the compressed video frames display some substantialburstiness on several time scales -ere will also befrequent route recalculation that is as well energy in-tensive for nodes frequently capturing and transmittingenergy -e solution here could as well possibly be theuse of the MBC protocol Route recalculation tooshould be dynamic to cater for the mobile nodes whichmight be in and out of the range at different instants oftime Recently the use of smart dumpsters for effectivewaste management in smart cities has also been pro-posed as a possible solution [46]

(e) For WSNs their routing protocols do not look atmultimedia applications which require enormousbandwidth and processing energy and should betransmitted in real time with utmost fidelity Extrachallenges in routing are imposed in WMSNs indeployment of heterogeneous sensors since therecould be audio video and still pictures in addition toscalar data during transmission Heterogeneous datahandled also have different QoS requirements andwith many different business needs different ser-vices with varying requirements give rise to a bigchallenge for the routing design -ere is a need todesign protocols that can support WMSNs such asEQoSA [14] for this purpose

Table 2 summarizes the reviewed MAC protocols forWSNs and the ones designed for suitability in WMSNs

3 QoS Strategies at the Application Layer

We discuss many protocols used to solve different needs ofcommunication between machines

31 ConstrainedApplication Protocol (CoAP) It is a requestresponse protocol designed by the Internet Engineering TaskForce (IETF) for constrained devices such as those havinginsufficient RAM or CPU and WPANs with low-powerconstraints resulting in poor transfer of packets andenormous overhead -e protocol was designed for ma-chine-to-machine (M2M) apps and system automation toreduce bandwidth requirements increase packet transferdecrease the overhead and make work simpler leading tolightweight implementation HTTP commands like GETPOST PUT and DELETE are used in a client-server ar-chitecture Using the publishsubscribe architecture CoAPcan support multiple users and hence yield better perfor-mance via multicasting enabling asynchronous messageexchange -is protocol can support both multicasting andunicasting via UDP and hence is a reliable and simpleprotocol -is is done via the messaging sublayer (for re-liability based on UDP) and the requestresponse sublayer(for communication and interaction) [16]

It uses the following messages

(i) Confirmable message this guarantees reliabilityusing ACK messages sent by the receiver node-ese could be synchronous or asynchronous ifthere is a need for more computational time

16 Journal of Computer Networks and Communications

(ii) Nonconfirmable message here no ACK messagesare sent back

(iii) Acknowledgment (ACK) message this confirmsreceipt of a confirmable message

(iv) Reset message this confirms receipt of an un-processed message in case any critical parts nec-essary for interpreting the message are missing

(v) Piggybacked response the receiver directly re-sponds while getting the ACK message

(vi) Separate response the reply by the receiver is dis-crete from the ACK message

CoAP has no in-built security features It is secured bythe datagram transport layer security (DTLS) that runs ontop of UDP though it was not designed for the IoT -ismakes its suitability debatable since DTLS offers no supportfor multicasting It also needs more packets for handshakingthereby increasing the traffic volume and computationalresources and so shortening the lifespan of mobile devicessince they use batteries [17]

Advantages

(i) It can support multiple users and hence yieldbetter performance via multicasting

Table 2 Comparison of MAC layer protocols used in QoS provisioning

Protocol Features QoS parameter Comments Access Priorityassignment Energy eff Designed for

S-MAC[22] Collision avoidance Energy latency Requires periodic

synchronization CSMA Hybrid Medium WSNs

B-MAC[23] Collision avoidance Energy delay

throughput Flexible interface CSMA Dynamic High WSNs

T-MAC[25] Collision avoidance Energy Complex nonscalable CSMA Dynamic High WSNs

C-MAC[13]

Dynamic channelallocation Energy Multichannel MAC CSMA Dynamic Medium WSNs

CU-MAC[24]

Multichannelmechanism

-roughputlatency High packet delivery ratio TDMA Dynamic mdash WSNs

A-MAC[39]

Per flow serviceguarantees Delay m-ary tree model

dynamic prioritiesTDMACSMA Dynamic Medium WSNs

ER-MAC[30]

Emergencyresponse

Energy delaydelivery ratio

Adapts to traffic andtopology changes

TDMACSMA Hybrid Higher than Z-

MAC WSNs

RI-MAC[31] Collision detection Delay packet

delivery ratioHigh packet delivery ratio

and throughput CSMA Dynamic Medium WSNs

SRI-MAC[33] Collision avoidance Energy delay Good data transfer and

energy consumption TDMA Dynamic High WSNs

H-MAC[34] Slotted sleep time -roughput

latencyData transmission only in

sleep timeCSMAAloha Dynamic Medium WSNs

Z-MAC[36] Collision avoidance Energy delay

throughputVery good channel usage

and error syncCSMATDMA Dynamic Medium WSNs

Q-MAC[37]

Intrainternodescheduling Energy High QoS improved

energy efficiency CSMA Hybrid High WSNs

X-MAC[32]

Overhearingavoidance

Energy delaythroughput

Reduces excessivepreamble by strobing CSMA Adaptive

static Medium WSNs

L-MAC[40] Collision avoidance Network

lifetime energyReduces the number oftransceiver switches TDMA Static Medium WSNs

WiseMAC[38]

No downlinkchannel collision Power delay Significantly low-power

consumptionAlohaCSMA Static Higher than

ZigbeeInfrastructure

WSNsAR-MAC[27] Collision avoidance Energy delay Uses data packets and

control packets TDMA Static Higher thanIEEE 802154 WBANs

BB-MAC[42]

Real-time trafficprioritization Delay Employs round robin in

real-time nodesCSMATDMA Dynamic High WMSNs

COM-MAC [41] Collision avoidance Packet delay

throughputBalances reliability and

retransmissionTDMAFDMA Dynamic High WMSNs

EQoSA[14]

Busty traffic loadaccommodation Energy Provides QoS support for

video and imagesTDMACDMA mdash High WMSNs

QE-MAC[42] Collision avoidance Energy jitter

latencyRuns central polling-like

channel access CSMA Dynamic High WMSNs

R-MAC[26] Collision avoidance Energy delay Reservation based mdash Static High Underwater

SNs

Journal of Computer Networks and Communications 17

(ii) It enables asynchronous message exchange(iii) It is a reliable and simple protocol

32 Message Queue Telemetry Transport (MQTT)Developed by IBM and later improved by OASIS with anintention of decreasing bandwidth MQTT is a simple open-source publishsubscribe protocol analogous to the client-server version It is appropriate for lightweight IoT appli-cations and M2M transmissions in constrained situationslike limited power computation capability and memory orbandwidth -e network bandwidth and the computationalresource usage are decreasing since clients are not updatedmaking publishsubscribe protocols outperform requestresponse protocols via IoT requirements [16] -is protocolcan run over TCPIP and so has the following advantages

(i) Delivering of packets with reliable guarantee(ii) Enabling multicasting(iii) Remote device session establishment(iv) Reduction of transport overhead and protocol ex-

change to minimize network traffic(v) Provision of the notification mechanism when

anomalies occur [47]

MQTTrsquos design targets economical use of bandwidthand battery so it has a major application in FacebookMessenger MQTT ensures QoS and reliability of messagedelivery by providing the following three QoS levels

(1) Fire and forget a message is sent once depending onthe networkrsquos best effort and there is no need forACK (lowest QoS level)

(2) Delivered at least once a message is sent at least onceand an ACK message is needed to avoid duplicates

(3) Delivered exactly once by a four-way handshake itensures a message is sent exactly one time (highestQoS level)-e architecture it has three key components (i)publishersmdashlightweight nodes sending data to thebroker before returning to sleep (ii) broker whichorganizes sensory data in topics and sends them tointerested subscribers and (iii) subscribersmdashIoTapps interested in data sent by sensors through thebroker as shown in Figure 3

MQTT is a simple lightweight protocol that is betterthan CoAP via low and high packet loss probability interms of throughput and exhibits lower latency in lowsampling rates [48] It supports a number of imple-mentations for embedded devices and mobile applicationsand offers web-socket support too Nevertheless it hassome shortcomings

(i) It is only employed in simple data types(ii) It exhibits high latency in high sampling rates(iii) It is unsuitable for real-time apps in such cases(iv) It does not provide automatic discovery andor

sufficient security at the protocol level [49]

33 Secure Message Queue Telemetry Transport (SMQTT)-is is an extension of MQTT that is based on lightweightattribute-based encryption using elliptic curve cryptogra-phy -e protocol has three entities as shown in Figure 4(i) publisher that publishes information under a certaintopic (ii) subscriber that obtains the same topic in-formation via a broker and (iii) public key generator(PKG) or broker that acts as the trusted third party -eprotocol has four phases namely (a) set-up phase (b)encrypt phase (c) publish phase and (d) decrypt phase

(a) Set-up phase publisher and subscriber provide aunique ID with attributes for registration with thePKG -is generates a secret master key set andpublic parameters that it publishes with set U A1A2 An For ciphertext policy-attribute-basedencryption (CP-ABE) the PKG gives a key set todevices with attributes [50]

(b) Encrypt phase based on the access tree the publishercreates an access policy An access tree is sent by thepublisher to the PKG for generating the key-gen-eration policy in the key policy- (KP-) based ABEalgorithm -e publishers know beforehand thetopics and subscribers to access as well as the rulesfor access Subscribers are issued with private keysfor all communication sessions Publishers in theCP-ABE algorithm come up with access trees andrules where the payload is encrypted by the pub-lisher who also issues more info needed for de-cryption and the rules With the help of public keycryptography data are encrypted giving rise to ci-phertext on the basis of KPCP-ABE and the receivergets SUBACK from the broker [51]

(c) Publish phase ciphertext is embedded in theSPublish command as payload and topic name isdefined in the header by the publisher -e brokerreceives the SPublish packet and replies withPUBACK prompting the publisher to respond withthe PUBREL packet -e broker then broadcasts totopic subscribers and deletes the data forwarding thePUBCOMP packet to the sender node

(d) Decrypt phase the encrypted information isdecrypted by the subscriber with the help of a secretkey Considering KP-ABE on verification that itsatisfies the access policy the subscriber invokes thePKG for a corresponding key which it sends afterverifying the request With CP-ABE subscribers usethe private key for decryption after authenticationthereby enabling offline interaction [51]Advantages

(i) -e information being encrypted makes it moresecure

34 MQTT for Sensor Networks (MQTT-SN) -is is a ver-sion of MQTT designed to operate under environments oflow bandwidth high link failures and short message lengthIt is a publishsubscribe (pubsub) protocol designed for

18 Journal of Computer Networks and Communications

UDP-based WSNs to visualize communication for low-power devices [5152] It comprises a publisher device actingas the MQTT-SN client to forward communications to agateway that also changes the MQTT-SN message to theMQTTmessage before delivering it to the broker that sendsit to MQTT-SN subscribers [51]

It has a gateway for changing protocols from MQTT toMQTT-SN (Figure 5) [52] -is protocol supports nodes in

the sleeping mode using an offline keep-alive procedure thatcan be used by battery-operated devices to get into thesleeping mode To support sleeping clients MQTT-SN has anew offline keep-alive procedure with which battery-oper-ated devices can go to the sleeping state in which theirmessages are buffered at the server until they are awake InMQTT-SN clients with no preconfigured gateway addresscan figure out the real address of their operational gateway

Publisher

Subscriber

Subscriber

Broker

Publisher

4G

Figure 3 Architecture of the MQTT protocol

Register by mentioning URI Register by mentioning URI

Access tree

Set-up

Publisher ReceiverPKG(broker)

Key management Key management

Encrypt data using public parametersand random key This is done using

CP-ABE Publish it as

Subscribe (topics)SUBACK

S-Publish (TopicEncrypt (data))S-Publish (TopicEncrypt (data))

Encryptand publish

PUBACK

PUBACK

PUBRELPUBRELPUBCOMP

PUBCOMP

Figure 4 SMQTT protocol

Journal of Computer Networks and Communications 19

within the network through a discovery procedure Simul-taneously we may have several gateways in a single wirelessnetwork that can be cooperative via load sharing

MQTT-SN architecture it is composed of MQTT-SN clients gateways and forwarders (Figure 5)rough the MQTT-SN gateway (GW) the MQTT-SN protocol is used by the MQTT-SN clients toconnect to the MQTT server Integration betweenthe MQTT-SN GW and the MQTT-SN server mightbe possible or not Translating between the MQTT-SN and the MQTT for a stand-alone GW is doneusing the MQTT-SN protocol If a GW is detachedfrom the network it can be accessed by the MQTT-SN clients through a forwarder is just encapsu-lates the received MQTT-SN frames that are on thewireless side before forwarding them to the GW un-altered It then decapsulates the received frames tooresending them to the client as sent by the GW [52]ere are 2 types of gateways (a) transparent gatewayand (b) aggregating gateway as shown in Figure 6

(i) Transparent gateway this launches and sustains adirect MQTT connection between the MQTT-SNclient and the MQTT server which is entirely re-served for end-to-end client-server messaging enumber of servers and that of MQTT-SN clientsconnected to the GW are the same is transparent

GW translates syntax between these protocols and iseasier to implement compared to the aggregatingGW But the MQTT server must support dierentsessions for all connected clients Some MQTTservers are congured to enable few parallel sessions

(ii) Aggregating gateway this maintains a single MQTTconnection to the server on behalf of all clients Allexchanged messages from the MQTT-SN client to theaggregating GW are directed to the GW which sortsthe information to forward to the server e aggre-gating GW decreases the number of MQTT sessionssimultaneously supported by the server in WSNs withmany battery-operated sensors and actuators (SAs)having limited storage and processing capabilities It ishowever more complex to implement compared to atransparent GWe general format of the MQTT-SNprotocol message has two or four message headeroctets and n octets of variable header that is optionaldepending on the message in question MQTT-SN has256 message types (MsgType) with those from0X1Eminus 0XFD 0X19 and 0XFF reserved for futureusage [51]

Advantages

(i) e protocol can operate under environmentsof low bandwidth high link failures and shortmessage length

MQTT-SN gateway

MQTT-SN

MQTT

MQTT-SN

MQTT-SN

MQTT-SN

MQTT-SN client

MQTT-SN clients

MQTT broker

MQTT-SN client

MQTT-SN client

MQTT-SN gateway

MQTT-SN forwarder

Figure 5 MQTT-SN architecture

20 Journal of Computer Networks and Communications

(ii) It is a pubsub protocol designed for UDP-basedWSNs to visualize communication for low-power devices

(iii) It can be optimal for low-cost battery-operatednodes having low processing power and lessstorage

(iv) It supports nodes in the sleeping mode using theobrvbarine keep-alive procedure used by battery-operated devices

35e Extensible Messaging and Presence Protocol (XMPP)e standardization of this protocol by the IETF made itpopular for usage over the Internet especially in instantmessaging and chatting platforms It is however susceptibleto spamming attacks and its failure to have worldwidesupport has made Google too withdraw its support of thestandard [17] e protocol is mainly useable in the client-server architecture but can also be applicable in client-clientXMPP communication [53] Among the application layerprotocols XMPP is the only one that supports both asyn-chronous (publishsubscribe) and synchronous (requestresponse) services while running over TCP but can still runover other protocols

Given the simplicity of its implementation and ease ofusability in XMPP instant messaging (IM) is suitable forXMPP and is implemented using the message stanzaltmessagegt ese stanzas are usable in one-to-one ormultiuser sessions or chatting or in error reporting andsending headlines Attribute type of the stanza denes themessage type e attribute type could be a chat group chatheadline normal or error An instant messaging network inXMPP can support many XMPP clients and servers with agateway congured to enable other IM network users toconnect to other XMPP users e XMPP server mainlycontrols sessions and XML streams to clients and servers inaddition to routing XML stanzas in the streams [53] eTCP ports 5269 and 5222 have been respectively registeredfor the XMPP server-to-server connections and XMPP clientconnections by the Internet Assigned Numbers Authority(IANA) XMPP enables specication of XMPP ExtensionProtocols (XEP) to enhance its usefulness [17]

XMPP security clients connect to the server usingtransport layer security (TLS) and simple authentication andsecurity layer (SASL) protocols which are at the same timeused by the servers for interdomain communications Fordata condentiality and integrity TLS will encrypt thestreams of XML for all sessions To conrm the identity of anyclient that needs to access the server the SASL protocol doesthe authentication of the XML stream For any communi-cating session with a server to start the client needs to resolvethe serverrsquos DNS hostname For protection of the server of theclient from third-party attacks the server maintains the au-thentication details secret ie IP address and access methodand only original connections of the server are required

In server-to-server communications the SASL protocolis employed for authentication and condentiality Server-to-server communications may as well be deactivated from aserver by an administrator depending on the organizationalpolicy To guard against domain-spoong attacks serverdialback can be a good solution since servers support itsecuring more the XML stanzas Nevertheless it cannot beused to authenticate secure or encrypt in-server streamsand the server identication resulting therein is not strongenough to be relied on It further cannot secure against DNSpoisoning attacks and IP session hijacking for remote do-mains [53] High-security domains can use TLS and SASL

Advantages

(i) It is popular for usage over the Internet especially ininstant messaging and chatting platforms

Disadvantages

(i) It is susceptible to spamming attacks(ii) It has failed to have worldwide support making

Google withdraw support of the standard(iii) It is not secure against DNS poisoning attacks and

IP session hijacking for remote domains

36 Advanced Message Queuing Protocol (AMQP) isprotocol can be utilized by a number of platforms formessage exchange since it is open source and written inmany languages It is a publishsubscribe model depending

MQTT-SNclients Broker

MQTT

(a)

MQTT-SNclients Broker

(b)

Figure 6 (a) Transparent gateway and (b) aggregating gateway

Journal of Computer Networks and Communications 21

on a reliable and efficient messaging queue with capacity forinstant messaging or e-mail for different applications es-pecially in IoT environments It differs from others in thesense that message types can be specified and have theirsource traceable and how to do trade-off between perfor-mance security and reliability For reliability of messagetransmission AMQPmakes use of some delivery guaranteeslike at most once at least once and exactly once It can alsohowever utilize the TCP transport layer for guaranteeingreliability -e AMQPrsquos publishsubscribe method has twocomponents namely (i) exchange queue meant for routingof messages to an appropriate order in the queue and (ii)message queue that acts as storage pending their delivery toreceivers [16] In this protocol model the broker can makerouting decisions contrary to othermessaging systems whereapps using the queue have in them embedded the decision-making logic for recipients or senders -e developer mayneed to check each of the affected apps so as to change thelogic for routing and delivery of the messages [54]

Advantages

(i) It is convenient(ii) It has capacity for interoperability between dif-

ferent vendors(iii) It can use open standards to enable connection

between business partners(iv) It can enable innovations based on AMQP foun-

dations [55](v) It can enable supporting of mission-critical ap-

plications in e-commerce(vi) It can guarantee reliable delivery of messages

Disadvantages

(i) It is not very appropriate for real-timeapplications

(ii) It may not offer automatic discovery(iii) Interoperability is not guaranteed(iv) It lacks open-source libraries for constrained

gadgets [49]

37 Data Distribution Service (DDS) A product of theObject Management Group (OMG) DDS is a pubsubprotocol meant for M2M communications with two sub-layers -ese are the data-centric pubsub sublayer re-sponsible for connecting anonymous data publishers tosubscribers and the data-local reconstruction sublayer that isoptional and integrates DDS into the application layer [56]-e DDS architecture connects system components asshown in Figure 7 Publishers and subscribers are discon-nected with respect to (i) time (ii) space (iii) flow andbehavior (iv) platforms and (v) programming languages[57] A DDS data publisher generates topics and publisher-subscriber coupling is represented as topic name data typeschema and publishers and subscribersrsquo QoS attributesDelivering of data is done by the publisher layer -e datawriter and publishers together make a decision about anynecessary alterations for subsequent forwarding to thesubscribers who need to send data to IoT applications Data

readers are responsible for delivering topics to the sub-scribers after reading through them [56]

Some of the major components of the DDS architectureinclude the following [57]

(i) Topic this is a logical path specifying publicationand subscriptionrsquos data type between DataWritersand DataReaders For a successful session to begintopic names types and DataWriters and Data-Readersrsquo QoS should correspond

(ii) Domain this is a logical setting for transmissionthat can be utilized for separating and optimizingnetwork transmission in a setup of similar appli-cations If DDS apps share a domain ID then theycan exchange data among themselves

(iii) DataWriter and DataReader these are terminalobjects for writing and reading messages re-spectively to and from a universal data space

(iv) Participant this is an object that stands for apublisher or subscriber of the DDS app in a do-main acting like a container for other objects

Advantages

(i) It is highly reliable and offers QoS support(ii) It is highly scalable(iii) It is appropriate for real-time apps(iv) It is fault tolerant(v) It has high level of interoperability(vi) It does not need brokers that may act as

bottlenecks(vii) It offers automatic discovery(viii) It has a user-defined data structure for topic(ix) It supports publisher and subscriber decoupling

Disadvantages

(i) It is originally developed for only stand-aloneLANs

(ii) A lot of memory is needed(iii) It has no open-source libraries to support con-

strained devices(iv) It is hard to design and configure compared to

other protocols [49]

38 RESTful Services REpresentational State Transfer (REST)-is is an architectural style based on a set of principles thatdefine and address networked resources It offers web servicesthat enable transmission from the HTTP device to the devicewithin the IoT architecture and enables clients to reach theserver [16]-is protocol utilizes HTTP commands to backupthe requestresponse messaging model HTTP is a popularand secure protocol in the WWW using TLSSSL [16] ARESTful application has the following characteristics

(i) State and functionality are separated into distrib-uted resources

(ii) All resources are distinctively named by the use ofunique HTTP commands (GET POST PUT orDELETE over the Internet)

22 Journal of Computer Networks and Communications

(iii) It is a clientserver stateless layered protocol thatsupports caching REST uses the same methods asHTTP for all requests or responses with POST andGET for creating and retrieving resources PUT forupdating and changing the state of resources andDELETE for removing resourcesAdvantages of REST over arbitrary web services(SOAP) in the IoT context

(i) It reduces overhead(ii) It reduces parsing complexity(iii) It is stateless(iii) It integrates more tightly with HTTP [58](iv) It is easy to implement and learnis together

with its natural t over HTTP makes it apreferred method for Web 20 applications toshowcase their data and secure M2M sessionsin the IoT

(v) Applications that support RESTful web ser-vices exhibit better performance than the re-source-constrained WSN nodes [59]

(vi) REST is important for the IoT since it supportscommercial M2M cloud platforms

(vii) Its implementation in smart phones and tabletapps is also easy since it needs only an HTTPlibrary that is accessible to all operating systemplatforms [56]

Disadvantages

(i) Although it enjoys wide usage in commercialM2M platforms it is not likely to dominatebecause it is hard to implement

(ii) Its HTTP usage makes it incompatible withconstrained M2M devices leaving its usage toend applications

39 WebSockets It is a web-based protocol operating on asingle TCP channel and providing full-duplex transmissions

It permits secure duplex communication between a clientexecuting unreliable code in a restricted environment and adistant node communicating from that code while using theorigin-based security model employed by web browsers overTCP [60] e protocol is not a requestresponse model andis not a publishsubscribe model So unlike other protocolsa web-socket session is established by the client initiating ahandshake to the server to launch the session is onceestablished a full-duplex client-to-server connection beginsasynchronously and runs until terminated by either party[56] It gives a model for browser-based apps utilizing bi-directional communication with no need of many HTTPconnections to servers

WebSocket comprises two parts ie (i) the handshakehaving a clientrsquos message and the serverrsquos handshake responseand (ii) the data transfer HTTP headers should be repeated byapplications in all client requests and server responses arisingfrom the endless polling which might increase the commu-nication overhead subject to the application [61] WebSocketcan be used to come up with scalable real-time web appli-cations since it can enable full-duplex transmission via onesocket across the Web Its security can be ensured with thehelp of TLSSSL is protocol is suitable for resource-con-strained devices and IoTapplications will not have a good userexperience with it since it operates a clientserver architectureNevertheless it can outcompete all TCP-based protocolsbecause of the following advantages

(i) It is t for real-time transmission(ii) It has improved security(iii) It minimizes transmission overhead(iv) It can provide ecient messaging systems using

WebSocket ApplicationMessaging Protocol (WAMP)[17]

310 Streaming (Simple) Text-Oriented Messaging Protocol(STOMP) is is a TCP-based protocol designed for textmessage-oriented middleware (MOM) with the use of

Topic

Datareader

Datareader

Datareader

Datawriter

Datawriter

Datawriter

Subscriber PublisherPublisher

Data bus (DDS domain)

Subscriber

Topic Topic

Figure 7 DDS architecture

Journal of Computer Networks and Communications 23

HTTP-like commands for interoperability among platformslanguages and brokers Data communication takes placebetween the client and the broker in line with commands(like CONNECT DISCONNECT ACK NACK SUB-SCRIBE UNSUBSCRIBE SEND BEGIN COMMIT orABORT) then headers in the form ltkeygt ltvaluegt (one perline) then a blank line and then body content terminatingin a null character Server-client communication takes placeover a MESSAGE RECEIPT or ERROR frame with headerand body content format being analogous MOM productssupporting STOMP include Apache ActiveMQ or FuseMessage Broker HornetQ Open Message Queue(OpenMQ) RabbitMQ and syslog-ng through its STOMPdestination plugin [62] In addition to these other appli-cation layer protocols include simple media control protocol(SMCP) lightweight local automation protocol (LLAP)simple sensor interface (SSI) lightweight M2M (LWM2M)XMPP-IOT and simple object access protocol (SOAP)

Advantages

(i) It is lightweight for constrained networks(ii) It has flexibility to choose quality of services with

the given functionality(iii) It is standardized by OASIS Technical Committee(iv) It is easy and quick to implement

Challenges and opportunities

Here security threats are imminent because of the open-endedness of the application layer [63] -ese could includethe following

(i) -ere is poor or lack of a security design for anapplicationrsquos function Coming up with a gooddesign to address this could be a good ideaopportunity

(ii) Some of the programs might have backdoorsevading otherwise secure controls to allow un-authorized access to network resources We requiremechanisms to patch up such loopholes

(iii) Applications that lack or have weak authenticationmechanisms can be easily targeted by unlicensedusers to abuse the network Such mechanisms needto be strengthened further

(iv) Excessively complicated access control rules maybecause of their misunderstanding or poor writinggive unauthorized access allowing dubious activityrather than protecting the network they purport toprotect -ese should be simple and clear to un-derstand by all users

(v) In booting workstations without disks and managingnetwork devices we usually use the TFTP protocolalthough access does not require authenticating theuser by usernamepassword -is makes an intruderable to gain easy access to configaccess info as longas he can guess the filenames -is needs to be fixed

A few steps can be used to secure the application layerTo protect network data transmitted across the network

applications need to strengthen their encryption and au-thentication mechanisms Apps too need to enable strictcontrols of privileges for data access using good mechanismsto balance between usability and effectiveness [63] Appli-cations handling sensitive data need detailed logging andaudit capability as well as testing and reviewing

Table 3 summarizes the application layer protocols dis-cussed with their weaknesses and differences among them

4 Cross-Layer QoS Strategies

41 Service-Differentiated Real-Time Communication Scheme(SDRCS) -is is an event-driven routing protocol that routesreal-time traffic by featuring a cross-layer packet-forwardingdesign in which it integrates the real-time routing functionalitywith a new priority-based MAC scheme -e protocol ap-proximates distributed packet traversal speed for traffic clas-sification and admission control on the basis of this design Itfurther localizes decision-making by prioritizing packet for-warding to maximize the speed of packets [64]

Advantages

(i) It improves bandwidth utilization bandwidthdegradation due to unschedulable packets can beavoided

(ii) It offers a lightweight packet schedulability estima-tion mechanism Here it uses received signalstrength and admission control and then earlymissed-deadline packet-dropping policies are made

(iii) It delivers high end-to-end throughput via highersource data rates alongside strict end-to-end latencyneeds

(iv) Nodes which meet real-time requirements of lesstraffic load and better channel quality are givenhighest priority when it comes to packet forwarding

-e SDRCS operates via 5 mechanisms namely (1) RSS-(Rich Site SummaryReally Simple Syndication-) basedgrouping (2) admission control (3) prioritized queuing(4) real-time MAC and (5) dynamic forwarding -isprotocol has a queuing policy that employs per-hopdeadline-based queues and the nodes schedule theirpackets using FIFO priority-based queues [1] -e prin-cipal features of this protocol are mentioned as follows

(i) It is event driven and is easily adaptable to anynetwork changes

(ii) It does not need additional hardware for locali-zation or multichannel transmission

(iii) -e cross-layer packet-forwarding strategy can beused when transmitting multimedia traffic

(iv) It is able to efficiently circumvent any voids therein

Early deadline miss (EDM) policy drops all unscheduledpackets -e SDRCS circumvents wrong packet dropsdue to EDM since it adjusts to network changes easily

42 Sensor Fuzzy-Based Image Transport (SUIT) -is pro-tocol regulates congestion using a scheme based on fuzzy

24 Journal of Computer Networks and Communications

logic It transmits packets with reduced quality to allowablelevels if there is congestion enabling it to transmit lower-quality packetsframes without being dropped-e numberof packets delivered per second thus rises and while thisenhances video streaming apps it is more appropriatelymade for those that quickly transmit non-real-time videosuch as in video surveillance SUIT can stream JPEG pic-tures other than the traditional predictive video codingmethods that are not suitable for sensor devices because oftheir constraints of energy memory and computationalspeed Progressive JPEG-PJPEG performs better via packettransmission rate and delay though it degrades the imagequality because of congestion [65] -is protocol employscross-layer methods for interlayer data exchange though itstill employs some layering making it still possible topreserve all advantages of the layered design since it hasapplication and transport layers but has noMAC or routinglayer

For congestion detection and control SUIT bases on thepercentage number of in-bound and out-bound frames perwindow number of contenders and buffer occupancy of thenext-hop node Since buffer management has an effect oneffectiveness and QoS inWMSNs it is useful in the transportlayer so the protocol prioritizes packets for better QoS SUITreorders packets by tracking the sequence numbers of thedelivered packets since they all have a source ID framenumber and sequence number that it uses to detect anymissing packets -e protocol however does not providepacket reliability [65]

Advantages

(i) -e protocol prioritizes packets for better QoS(ii) It can transmit lower-quality packetsframes

without being dropped

Disadvantages

(i) It does not provide packet reliability

43 Network Layer QoS Support Enforced by a Cross-LayerController (NLQS) -is scheme permits packet-level servicedifferentiation as a function of throughput end-to-endpacket error rate and delay [66] (see Figure 8) -is en-hances network layer QoS It comprises a cross-layer controlunit (XLCU) for configuring and controlling networkingfunctionalities at physical MAC and network layers [18]-is depends upon the unified logic which affects choices forapplication layer needs and status of functional blocks whichdo implementation of networking functions [66]

Advantages

(i) Cross-layer interactions can be controlled withoutweakening the upgradability modularity andsimplicity of designing the system

44 Cross-Layer Signaling Shortcuts (CLASS) CLASS has ahigh level of effectiveness flexibility and comprehensive-ness According to [67] it has the following features

(i) Direct signaling among remote layers leading tofaster transmission

(ii) Lightweighted internal message format standard-ized protocols are not suitable for use in internalsignaling since they are not usually lightweight forinstance transmitting against faults in the networkCLASS requires only three fields namely destina-tion address event type and event contents

(iii) -e setup of its exterior messages is standardconsidering external signaling ICMP and TCPIPare suitable for general messages and short notifi-cations respectively

(iv) Message control protocol should ensure optimizedand organized exchange of dense simultaneousmessages across layers for high efficiency andavoidance of possible conflicts -e protocol appliesto many cross-layer signaling situations

Table 3 Differences and problems with application layer protocols

Protocol Architecture Transport QoS options Security Weaknesses

CoAP Requestresponse UDP DTLS No in-built security features DTLS does not supportmulticast

MQTT Publishsubscribe TCP TLSSSL Insufficient security at the protocol level sincepayload values are not encrypted

SMQTT Publishsubscribe TCP CPKP-ABE Key revocation and group pubsub for distributedSMQTT are still a challenge

MQTT-SN Publishsubscribe UDP TLSSSL SSLTLS suffers from attacks like BEAST CRIMERC4 and Heartbleed

XMPP RequestresponsePublishsubscribe TCP X TLSSSL Susceptible to spamming attacks and lacks worldwide

supportAMQP Publishsubscribe TCP TLSSSL Inappropriate for real-time applications

DDS Publishsubscribe TCPUDP TLSSSL Memory intensive and no open-source libraries forconstrained devices

REST Requestresponse HTTP X HTTPS Hard to implement Uses HTTP so incompatiblewith constrained apps

WebSockets PublishsubscribeClientserver TCP X TLSSSL Bad user experience for IoT apps since it runs the

clientserver architectureSTOMP Clientserver TCP HTTP -e broker can act as a bottleneck

Journal of Computer Networks and Communications 25

Advantages

(i) e protocol is scalable and has a very lowpropagation delay

(ii) It is very ecient and exible since it uses directsignaling between any two layers

(iii) It is very comprehensive and eective

45 Cross-Layer and Multipath-Based Video Transmission(CMVT) CMVT is a hybrid of application and networklayers At the application layer it carries out the encoding ofvideo streams to video data frames (I- P- and B-frames) bymeans of the MPEG-4 encoding format At the networklayer route discovery and data transfer take place Twoalgorithms greedy forwarding and rollback are used todiscover a number of paths from the source to the sink nodevia the route discovery method [20] A specied node icalculates the evaluation of its adjacent node j using thefollowing equation

fij (1 minus α)d2(j D) minus d2min(i)d2max(i) minus d

2min(i)

+ αeinit(j) minus eres(j)

einit(j)

(19)

where fij evaluation value for the node i to j d2(j D) isthe distance from the node j to the destination node D

d2min(i) and d2max(i) are the minimum and maximum dis-

tances of neighbors of the node i to D respectively einit(j) isthe initial energy of the node j eres(j) is the current residualenergy of the node j and α is the energy coecient given asfollows

α emax(i) minus emin(i)

emax(i) (20)

where emax(i) and emin(i) are respectively the maximumand minimum energy left of all node irsquos neighbors

e network layer further transmits video streams withCMVT doing status evaluation choosing an appropriatecommunication path for any packets and the QoS guaranteelevel for the path i is calculated as

fi (1 minus ω)hisum hi

+ ωnisum ni

(21)

where fi is the evaluation value for the path i hi are the hopsfor the path i ni is the summation of packets sent via isum ni isthe summation of packets sent by sources and ω is theenergy consumption factor [18]

Advantages

(i) e CMVT protocol is very superior in mediatransmission particularly large-scale WMSNs

To network peers

Flow requirements

-

Atimes

Admissioncontrol

Capacityassignment

Next-hopselection

Coding rate

Data rate

QoScontracts

Cross-layer controlunit (XLCU)

Localization subsystem

Synchronizationsubsystem

Networkcoordinationunit (NCU)

UWB modulator

Channel coder

QoS scheduler

Georouting

Source coder

Rate control

Figure 8 Cross-layer controller architecture

26 Journal of Computer Networks and Communications

46 Cross-Layer Cooperative MAC (CoopMAC)CoopMAC [68] is based on the IEEE 80211 DCF approach-ere is a reduction in interference of neighboring cellsleading to uniform coverage in densely deployed networks[68] A connection is established by two-way authenticationbetween the sender and the receiver When a Clear-to-Send(CTS) packet and short interframe space (SIFS) are obtainedby the sender packets are transmitted right to the receiverwith no valuable cooperation If available the two nodes findout whether they can exchange data -is is done by the useof a helper identification signal to establish how feasiblecooperative communication is It is established in thepresence of a signal and if it is not available direct com-munication is activated For distributed systems helper-initiated cooperation is preferable because of the presence ofthe RTSCTS packets [69]

Advantages

(i) CoopMAC has an advantage over others due to thespatial diversity amongst the three nodes [70]

(ii) It has the potential to achieve significantthroughput and delay with no complexity in thedesign of the system

47 Cooperative MAC Protocol for Multihop Networks (M-CMAC) Just as CoopMAC M-CMAC is made in such away that high data rate stations help low data rate stations toforward traffic for broadcasting [71] -ere are also helperschosen in such a way that 2 fast-hop transmissions replace aslow hop transmission -e helper via whom there existsleast delay from the sender to the receiver is considered bestbecause of theminimum two-hop transmission rate and so isthe chosen neighbor to the duo Here the working as-sumption is that all nodes have their position coordinatesknown and thus the Euclidean distances between all pairscan be calculated and converted to the data rate for that linkAll nodes have cooperative tables (CT) of potential helperswith destination and helper MAC address Euclidean dis-tance and total distance through the helper Any sender withdata to transmit checks out for the existence of any helper inthe CT for the destination and if available forwards RTS forchannel reservation for single-hop duration M-CMAC hasgot an RTS format having a shape with five fields similar tothe one given as follows

Framecontrol Duration Source

addressDestination

addressHelperaddress

-e destination nodersquos address is kept by the source in thepart for the helper address that is in turn saved as the desti-nation address Nodes inspect helper and destination addressfields on receipt of RTS whereby the node acts as the othernodersquos helper in case the helper address field is not the sameand the helper transmits CTS back to the source if at all it wantsto forward data On delivery of the CTS packet the sourcetransmits to the helper which in turn sends the packets to thedestination that also sends ACK to the helper on receipt of thepacket In M-CMAC there exists a higher level of channelreuse (parallel transmissions) due to the increased number of

nodes therein that subsequently raises availability of helpers fordata forwarding -is results in increased throughput incomparison with CoopMAC and IEEE 80211 DCF [71]

Advantages

(i) It promises a higher throughput compared toCoopMAC and IEEE 80211 DCF

48 Cluster-Based Cooperative Routing (CBCR) Protocol-e CBCR protocol has a multihop data-forwarding func-tion realized at the link layer with cooperative links that useM-CMAC -is protocol encompasses two stages namelythe routing relay selection phase and data forwarding phase

(i) Routing relay selection phase all nodes announcetheir presence to their neighbors by broadcastingperiodical beacon messages that carry the MAC ad-dress of the node Each of these constructs a relaytable that contains all neighboring nodes with whichit is able to communicate -e node further broad-casts its neighbor list in case of any change to itsentries since the previous broadcast -e MAC ad-dresses of nodes next to the node X are contained inits relay tablersquos column one and the neighbor nodersquosrow has MAC addresses for neighbors of the adjacentnode Based on its relay table each node in-dependently chooses routing relays Choosing a nodeas a relay node depends on the number of nodes itconnectsmdashit should be the highest number [71]

(ii) Data forwarding phase a node with packets totransmit needs to first verify if the receiver is in thesame cluster and if that receiver has a helper to whichthe packet is thence forwarded but in whose absencethe packet is delivered right to the receiver In casethe intended receiver is in a different cluster therelay table will be checked to see whether that re-ceiver is reachable via other routing relays If it isreachable packets will be sent to the routing relaydirectly or via the helper if at all the relay has one Incase of destination unreachable via relays packetsare multicast to all relays by the node [71]Advantages

(i) It enables multicasting

49 MAC-PHY Cross-Layer Protocol A cooperative cross-layer standard for cooperation at the physical layer in next-generation WMSNs is developed in [70] It provides acomplete MAC layer algorithm that gives a supportive shellfor the PHY-MAC layer Like CoopMAC its scheme de-pends on the middle node for communication betweennodes-eMAC layer protocol has been changed to regulateinformation exchanges at the physical layer -e receivernode receives duplicate packets from source and helpernodes to decode the data [70]

410 MAC-Centric Approach -is cross-layer protocoltargets multimedia applications by using the MPEG-4

Journal of Computer Networks and Communications 27

scheme [21] It is characterized by 4 access categories(AC3ndashAC0) according to their priority in transmission It ismeant for supporting different QoS needs in upcomingvideo applications and enabling MAC layer differentiationfor H264 partitioning [18] Because of low bandwidth delayand other QoS challenges that cause inefficiency in trans-mittingmultimedia data inWMSNs a number of algorithmsdepending on IEEE 80211e have been proposed to supporttransmission of quality videos [72] Selecting an AC dependson the measures for QoS such as the loss rate and delay-us AC3 having the highest priority is mapped to theparameter set concept since the stream is sensitive to bit lossespecially in video transmission [21]

411 Adaptive Cross-Layer Forward Error Correction(ACFEC) In the ACFEC [73] model data packets are ex-changed among nodes through the access point (AP) whichoperates the infrastructure mode in which it adds FEC to videodata [72] -ese data are dealt by encapsulation using astreaming server to the receiver through thewireless AP as RTPpackets An adaptive FEC controller senses the category ofpackets out of the RTP header and recovers the header ofpackets from UDP -e encoder generates some error-cor-recting packets whose number is determined by the sourcepacket number of the block Multimedia transmissions aremonitored by the controller using MAC failure data and itscounter is increased by one if there is a failed transmission-econtroller uses the failure counter to change the packet numberproduced after transmitting a block [18] In case there are lostpackets it changes the redundancy rates and produces extrapackets to replace the lost ones and satisfy the requirements ofthe receiving node [72] -e FEC packet number is enhancedor reduced to satisfy the receiverrsquos requirements and stoppacket losses-is is done through accurate detection of packetlosses and adjustment of redundancy rates If all video datapackets are well received there will not be generation of FECpackets [73]

Advantages

(i) It promises a good QoS through packet loss re-duction and redundancy rate adjustment

412 Balanced Cross-Layer Fuzzy Logic (BCFL) DesignRouting Algorithm in WSNs A new fuzzy logic-basedrouting algorithm (BCFL) [74] was designed using disper-sion of the cross-layer parameter as the fuzzy logic inferencesystem input Each cross-layer parameter has a dynamicweight depending on the value of dispersion -e designcomes with some innovations as follows (i) for fuzzy logicinference system input the absolute parameter value issubstituted by parameter dispersion thereby significantlyreducing algorithmic complexity (ii) dispersion does notchange with the order of magnitude according to the dis-persion formula and (iii) the weight of the parameter de-pends on the size of its dispersion and the two are inverselyproportional to each other -is enables BCFL to have somenobility unlike other algorithms as per the following dis-tinguishing properties making it advantageous

(i) It has got simple if-then rules which remain con-stant even when the constraints increase

(ii) It is capable of dealing with many constraints withno increment in complexity

(iii) It has capacity to yield a more balanced solutionthan other algorithms

(iv) It is easily adaptable to changes in network con-ditions and topology even when the changes arefrequent like in underwater WSNs -e algorithm isuseable in choosing the CH in cluster-based routingprotocols [74]

413 Minimum Hop Disjoint Multipath Routing Algorithmwith Time Slice Load-Balancing Congestion Control SchemeMHDMwTS is a two-phased routing protocol comprisingpath build-up and path acknowledgment phases It consists ofmultiple sources each of which has three build-up disjointpaths namely primary alternate and back-up paths Onactivation the source node requests to build up a route to thenearest hop neighbor which is the path build-up phase [19] Inthis phase step one has source activation in which therequested node for path building adds its number and time-stamp and sends to the least hop-count neighboring node-isgoes on until the least time latency sink having the informationneeded to construct the primary route is reached [75] In steptwo there is path extraction when the new package fromanother path arrives whereby the extracted path is comparedwith the primary path If the node is shared the package will berejected or an alternative path will be searched for to get abackup through comparison of the preceding two paths Phasetwo is path acknowledgment where we have the third step inwhich the sink returns the ACK packet to the sender with pathinformation having nodes and their time information after ithas computed it using the timestamp [1819]

Advantages

(i) -e protocol minimizes end-to-end latency(ii) It enhances congestion control

414 Cross-Layer Optimal Design (CLOD) Authors in [76]come up with a CLOD for scheduling at the data link layerrouting at the network layer and controlling congestion atthe transport layer with an assumption of fixed link capacity-rough congestion control energy efficiency is improvedTransport layer congestion at the nodes is minimized bycompressed sensing (CS) in which transmitted bits are re-duced whereas optimally allocating resources decreasescongestion on the links at the data link layer CLODpromises minimized computational complexity and betterperformance in light traffic scenarios [77] Generating andstoring CLOD is more efficient compared to Gaussianrandom matrices It prolongs network lifetime and savesenergy

Advantages

(i) It does congestion control subsequently increasingenergy efficiency

28 Journal of Computer Networks and Communications

(ii) It reduces computational complexity(iii) It improves performance in light traffic scenarios

415 Challenges and Opportunities of the Cross-Layer Design(CLD)

(i) Different layers in a real network assume differentfunctions andor services A layer only communi-cates with its neighboring layers -e layered modeldegrades system performance because of many

features in wireless transmission QoS parameters inthe layers of WMSNs are shared by routing pro-tocols to optimize the performance Neverthelessthere is a need to design cross-layer models forincreased efficiency in routing [1]

(ii) -e physical layer plays a highly vital role in CLDRate adaptation and channel allocation take place atthe physical layer via signal processing to enhanceQoS End-to-end performance in wireless media isaffected by the changes therein given the effect on

Table 4 A summary of some of the reviewed cross-layer models

Protocol Layers Aims Comments QoS parameters

SDRCS MACPHY Routing real-time traffic Transmits multimedia traffic viacross-layer packet forwarding

-roughputlatency

SUIT Applicationtransport Transmission of non-real-time video Uses fuzzy logic to regulate

congestionTransmission rate

delay

CLASS Any twoDesign serves as a framework fordifferent implementations ofdifferent application scenarios

Scalable very efficient flexible andhas very low propagation delay

Propagation delayjitter

NLQS PHYMACnetwork

Permits packet-level servicedifferentiation as a function of

throughput packet error rate anddelay

Network layer QoS is enhanced -roughput delaypacket error rate

CMVT Applicationnetwork

Encodes video streams to video dataframes via MPEG-4 encoding anddoes route discovery and data

transfer

Greedy forwarding and rollback areused to find source-to-sink paths Energy

CoopMAC PHYMAC Offers spatial diversity among thethree nodes

Helper ID signal finds out howfeasible cooperative communication

is

-roughput ratedelay

M-CMAC PHYMAC Increases end-to-end throughputand packet delivery ratio

Euclidean distances between nodesare calculated and converted to the

data rate for a link

-roughputpacket delivery

ratio

CBCR PHYMACMinimizes control overhead andtime consumed in establishing thecooperative paths than M-CMAC

Energy consumption is moreuniformly distributed in a network

enhancing network lifetime

-roughputpacket deliveryratio energy

MAC-PHY MACPHYMeant to enable PHY layer

cooperation and maximize gains ofcooperation at the MAC layer

Leverages both spatial diversity andcoding gain -roughput delay

MAC-centric MACAPPL

Meant to support QoS needs in newvideo apps and enable MAC layer

differentiation for H264partitioning

Targets multimedia applications byusing the MPEG-4 scheme

Delay packet lossrate

ACFEC MACnetwork(UDP)

Meant to enhance the quality ofvideo streaming over 80211WLANs

and overcome packet losses

Adjusts redundancy rates toovercome channel fluctuations and

detect and reduce packet lossPacket loss rate

MHDMwTS mdash

Meant to provide reliable datatransfer with the multipath routingand load-balancing congestioncontrol method in WMSNs

More reliable than basic routingschemes for transport multimedia

data

Latency packagetransmit rate

BCFL mdash

Introduces dispersion into fuzzylogic-based routing and sets every

cross-layer parameter with adynamic weight

Can be used to select a CH in cluster-based routing protocols andproposes a dispersion formula

Node utilitydispersion

CLODDatalinknetworktransport

Prolongs network lifetime andachieves congestion control anddesigned for lightly loaded WSNs

Assumes fixed link capacity andintegrates compressed sensing

technology

-roughputaverage energy CS

error ratio

Journal of Computer Networks and Communications 29

the purposefulness of the protocols at the networklayer CLD offers solutions to conserving powerminimizing energy and controlling both the flowand congestion in the network as well as fault tol-erance so it is an opportunity for designers con-sidering other layers We further desire to developCLDs that shall make an impact on network oper-ation and get close attention [18]

(iii) CLD interface standardization the architecturemust deliver module functionality although thereexist some queries amongst modules about possibleinterfaces predicted by the necessity to share in-formation among remote protocol layers Technicalchallenges include developing designing andstandardizing of cross-layer interfaces and algo-rithms that satisfy the cross-layer optimization re-quirements among protocol layers

Table 4 summarizes these cross-layer models with a fewcomments

5 Future Research Direction

51 At theMACLayer In WMSNs achieving duty cycling istricky since video traffic is volatile Listening to adjacentnodes wastes energy We need to switch states from awake tosleep -is requires further research

Self-organizing networks and green communicationalthough most CRWSN energy-saving algorithms aredesigned to reduce node transmission power energyconsumption is inevitable during the operational mode ofBSs because of internal processing Scholars are trying withpolicy-makers to encourage a shift to greenWSNs to reduceoperational costs and carbon footprints With self-orga-nizing CRWSNs for automatic switch-off when idle energyis saved which calls for cognitive switching algorithms Sointegrating cognitive switching with radio resource allo-cation in CRWSNs is a promising research area [78]

52 At the Application Layer Considering a WMSN ar-chitecture end users access the network via the BS to andfrom which they send their sensed data We desire to getguaranteed QoS and energy efficiency by designing a newarchitecture in the form of hardware and software whichneeds to be investigated further for a fix of the same [1]

QoS and energy efficiency are very crucial in WMSNsand mostly real-time apps that require guaranteed band-width and throughput in their network lifetime Mostprotocols ignore base station (BS) mobility and WMSNnodes Traffic management telemedicine and battlefieldsurveillance apps require mobile nodes or BSs so designingdynamic routing protocols to be adaptable in these cir-cumstances is necessary

53 Cross-Layer Security of data transmitted on WMSNseg in military surveillance and e-commerce is key withQoS and energy efficiency WMSNs are vulnerable to attacks

like worms and sinkhole Moreover their computationpower is not that high complicating implementation ofstrong secure protocols -is needs further attention

For fairness and priority issues in CRWSNs nodes areassigned diverse priorities depending on importance andurgency of their data For better network performance thereis a need to develop fair resource allocation mechanismswith priority-based fairness in sensors-ere is still potentialin this research area [78]

We must develop multiobjective adaptive protocols tooptimize QoS metrics involved in routing for best trans-mission results to get a trade-off between different opti-mization metrics We need to design standardized cross-layer algorithms which are able to meet cross-layer opti-mization standards

6 Conclusion

We have reviewed the different QoS strategies for WSNs inthe context of IoT from the MAC layer and application layeras well as the cross-layer paradigm For the MAC layer wehave reviewed protocols for WSNs such as the hybrid of thecontention-free and contention-based MAC protocols andthose for WMSNs We further reviewed a number of ap-plication layer protocols including many machine-to-ma-chine requestresponse and publishsubscribe protocols Forsystem optimization cross-layer QoS strategies are veryimportant in wireless communication We have reviewed anumber of cross-layer strategies For all categories reviewedchallenges and opportunities are discussed Finally somepossible future directions are discussed for research andapplication showing a promising potential for future re-search in this relatively new area especially as more WMSNapplications emerge recently

Conflicts of Interest

-e authors hereby declare no conflicts of interest regardingthe publication of this work

Acknowledgments

-is research was funded by the National Natural ScienceFoundation of China (Nos 61602398 61672447 and61711540306)

References

[1] V Bhandary A Malik and S Kumar ldquoRouting in wirelessmultimedia sensor networks a survey of existing protocolsand open research issuesrdquo Journal of Engineering vol 2016Article ID 9608757 27 pages 2016

[2] I F Akyildiz T Melodia and K R Chowdhury ldquoWirelessmultimedia sensor networks applications and testbedsrdquoProceedings of the IEEE vol 96 no 10 pp 1588ndash1605 2008

[3] M Iqbal M Naeem A Ahmed M Awais A Anpalagan andA Ahmad ldquoSwarm intelligence based resource managementfor cooperative cognitive radio network in smart hospitalsrdquoWireless Personal Communications vol 98 no 1 pp 571ndash5922018

30 Journal of Computer Networks and Communications

[4] S A Alvi B Afzal G A Shah L Atzori and W MahmoodldquoInternet of multimedia things vision and challengesrdquo AdHoc Networks vol 33 pp 87ndash111 2015

[5] Q F Hassan A U R Khan and S A Madani Internet ofings Challenges Advances and Applications CRC PressTaylor amp Francis Group A Chapman amp Hall Group BocaRaton FL USA 2018

[6] Q Wang Y Zhao W Wang et al ldquoMultimedia IoT systemsand applicationsrdquo in Proceedings of the 2017 Global Internet ofings Summit (GIoTS) vol 2 Geneva Switzerland June 2017

[7] A Aslam and E Curry ldquoTowards a generalized approach fordeep neural network based event processing for the internet ofmultimedia thingsrdquo IEEE Access vol 6 pp 25573ndash25587 2018

[8] M Jridi T Chapel V Dorez L Bougeant and A Le BotlanldquoSoC-based edge computing gateway in the context of theinternet of multimedia things experimental platformrdquoJournal of Low Power Electronics and Applications vol 8no 1 p 1 2018

[9] H Noura A Chehab L Sleem andM N Rapha ldquoOne roundcipher algorithm for multimedia IoT devicesrdquo MultimediaTools and Applications vol 77 no 14 pp 18383ndash18413 2018

[10] T Balan D Robu and F Sandu ldquoMultihoming for mobileinternet of multimedia thingsrdquo Mobile Information Systemsvol 2017 Article ID 6965028 16 pages 2017

[11] H Shen and G Bai ldquoRouting in wireless multimedia sensornetworks a survey and challenges aheadrdquo Journal of Networkand Computer Applications vol 71 pp 30ndash49 2016

[12] A Ahmad M H Rehmani H Tembine O A Mohammedand A Jamalipour ldquoIEEE access special section Editorial optimization for emerging wireless Networks IoT 5G andsmart grid communication networksrdquo IEEE Access vol 5pp 2096ndash2100 2017

[13] B A Muzakkari M A Mohamed M F A Kadir andZ Mohamad ldquoRecent advances in energy efficient-QoS awareMAC protocols for wireless sensor networksrdquo InternationalJournal of Advanced Computer Research vol 8 no 38pp 212ndash228 2018

[14] M A Yigitel O D Incel and C Ersoy ldquoQoS-aware MACprotocols for wireless sensor networks a surveyrdquo ComputerNetworks vol 55 no 8 pp 1982ndash2004 2011

[15] A M Abbas and O Kure ldquoQuality of Service in mobile ad hocnetworks a surveyrdquo International Journal of Ad Hoc andUbiquitous Computing vol 6 no 2 p 75 2010

[16] M Q Shatnawi ldquoApplication layer protocols for the internetof things a surveyrdquo in Proceedings of the InternationalConference on Engineering amp MIS (ICEMIS) Agadir Mo-rocco September 2016

[17] V Karagiannis P Chatzimisios F Vazquez-gallego andJ Alonso-zarate ldquoA survey on application layer protocols forthe internet of things research motivationrdquo InternationalJournal of Advanced Research in Computer Science vol 8no 3 pp 1ndash10 2015

[18] M S Bernard T Pei Z Li and K Li ldquoQoS strategies forwireless multimedia sensor networks in the context of IoTrdquoLecture Notes of the Institute for Computer Sciences SocialInformatics and Telecommunications Engineering vol 4pp 228ndash253 2019

[19] A AlAmri and M Abdullah ldquoCross-layer quality of serviceprotocols for wireless multimedia sensor networksrdquo in Pro-ceedings of the International Conference on CommunicationManagement and Information Technology pp 649ndash658Cosenza Italy April 2017

[20] J Guo L Sun and R Wang ldquoA cross-layer and multipathbased video transmission scheme for wireless multimedia

sensor networksrdquo Journal of Networks vol 7 no 9pp 1334ndash1340 2012

[21] A Ksentini A Naimi and M Gueroui ldquoToward an im-provement of H264 video transmission over IEEE 80211ethrough a cross-layer architecturerdquo IEEE CommunicationsMagazine vol 44 no 1 pp 107ndash114 2006

[22] W Ye J Heidemann and D Estrin ldquoAn energy-efficientMAC protocol for wireless sensor networksrdquo in Proceedings ofthe 21st International Annual Joint Conference of the IEEEComputer and Communications Societies (INFOCOM 2002)New York NY USA June 2002

[23] J Polastre J Hill and D Culler ldquoVersatile low power mediaaccess for wireless sensor networks categories and subjectdescriptorsrdquo in Proceedings of the Second InternationalConference on Embedded Networked Sensor Systems Sen-Sysrsquo04 pp 95ndash107 Baltimore MA USA November 2004

[24] T Danmanee K N Nakorn and K Rojviboonchai ldquoCU-MAC a duty-cycle MAC protocol for internet of things inwireless sensor networksrdquo ECTI Transactions on ElectricalEngineering Electronics and Communications vol 16 no 2pp 30ndash43 2018

[25] T Van Dam and K Langendoen ldquoAn adaptive energy-efficientMAC protocol for wireless sensor networksrdquo in Proceedings ofthe 1st International Conference on Embedded NetworkedSensor Systems Los Angeles CA USA November 2003

[26] P Xie and J Cui ldquoR-MAC an energy-efficient MAC protocolfor underwater sensor networksrdquo in International Conferenceon Wireless Algorithms Systems and Applications (WASA2007) Chicago IL USA August 2007

[27] A Rahim N Javaid M Aslam U Qasim and Z A KhanldquoAdaptive-reliable medium access control protocol forwireless body area networksrdquo in Proceedings of the AnnualIEEE Communications Society Conference on Sensor Meshand Ad Hoc Communications and Networks (SECON) SeoulKorea June 2012

[28] H S Kwak S Ullah D H Kwak and C H Lee ldquoA powerefficient MAC protocol for wireless body area networksrdquoJournal of the Korean Institute of Intelligent Systems vol 8no 6 pp 131ndash140 2009

[29] C Omeni A J Burdett and M Park ldquoEnergy efficientmedium access protocol for wireless medical body area sensornetworksrdquo in Proceedings of the 4th IEEEEMBS InternationalSummer School and Symposium on Medical Devices andBiosensors pp 29ndash32 Cambridge UK August 2007

[30] L Sitanayah C J Sreenan and K N Brown ldquoER-MAC ahybrid MAC protocol for emergency response wireless sensornetworksrdquo in Proceedings of the Fourth International Con-ference on Sensor Technologies and Applications pp 244ndash249Mestre Italy July 2010

[31] Y Sun and D B Johnson ldquoRI-MAC a receiver-initiatedasynchronous duty cycle MAC protocol for dynamic trafficloads in wireless sensor networksrdquo in Proceedings of the 6thACM conference on Embedded network sensorsystemsmdashSenSysrsquo08 pp 1ndash14 Raleigh NC USA November2008

[32] M Buettner G V Yee E Anderson and R Han ldquoX-MAC ashort preamble MAC protocol for duty-cycled wireless sensornetworksrdquo in Proceedings of the 4th International Conferenceon Embedded Networked Sensor Systems SenSys 2006pp 307ndash320 Boulder Colorado USA October 2006

[33] S Boulfekhar and M Benmohammed ldquoSynchronous receiverinitiated MAC protocol for long-lived sensor networksrdquoComputers amp Electrical Engineering vol 40 no 2 pp 504ndash516 2014

Journal of Computer Networks and Communications 31

[34] S Mehta and K S Kwak ldquoH-MAC a hybrid Mac protocol forwireless sensor networksrdquo International Journal of ComputerNetworks amp Communications vol 2 no 2 pp 108ndash117 2010

[35] I Rhee A Warrier M Aia J Min and M L Sichitiu ldquoZ-MAC a hybrid MAC for wireless sensor networksrdquo IEEEACM Transactions on Networking vol 16 no 3 pp 511ndash5242008

[36] S Arshad A Al-sadi and A Barnawi ldquoZ-MAC performanceevaluation and enhancementsrdquo Procedia Computer Sciencevol 21 pp 485ndash490 2013

[37] Y Liu I Elhanany and H Qi ldquoAn energy-efficient QoS-aware media access control protocol for wireless sensornetworksrdquo in Proceedings of the IEEE International Confer-ence on Mobile ad hoc and Sensor Systems pp 9ndash11 Wash-ington DC USA December 2005

[38] J Decotignie ldquoWiseMAC an ultra low power MAC protocolfor the downlink of infrastructure wireless sensor networksrdquoin Proceedings of the Ninth International Symposium onComputers And Communications pp 244ndash251 AlexandriaEgypt June 2004

[39] R A Rashid ldquoDevelopment of energy aware TDMA-basedMAC protocol for wireless sensor network systemrdquo EuropeanJournal of Scientific Research vol 30 no 4 pp 571ndash578 2009

[40] H P Van Hoesel ldquoA lightweight medium access protocol(LMAC) for wireless sensor networksrdquo in Proceedings of theInternational Workshop on Networked Sensing Systems Bal-timore MD USA November 2004

[41] C Li P Wang H Chen andM Guizani ldquoA cluster based on-demand multi-channel MAC protocol for wireless multi-media sensor networksrdquo in Proceedings of the IEEE In-ternational Conference on Communications pp 2371ndash2376Beijing China May 2008

[42] J L Sobrinho and A S Krishnakumar ldquoQuality-of-service inad hoc carrier sense multiple access wireless networksrdquo IEEEJournal on Selected Areas in Communications vol 17 no 8pp 1353ndash1368 1999

[43] J P John ldquoProviding MAC QoS for multimedia traffic in80211e based multi-hop ad hoc wireless networksrdquo ComputerNetworks vol 51 no 1 pp 153ndash176 2007

[44] S Lofty and N Padmavati ldquoA survey on mobility basedprotocols in WSNsrdquo in Proceedings of the InternationalConference Communication and Manufacturing pp 12ndash15Hangzhou China July 2014

[45] S Deng L Shen and J Li ldquoMobility-based clustering pro-tocol for wireless sensor networks with mobile nodesrdquo IETWireless Sensor Systems vol 1 no 1 pp 39ndash47 2011

[46] M Khalil A Khalid F U Khan and A Shabbir ldquoA review ofrouting protocol selection for wireless sensor networks insmart citiesrdquo in Proceedings of the 24th Asia-Pacific Confer-ence on Communications (APCC) Ningbo China November2014

[47] O Message Q Telemetry and T Mqtt Committee Specifi-cation Draft 02 April 2014

[48] M Collina M Bartolucci A Vanelli-coralli andG E Corazza ldquoInternet of things application layer protocolanalysis over error and delay prone linksrdquo in Proceedings ofthe 7th Advanced Satellite Multimedia Systems Conference andthe 13th Signal Processing for Space CommunicationsWorkshop Livorno Italy September 2014

[49] A Talaminos-barroso M A Estudillo-valderrama L M RoaJ Reina-tosina and F Ortega-ruiz ldquoA machine-to-machineprotocol benchmark for e health applicationsmdashuse case re-spiratory rehabilitationrdquo Computer Methods and Programs inBiomedicine vol 129 pp 1ndash11 2016

[50] V Goyal O Pandey A Sahai and B Waters ldquoAttribute-based encryption for fine-grained access control of encrypteddatardquo in Proceedings of the 13th ACM Conference on Com-puter and Communications Security pp 89ndash98 AlexandriaVA USA October 2006

[51] M Singh M A Rajan V L Shivraj and P BalamuralidharldquoSecure MQTT for internet of things (IoT)rdquo in Proceedings ofthe Fifth International Conference on Communication Systemsand Network Technologies Gwalior India April 2015

[52] A Stanford-Clark and H L Truong MQTT for Sensor Net-works (MQTT-SN) Protocol Specification httpmqttorgnewwp-contentuploads200906MQTT-SN_spec_v12pdf 2013

[53] RFC ldquoExtensible messaging and presence protocol (XMPP)rdquoin Proceedings of the IETF pp 1ndash9 San Diego CA USAAugust 2005

[54] S Vinoski and I Technologies ldquoQueuing protocolrdquo IEEEInternet Computing vol 10 no 6 pp 87ndash89 2006

[55] J L Fernandes I C Lopes J J P C Rodrigues and S UllahldquoPerformance evaluation of restful web services and AMQPprotocolrdquo in Proceedings of the Fifth International Conferenceon Ubiquitous and Future Networks (ICUFN) pp 2ndash7 DaNang Vietnam July 2013

[56] M Asim ldquoA survey on application layer protocols for internetof things (IoT)rdquo International Journal of Advanced Researchin Computer Science vol 8 no 3 pp 996ndash1000 2017

[57] K An A Gokhale D Schmidt S Tambe P Pazandak andG Pardo-castellote ldquoContent-based filtering discovery pro-tocol (CFDP) scalable and efficient OMG DDS discoveryprotocolrdquo in Proceedings of the 8th ACM InternationalConference on Distributed Event-Based Systems New YorkNY USA July 2014

[58] E Web What is a web service pp 52ndash57 2010[59] M Laine RESTful Web Services for the Internet of ings

pp 2ndash4 2012 httpmediaTkk[60] IETF -e WebSocket protocol pp 1ndash71 2011[61] V Pimentel ldquoCommunicating and displaying real-time data

with websocketrdquo IEEE Internet Computing vol 16 no 4pp 45ndash53 2012

[62] Wikipedia 2019 httpsenwikipediaorgwikiStreaming_Text_Oriented_Messaging_Protocol2019

[63] K Raghavendra and S Nireshwalya ldquoApplication layer se-curity issues and its solutionsrdquo IJCSET vol 2 no 6pp 1266ndash1269 2012

[64] Y Xue B Ramamurthy andM C Vuran ldquoSDRCS a service-differentiated real-time communication scheme for eventsensing in wireless sensor networksrdquo Computer Networksvol 55 no 15 pp 3287ndash3302 2011

[65] C Sonmez S Isik M Y Donmez O D Incel and C ErsoyldquoSUIT a cross layer image transport protocol with fuzzy logicbased congestion control for wireless multimedia sensornetworksrdquo in Proceedings of the 5th International Conferenceon New Technologies Mobility and Security (NTMS) pp 1ndash6IEEE Istanbul Turkey May 2012

[66] T Melodia and I Akyildiz ldquoCross-layer QoS-aware com-munication for ultra wide band wireless multimedia sensornetworksrdquo IEEE Journal on Selected Areas in Communica-tions vol 28 no 5 pp 653ndash663 2010

[67] Q Wang and M A Abu-Rgheff ldquoCross-layer signalling fornext-generation wireless systemsrdquo in Proceedings of the 2003IEEE Wireless Communications and Networking 2003WCNC 2003 vol 2 pp 1084ndash1089 IEEE New Orleans LAUSA March 2003

[68] P Liu Z Tao S Narayanan T Korakis and S PanwarldquoCoopMAC a cooperative MAC for wireless LANsrdquo IEEE

32 Journal of Computer Networks and Communications

Journal on Selected Areas in Communications vol 25 no 2pp 340ndash354 2007

[69] H Shan H T Cheng and W Zhuang ldquoCross-layer co-operative MAC protocol in distributed wireless networksrdquoIEEE Transactions onWireless Communications vol 10 no 8pp 2603ndash2615 2011

[70] F Liu T Korakis Z Tao and S Panwar ldquoAMAC-PHY cross-layer protocol for wireless Ad-Hoc networksrdquo in Proceedingsof the 2008 IEEE Wireless Communications and NetworkingConference pp 1792ndash1797 IEEE Las Vegas NV USA March2008

[71] L Jacob and H R Shamna ldquoEfficient cooperative MAC androuting in wireless networksrdquo Transactions on Networks andCommunications vol 3 no 5 2015

[72] S Rao and K Shama ldquoCross layer protocols for multimediatransmission in wireless networksrdquo International Journal ofComputer Science amp Engineering Survey vol 3 no 3pp 15ndash28 2012

[73] L Han S Park S Kang and H Peter ldquoAn adaptive cross-layer FEC mechanism for video transmission over 80211WLANsrdquo in Proceedings of the 2009 International Conferenceon Internet pp 209ndash215 Las Vegas NV USA December2009

[74] N Li J Martınez and V H Dıaz ldquo-e balanced cross-layerdesign routing algorithm in wireless sensor networks usingfuzzy logicrdquo Sensors vol 15 no 8 pp 19541ndash19559 2015

[75] G Sun J Qi Z Zang and Q Xu ldquoA reliable multipathrouting algorithm with related congestion control scheme inwireless multimedia sensor networksrdquo in Proceedings of the2011 3rd International Conference on Computer Research andDevelopment vol 4 pp 229ndash233 IEEE Shanghai ChinaMarch 2011

[76] M Li Y Jing and C Li ldquoA robust and efficient cross-layeroptimal design in wireless sensor networksrdquoWireless PersonalCommunications vol 72 no 4 pp 1889ndash1902 2013

[77] J Yan M Zhou and Z Ding ldquoRecent advances in energy-efficient routing protocols for wireless sensor networks areviewrdquo IEEE Access vol 4 pp 5673ndash5686 2016

[78] A Ahmad S Ahmad M H Rehmani and N U Hassan ldquoAsurvey on radio resource allocation in cognitive radio sensornetworksrdquo IEEE Communications Surveys amp Tutorials vol 17no 2 pp 888ndash917 2015

Journal of Computer Networks and Communications 33

International Journal of

AerospaceEngineeringHindawiwwwhindawicom Volume 2018

RoboticsJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Active and Passive Electronic Components

VLSI Design

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Shock and Vibration

Hindawiwwwhindawicom Volume 2018

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawiwwwhindawicom

Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Control Scienceand Engineering

Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Journal ofEngineeringVolume 2018

SensorsJournal of

Hindawiwwwhindawicom Volume 2018

International Journal of

RotatingMachinery

Hindawiwwwhindawicom Volume 2018

Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Navigation and Observation

International Journal of

Hindawi

wwwhindawicom Volume 2018

Advances in

Multimedia

Submit your manuscripts atwwwhindawicom

Page 5: ReviewArticle - downloads.hindawi.com · ReviewArticle QoSStrategiesforWirelessMultimediaSensorNetworksinthe ContextofIoTattheMACLayer,ApplicationLayer,and Cross-LayerAlgorithms MuwongeSsajjabbiBernard

cross-layer protocols and leave out the MAC layer andapplication layer protocols and also some cross-layer pro-tocols are left out -e surveys that have been reviewed arecompared in Table 1

-emajor contributions that make our survey importantare outlined below

15 Importance of the Survey

(i) We discuss and draw a comparison between the IoTand the IoMT by discussing the differences betweenthe two

(ii) We further discuss the newer paradigm of WMSNsand compareWSNs andWMSNs in our discussion

(iii) We make a comprehensive survey of the recentstate-of-the-art routing protocols focusing on theMAC layer the application layer and the cross-layer paradigm

(iv) We discuss the WMSN protocols as well as theWMSN-enabled protocols that have been de-veloped in the recent studies by several scholars

(v) In some of the reviewed protocols we talk about therelative advantages and disadvantages of some ofthe routing approaches for readers to be in positionto comprehend different techniques and so choosethe most appropriate technique depending on theuser requirements

(vi) In all these we cover the challenges and oppor-tunities existing for the discussed categories

(vii) We consider the possible future research trendsbefore concluding this paper

16 Challenges

(i) Since WSNs and WMSNs are usually distributed andad hoc networks their nodes are powered by batteriesmaking energy optimization a challenge that needs tobe addressed when designing their routing protocols

(ii) Considering WSNs their routing protocols do notput into consideration multimedia applicationswhich need a lot of bandwidth and processingenergy and must be transmitted in real time withutmost fidelity Extra challenges in routing areimposed in WMSNs in deployment of heteroge-neous sensors since there could be audio video andstill pictures in addition to scalar data duringtransmission Besides the heterogeneous datathereby handled all have different QoS re-quirements and with many different businessneeds different services with varying requirementsgive rise to a big challenge for the routing design

(iii) WMSNs have also got nonrechargeable batteriesplacing a limitation on the energy of the nodes yetmultimedia applications consume a lot of energyQuite a number of complications and limitingfactors exist practically despite the fact that sensorscan be self-powering as enabled by energy

harvesting Predictive video encoding using MPEG-4 and other standards is also an energy-consumingprocess and so causes enormous degradation inrouting efficiency complicating the design ofWMSN protocols even more

(iv) -e heterogeneous nature of the network is anotherchallenge since there is a difference in the categoriesof nodes needed for communication so as to enablethe facilitation of effective data collection andprocessing as well as efficient transmission It is noteasy to have a uniform communication protocolplatform because of different functionalities unlikethe case of the conventional WSNs compared to theWMSNs

(v) -ere must be a trade-off between the energy effi-ciency and the multimedia QoS when deciding onthe route taken Data aggregation and compressionrouting protocols may be applicable for energy-saving though they can lead to intolerable delays inWMSNs Furthermore there might be networkcongestion due to the many-to-many and mutualinterference in wireless routes and the scarceWMSN resources In case a given node gets so manyhigh-rate streams it lowers the performance of thenetwork and raises the possibility of node failuredue to energy drainage

-is papermainly discusses QoS strategies forWSNs andWMSNs in the context of IoT from the MAC layer appli-cation layer and cross-layer perspective -is paper presentsa state-of-the-art survey on routing in WSNs and WMSNs-e rest of this paper is organized as follows In Section 2 wediscuss the QoS strategies at the MAC layer In Section 3 wediscuss the QoS strategies at the application layer Section 4discusses the cross-layer QoS strategies and in Section 5 wegive some future research directions before concluding thispaper in Section 6

2 QoS Strategies at the MAC Layer

-e medium access control (MAC) protocol is meant toregulate admittance to the shared medium as well astransmission reliability and efficiency through error cor-rection in wireless transmissions among others It is alsoresponsible for performance of framing addressing andflow control -e MAC protocol design affects energy effi-ciency too -is is majorly categorized into contention-freeprotocols contention-based protocols and a hybrid of these

21 MAC Protocols forWSNs Hybrid MAC protocols thesecombine contention-free and contention-based MAC pro-tocols on whose advantages they capitalize Some of these arediscussed below

211 S-MAC (Sensor MAC) -is is a MAC protocoldesigned for WSNs -e main components of S-MAC in-clude (i) periodic listen and sleep (ii) collision and over-hearing avoidance and (iii) message passing Energy

Journal of Computer Networks and Communications 5

consumption and self-configuration are controlled usingthese three techniques Nodes periodically go into the sleepmode to control the energy consumed in listening to idlechannels and virtual clusters are formed by neighbor nodesfor sleep schedule autosynchronization -e protocol em-ploys in-channel signaling when the radio is in the sleepmode and message passing is applied to decrease contentionlatency for store-and-forwardWSN applications during datatransmission [22] For adjacent nodes to solve their clockdrift S-MAC requires periodic synchronization To reducesynchronization errors exchanged timestamps are relativenot absolute

Advantages

(i) S-MAC has a longer listening period than the clockerror

(ii) It has more capacity to conserve energy than IEEE80211

(iii) It can do trade-off between energy and latencydepending on traffic conditions [22]

212 B-MAC (Berkeley MAC) -is is a carrier-sense MACprotocol designed for WSNs featuring a simplistic scalableimplementation that does network variations and is effectivein channel valuation [23] -e protocol has an interface withsome flexibility leading to very low-power usage and con-trols collisions through collision avoidance mechanismsleading to effective use of the frequency band It uses asampling scheme that is adjustable to minimize duty cyclingand idle listening so as to reduce power wastage Nodes arescheduled with their duty cycle each within which they sendthe preamble to the channel in case of availability of data fortransmission In different duty-cycle schedules channelusage is checked by the destination nodes that is if thepreamble found is long it will remain powered on until thedestination address is obtained and in case it is the target itwill either await senderrsquos data or go to sleep-e protocol haslower latency than synchronous MAC protocols but has aproblem of overhearing where the neighbor node getslengthy preambles though not the intended receiver hencewasting a lot of power in doing so [24] It outperforms anumber of protocols via reconfiguration feedback andbidirectional interfaces for upper layer services It can run at

low duty cycles and its apps cannot experience synchro-nization overhead [23]

Advantages

(i) -e protocol features a simplistic scalableimplementation that does network variations andis effective in channel valuation

(ii) It has an interface with flexibility leading to verylow-power usage

(iii) It controls collisions through collision avoidancemechanisms leading to effective use of the fre-quency band

(iv) It uses an adjustable sampling scheme to minimizeduty cycling and idle listening reducing powerwastage

Disadvantages

(i) It has a problem of overhearing where the neighbornode gets lengthy preambles though not theintended receiver hence wasting a lot of power indoing so

213 T-MAC -is is a contention-based MAC protocol forWSNs that dynamically with little difficulty via fine-grainedtimeouts adapts a listensleep duty cycle It uses this dutycycle by ending its active part so as to vary load in time andposition thereby decreasing the energy wastage on idlelistening [25] When starting an active period it uses a shortlistening window thereby improving the use of energy byS-MAC -rough adaptive duty cycling T-MAC savespower but at a cost of less throughput and enhanced delayBoth S-MAC and T-MAC under similar conditions performin the same way though in variable workloads S-MAC uses 5times more power than T-MAC But T-MAC is complex andnot scalable If its active window is decreased its snoopingcapability on nearby traffic and adaptation to variableconditions of the network is reduced [23] All nodes in thisprotocol are configured to wake up periodically listen toadjacent nodes and go back into the sleep mode till the nextframe as more messages come into the queue Collisionavoidance and reliable communication are ensured throughRequest-to-Send (RTS) Clear-to-Send (CTS) and Ac-knowledgment (ACK) messages among nodes [25]

Table 1 Comparison of this survey with existing surveys

Author(s) Year CharacteristicsBhandary et al [1] 2016 Routing strategies in WMSNs properties and limitationsMuzakkari et al [13] 2018 WSN contention-based scheduling-based and hybridMAC protocolsAhmad et al [12] 2011 QoS-aware MAC protocols for WSNsMuzakkari et al [13] 2010 QoS provisioning at the levels of routing MAC layer and cross-layerShatnawi [14] 2016 IoT application layer protocolsKaragiannis et al [15] 2015 Application layer protocols for the Internet of thingsBernard et al [16] 2019 Cross-layer QoS strategies for the IoT in WMSNsAlAmri and Abdullah [17] 2017 Cross-layer QoS protocols for WMSNsGuo et al [20] 2012 Cross-layer and multipath-based video transmission for WMSNs

Ksentini et al [21] 2006 Toward an improvement of H264 video transmission over IEEE80211e through a cross-layer architecture for WMSNs

6 Journal of Computer Networks and Communications

T-MAC design the major forms of energy consump-tion in T-MAC are through (i) idle listening (ii) col-lisions (iii) protocol overhead and (iv) overhearing Incontrast in the active period the node remains lis-tening and sending data If no activation event occursfor a certain time t then the active period ends -eactivation event might be data reception on a radiocommunication sensing on a radio and end oftransmission for the data packet of the node or ACKknowing that the neighbor has finished transmitting

Clustering and synchronization on waking up a nodebegins by listening to the channel If after a certain timethere is nothing heard it chooses a schedule fortransmitting Synchronization (SYNC) packetsappended with the time of the next frame startup Ifduring this time it hears another nodersquos SYNC packetit follows its schedule subsequently transmitting itsown SYNC Following both schedules enables it to havean activation incident at the beginning of both frames

RTS and t choice if a node does not get a CTS messageafter sending an RTS then either (i) there is a collisionpreventing the destination node from hearing the RTS(ii) RTS or CTS is overheard and destination node barredfrom responding or (iii) the destination node is in thesleep mode So if after time t there is no answer thesource node switches to the sleepmode It should only doso if none of its neighbors is still communicating A newinterval t can be triggered by a neighborrsquos RTS or CTSreception A node may be out of range to hear theinitiating RTS-e length t of the interval should be longenough to get the CTS packetrsquos beginning [25] and t isthus given by the comparison in the following equation

tgtC + R + T (1)

where C contention interval length RRTS packet lengthand T turn-around time

Advantages

(i) It saves power through adaptive duty cycling

Disadvantages

(i) It is unscalable and more complex(ii) It offers less throughput and enhanced delay

214 CU-MAC -is protocol supports IoT standards re-quiring request-response transmissions It employs a multi-channel mechanism for constant bidirectional transmissionof packets at low duty cycle to resolve hidden-terminal issuesAccording to Danmanee et al [24] CU-MAC is an SI-MACprotocol that is aimed at improving bandwidth of the channelby initiating connections and transmitting data throughdifferent channels using the multichannel approach

Protocol design the protocol has the following threeproperties that make it perform more efficiently withWSNs in the IoT

(i) It enables multichannel transmission using 9useable channels in IEEE 802154

(ii) It can enable linked data transmission using oneconnection which reduces additional overheadand delay

(iii) It can support continuous transmission of packets-is is enabled depending on how the bufferoverflow protection policy is set in the buffer stack-is stack has 2 parts (a) bidirectional buffer slotsfor enabling sensor nodes to support linked datatransmission and (b) normal buffer slots fornormal data queuing Buffer stack policies are setas follows

(a) A packet is only dropped in the buffer when itexpires

(b) Unless the packet is from a target node to which areceiver is sending the data the receiver deniespacket reception when buffer slots fill up

(c) -e bidirectional buffer slot is used for bi-directional buffer transfer

CU-MAC can reduce power wastage by choosing toemploy only a single channel for transmitting both controlpackets and data in case there is only one sender in the sameregion of coverage It can further enable multiple trans-missions through another usable channel if there are othersending nodes in this area [24] -e protocol has two phases(i) advertising phase in which a source node announces toneighbors before transmission and (ii) data transfer phasewhich commences when the source node receives a Ready-to-Receive Information packet from the destination node

Advantages

(i) It has a higher packet delivery ratio compared toother MAC protocols

(ii) It performs well in high traffic situations since itsduty-cycle is lower

(iii) It reduces power wastage by employing only asingle channel for transmitting both controlpackets and data

(iv) It can enable multiple transmissions through an-other usable channel if there are other sendingnodes

215 R-MAC -is is a reservation-based MAC protocoldesigned majorly with an aim to be fair and efficient in termsof energy use To eliminate collision of data packets R-MAC[26] schedules control and data packet transmissions at thesource and destination nodes rather than using RTSCTSmessage exchanges Channel utilization in R-MAC is en-hanced by an ARQ method called burst-based acknowl-edgment which together with other scheduling algorithmsaddresses the exposed terminal problem besides enhancingthe network throughput

R-MAC design all nodes in this protocol intermittentlyfunction in listen and sleep states so as to avoid wastingenergy in overhearing and idle modes -e time takento listensleep is the same for all nodes each of which

Journal of Computer Networks and Communications 7

chooses a schedule for itself rendering central sched-uling and synchronization irrelevant So in the absenceof traffic in the locality the node simply does periodiclistening and sleeping and to eliminate collisionstransmissions are distributively synchronized using areservation-based model in case there is a sender totransmit data R-MAC differs from S-MAC in that ittransmits data to a number of nodes using the sleepmode [24] -ere are 3 stages in this protocol (i) la-tency detection (ii) period announcement and (iii)periodic operation of which (i) and (ii) are for nodesynchronization and (iii) is for listensleep procedures

(i) Latency detection phase here all nodes broadcast aNeighbor Discovery control packet (ND) at an ar-bitrarily chosen time since they are powered onWhen the ND is received from the neighbor nodethe receiver notes its arrival time and chooses arandom time to send an acknowledgment (ACK-ND)mdashthe same as the receivedND in sizemdashinwhichthe duration from ND arrival to ACK-ND trans-mission I2 is specified and the time from ND packettransmission to ACK-ND arrival I1 is computed

(a) Propagation latency between the 2 nodesL (I1 minus I2)2

(b) -us propagation latency can be defined as ldquotheinterval from the time the first node sends the firstbit of a packet to the time the second node receivesthe last bit of the packetrdquo [24]

(ii) Period announcement phase here nodes choosetheir own random listensleep time and start timeand broadcast these and when packets are re-ceived the node turns the received schedule intoits own All nodes will record their neighborsrsquoschedules in relation to theirs -ese two phasesmake a number of rounds to ensure all nodes areinformed about their neighboring nodes

(iii) Periodic operation phase here there is datatransmission [26] and nodes intermittently go intothe sleep mode and wake up One period is a listensleep cycle and nodes possess similar periods Twoparts make a period T0 listen window TL and sleepwindow TS as shown in the following equation

T0 TL + TS (2)

In this protocol the nodes make use of control packetsfor communication-ese include REV (reservation packet)ACKREV (acknowledgment packet for REV) and DATAACK-DATA (acknowledgment packet for data) messageexchanges all of which are similar in size

Advantages

(i) It is fair and efficient in terms of energy use(ii) It eliminates collision of data packets

216 AR-MAC (Adaptive-Reliable Media Access Control)-is is a TDMA-based MAC protocol for wireless body areanetworks (WBANs) designed to reduce energy consumption

as proposed by Rahim et al [27] -e protocol allocates aguaranteed time slot (GTS) to all nodes for transmissiondepending on their needs It utilizes periodic sleep andwakeup to minimize overhearing and idle listening based onthe needs of the node It has a central node (CN) having bigbatteries with higher computational power and single ordouble transceivers For double transceivers the sum of thetime frame TFrame is assigned for node transmission But inAR-MAC a single transceiver CN is considered with TFramehaving 3 subdivisions contention-free period (CFP) forsensor connection contention access period (CAP) foremergencies and time TMS to communicate with themonitoring station (MS)

Channel selection the CN first scans for redundantchannels from which it chooses one for transmissionand broadcasts its address and other information to thenodes -e destination nodes search for channels byscanning the radio frequency (RF) channels If free itswitches to the next and if busy it will wait for time TCPlistening to packets and again switch to the subsequentchannel if it does not get the channel packets When thenode finally succeeds in getting a channel packettransmission commences and an ACK packet is sent tothe CNTime slot assignment a Time Slot Request (TSR) packetis sent to the CN by the sensor node after it has selectedthe RF channel -e TSR packet has data rate and timeslot info for the node Static-size time slots with staticguard-band time TGB are proposed in [28] -isprotocol employs the adaptive algorithm of time slot(TS) and GB time and the CN allocates TS whilesending Time Slot Request Reply (TSRR) based on thenode traffic pattern -e size of the time slots variesdepending on the node requirements Based on thetransmissionmodel the data packet transmission ACKpacket reception and some amount of delay can betolerated by allocated TS To prevent interference be-tween adjacent time slots resulting from the node andCN clock drift TGB is inserted between them and itsvalue is given as follows

TGB1

FxTS1100

TGBn

FxTSn

100

TGBnn+1

F

100X12

TSn + TSn+11113858 1113859

(3)

where FGB factor depending on the mean driftvalue Nodes go to sleep after slots have been suc-cessfully assigned and wake up to transmit in assignedslots thus saving energy wasted in idle listening sinceeven allocated slots are collision freeSynchronization for periodic synchronization TDMAneeds additional energy [29] It needs a lot of energy tosynchronize nodes after several cycles To reduce en-ergy wastage and control collisions AR-MAC employs

8 Journal of Computer Networks and Communications

a new synchronization scheme When a data packetarrives the current and expected packet arrival timeswith allowable delay (D) are compared by the CN and adrift value (DV) is computed based on the differenceΔT -e DV is sent to the node in the ACK packet forchanging the slot in subsequent transmissions thoughthe value is based on allowable delay and F In caseΔTgtD the CN will send the DV in the SYNC-ACKpacket for subsequent SYNC to nodes or else a mereACK packet is sent by the CN for the received packetFor any forthcoming data transmissions the node willchange its wake-up time plan based on the DV Withsuch a sync approach a sleeping node will not lose syncfor all cycles N Acceptable delay (D) is given as

D min TS1 TSn( 1113857XF

100 (4)

where F is the guard-band factor

ΔT expected arrival time minus current arrival time

DV 0 if |ΔT|ltD

ΔT if |ΔT|gtD1113896

(5)

Frame format the AR-MAC protocol employs twoforms of packets namely (i) data packets in which anode transmits in its assigned time slot and employsCAP for emergency data and (ii) control packets thatinclude channel packet Time Slot Request (TSR)packet Time Slot Request Reply (TSRR) packet Syn-chronization-Acknowledgment (SYNC-ACK) packetData Request (DR) packet and Acknowledgment(ACK) packet [27]Energy consumption if N is the number of cyclesenergy consumption for these is measured as follows

ETotal 1113944N

k1ESleepk

+ 1113944N

k1EActivek

ESleep TSleep x ISleep x V

TSleep Tframe minus Tactive

(6)

where ISleep current from the voltage source V whenthe node is asleep If Esw is the switching energy Etransthe transmission energy Erec the receiving energy andETout the time-out energy then

EActive 2Esw + Etrans + Erec + ETout (7)

Advantages

(i) It reduces energy consumption(ii) It minimizes overhearing and idle listening based

on the needs of the node via periodic sleep andwakeup

217 ER-MAC It is a combination of TDMA and CSMAmethods making it flexible and adaptable to topological andtraffic changes and hence applicable in rapid response

situations inWSNs Collision-free slots can be planned usingTDMA [30] ER-MAC is designed in such a way that thefollowing are met

(a) It has a higher ratio of packet delivery and low delayunder lower energy consumption levels out-performing Z-MAC

(b) It keeps two priority queues in order to distinguishhigh from least priority packets

(c) It permits contention in TDMA slots so as to handleenormous traffic volumes

(d) It has a harmonized slot shape and the nodes areable to locally change their plans enabling themconnect to or exit the network with easeProtocol design it discovers the topology in such away that it uses the simple flooding mechanism tobuild the tree as initiated by the base station (BS) -etopology discovery aims at neighbor discovery andchange tracking in addition to setting up a routingtree A TOPOLOGY_DISCOVERY message com-prising hop_count new_parent_id and old_-parent_id is generated by the BS A node broadcasts itto discover if it has any potential children and aresponse to the parent In case of the desire to changethe parent the former parent is notified For all nodesthe hop count to the BS parent ID children and one-hop neighbor are registered with the BS at this stageTDMA slot assignment when a BS sends a SYNCmessage it switches to the TDMAmode which whenreceived by the child uses the broadcast slot for itschildrenrsquos synchronization Neighboring nodescannot share a slot since slots are assigned andschedules are exchanged accordingly Slot assign-ment begins with a leaf node in a bottom-up modesuch that the flow of messages to the BS may besmooth Before getting the child-node schedules anonleaf node will not transmit its data via a unicastslot descendantrsquos data via many unicast slots andchild synchronization via a broadcast slot Childnodes send SCHEDULE_NOTIFICATION mes-sages to the BS to end this phase

Local time synchronization is done via parent-childbroadcast sync since they both need to operate the same clockso that when one is transmitting the other is ready to receivePacket prioritization happens in such a way that a highpriority queue is first emptied before a lower priority queue isallowed to transmit In case a queue is filled up the shortestslack packet is dropped and then fairness over source isconsidered as queue modification is done to enable the BSbalance info from different nodes Energy conservation is bysender turning off the radio in absence of any info to transmitand the receiver will switch to the sleep mode in case nopackets are received after a certain timeout [30]

Advantages

(i) It is flexible and adaptable to topological and trafficchanges

Journal of Computer Networks and Communications 9

(ii) It is applicable in rapid response situations inWSNs

(iii) It has a higher ratio of packet delivery and lowdelay under lower energy consumption levels

218 RI-MAC (Receiver-Initiated MAC) -is is an asyn-chronous duty-cycle protocol that tries to reduce timeoccupied by the communicating nodes on the medium toagree on time for data exchange-e sending node waits forthe receiving node to signify the beginning of the com-munication session through a short beacon frame that itsends -e channel is only occupied by the beacon and datatransmission reducing channel occupancy and thus en-abling data exchange for other nodes too -is implies animprovement in channel utilization that subsequently leadsto increment in throughput packet delivery ratio andpower efficiency over a wide-ranging traffic load [31] Beinga receiver-initiated model it ensures substantial reductionin overhearing and lowers the probability of colliding andthe cost of recovery compared to B-MAC [23] or X-MAC[32] RI-MAC is further capable of using a beacon withACK and Ready-to-Receive packets though it has aproblem if a sending node with Ready packets needs to havethe radio on till a ready-to-receive node is awake [13] whichthen sends it a beacon and transmission begins immediatelyand later is acknowledged by another beacon Nodes in-termittently wake up to listen to any possible frames meantfor them depending on their schedules A node announcesits readiness to receive a frame through a broadcast beaconon waking up when it switches on the radio When a dataframe has been received the receiver will spend some moretime known as ldquodwell timerdquo in the active state so as toreceive queued packets which is the time in RI-MAC thatdepends on the number of contending senders If there areseveral contending senders that might not have beenpredicted there is a challenge in trying to reduce the re-ceiverrsquos active time to optimize power efficiency and reducethe cost of collision detection and lost data recovery even inthe case of hidden senders For this protocol this can beaddressed by a beacon frame from the receiving node forcoordinating data frame transmissions of contendingsending nodes-e protocol also beats B-MAC and X-MACin minimizing the cost of collision detection and datarecovery through coordination of the receiversrsquo dataframes -is is because the receiving node is aware of themaximum delay prior to the arrival of the frame-e abilityto broadcast is supported by RI-MAC when it sends data asunicast to each senderrsquos neighboring node or continuouslytransmits the frame in a back-to-back mode for as long asthe sleep interval [31]

Advantages

(i) It improves channel utilization leading to in-crement in throughput packet delivery ratio andpower efficiency over a wide-ranging traffic load

(ii) It ensures substantial reduction in overhearing andlowers the probability of colliding and the cost ofrecovery

219 SRI-MAC (Synchronous Receiver-Initiated MAC)-is is a synchronous duty-cycle protocol that implementsthe method of receiver-initiated data transmission with anaim of enhancing network lifetime through energy wastageavoidance via collision overhearing and idle listening To dothis and also minimize duty cycle it uses a sequence ofadaptive beacons and RTSCTS packets In this protocolsending nodes do not go to sleep till they have got CTSpackets from receiving nodes to begin transmitting exceptfor nonsending nodes that return into the sleep mode rightaway as if the channel was idle [33] Since sending nodeshave separate schedules it reduces the contention proba-bility leading to more energy conservation A time frame isdivided into the information period the allocation periodand the communication period which are discussed asfollows

Information period at frame initiation a beacon is sentby the receiving node declaring its return from sleepand readiness to accept data packets in case the channelis not busy -e beacon is a small packet alerting otherlistening nodes of the transmitterrsquos readiness to get arequest for channel It consists of the receiverrsquos ID andthe duration allocation period (DAP) that relies onnumber of the receiving nodersquos neighbors [13]Allocation period here the receiving node stays awake forsome little time after broadcasting its beacon to confirm ifthere is any node having anRTS packet to transmitWhenthe beacon has been received an RTS packet (with 3fieldsmdashsender ID receiver ID and data size) is sent by thesender prompting the receiver to send a CTS for SYNC asthere is neighbor information exchangeCommunication period SRI-MAC is capable of or-ganizing and scheduling transmissions by means ofsensor node IDs -e receiver assigns order andtransmitting time for every sender such that they switchoff their radios till their time to transmit is dueHowever the receiving node remains in the wake-upmode to get all sending nodersquos data SRI-MAC canenable sensors have lengthy sleeping times duringtransmission such that energy can be saved as long as itis neither sending nor receiving data -e protocolrsquoscommunication period is premised on the TDMAtechnique which conserves energy since the radiorsquosduty cycle is lowered and collision is eliminated [33]

Advantages

(i) Since it employs the TDMA principle it eliminatescollisions while maintaining energy conservationtoo

(ii) Noncommunicating nodes switch off their radioand get into the sleep mode which eliminatesoverhearing

(iii) Since receiving nodes broadcast SYNC signals toall nodes in the form of CTS packets overemittingis avoided

(iv) It decreases duty cycling as sleep time increasesand lowers idle listening

10 Journal of Computer Networks and Communications

2110 H-MAC -is protocol is built on the power savingmechanism (PSM) of IEEE 80211 and slotted Aloha pro-tocols to enhance performance using multiple slots Timehere comprises large frames each having an active and asleeping part [34] If any node has got data to send it will askfor slots from the receiver node during active time and latertransmit the packets in the prenegotiated slots Nodes gointo the sleep mode during sleep-time slots if they do nothave data to transmit

Latency in IEEE 80211 if N is the number of hops nthe value for the current hop tcsn the back-off time andttx the transmission delay the total latency is calculatedusing the following equation

D(N) 1113944N

n1tcsn + ttx1113872 1113873 (8)

and the average latency is given by the followingequation

E[D(N)] N tcsn + ttx1113872 1113873 (9)

-roughput in H-MAC is given by the followingequation

Th npnm

tact + Cts (10)

where np is the number of transmitted packets nm is themaximum number of nodes connected in a time frameC is the number of slots with the same length tact is thelistening time and ts is the sleep delayAdvantages

(i) It enhances performance using multiple slots

2111 Z-MAC (ZebraMAC) -is rides on the strengths andoffsets weaknesses of TDMA and CSMA It yields very goodchannel usage and low delay under low contention and lowerscollision among two-hop neighbors at a reduced cost Itsperformance is strong via error synchronization failure inassigning slots and channel conditions that change with timeIn the set-up phase the protocol operates through certainsteps including neighbor discovery slot assignment localframe exchange and global time synchronization whichunless the topology of the network changes significantly donot run again When the data are being transmitted there isimprovement in network throughput and energy efficiencywhich covers up for the initial upfront operational costs [35]

Disadvantages

(i) Initial slot assignment is quite a cumbersome taskhere since it is done at the beginning yet this maynot be at the same time in reality It further as-sumes static links until setup happens again whichtoo is impractical

(ii) It also suffers from the hidden-terminal problemsince slot owners have a small contention windowand priority for slots allocated to them due to theECN messages

(iii) Synchronizing time also poses a challenge in thisprotocol due to a large clock drift between nodes

(iv) Source and destination nodes too are not wellcoordinated in Z-MAC [36]

2112 Q-MAC (QoS-Aware Media Access Control) -isprotocol offers high QoS with improved energy efficiencyand accesses the wireless channel using MACAW as theunderlying protocol [37] Traffic from different nodes isprioritized for QoS satisfaction depending on how criticalthe data are It mainly has two scheduling algorithms inWSNs (i) intranode and (ii) internode scheduling -eintranode scheduling scheme adopts a first-in first-out-(FIFO-) based queuing algorithm based on application andMAC layer abstraction for data packet classification In-ternode scheduling reduces idle listening and collision on thechannel for improved energy consumption

Advantages

(i) It gives a high QoS with improved energy efficiency

2113 WiseMAC (Wireless Sensor MAC) WiseMAC isbased on the Aloha protocol and employs preamble sam-pling in achieving low-power transmissions in infrastructuresensor networks by occasionally sampling the channel tocheck for activity [3238] -e protocol employs the samemethod as B-MAC except that the sender shortens theextended preamblersquos length by scheduling its transmissionsaccordingly after learning the receiverrsquos awake schedules-is is done by the receiver node putting its subsequentawake time on the ACK packet such that when the senderneeds to send to it again it starts the preamble shortly beforeit wakes up so as to avoid energy wastage in transmitting thepreamble [32] -e data frame is repeated instead of theextended preamble for less traffic transmissions that have alengthier preamble than the data frame in WiseMAC Afterprocessing the frame a node that is not the target recipientwill go back into the sleep mode while the recipient staysawake till the session ends thereby sending an ACK packetFor an idle channel the power consumption is rather lowwith this method Nevertheless there is a problem of longwake-up preamble limiting throughput and causing anenormous power consumption overhead in reception [38]All nodes overhearing a transmission bear the overheadtogether with the recipient node -e protocol can addressissues concerning low-power transmission but has nomechanism of nodes adapting to varying traffic patterns[32] To compensate for drift between clock at the accesspoint and on the sensor node there is a need to compute thewake-up preamble duration -e drift varies directly withtime since previous resynchronization and is given by thefollowing equation

TP min 4θl TW( 1113857 (11)

where θ is the time-base quartz frequency tolerance and l isthe interval between two transmissions Increase in trafficvolume naturally reduces overhearing

Journal of Computer Networks and Communications 11

Radiomodel for protocols with lower power transitiondelays between transceiver states and their powerconsumption are modeled Some of the states includeDOZE where the transceiver neither transmits norreceives but is prepared to power on to the receivetransmit state very fast RXwhere the transceiver listensto the channel receives data or demodulates data fromthe noisyidle channel and TX which is the trans-mission state by the transceiver Let TS be the set-uptime needed to turn the transceiver from the DOZE tothe RX or TX state TT the turn-around time needed toswitch between RX and TX states and B the trans-ceiverrsquos bit rate PZ PR and PT are the values of powerconsumed in the DOZE RX and TX statesrespectively

1113954PR PR minus PZ

1113954PT PT minus PZ(12)

where 1113954PR and 1113954PT are the changes in power as a result ofbeing in RX and TX states respectivelyTraffic model let N be the number of sensor nodesconnected to an access point (AP) and λ be the global rateat which the downlink Poisson traffic arrives at the APfrom the fixed network Traffic toward each sensor node isthe same and is given by λN Average packet interarrivaltime Lwith which data packets arrive at a node is given byL Nλ If TC and TD are the control packets and datapackets respectively assuming a small traffic volume theglobal interarrival 1λ far exceeds the summation ofdurations for the data packet and control packet and theturn-around time Mathematically it is given as

1λ≫TD + TC + TT (13)

-e average power consumed by WiseMAC andtransmission delay incurred are given by equations (14)and (15) [38]

PW PZ +1113954PR TS +(1B)( 1113857

TW+

1113954PR X + TD + TT( 1113857

L

+ 1113954PR(N minus 1)Y

L

(14)

where

X 2θL 1 minus eminus TD4θL( )1113874 1113875

Y T2D + 12TDθL

2TW1 minus e

minus TW4θL( )1113874 1113875

DW TD +TW

21 minus e

minus TW4θL( )1113874 1113875

+ 2θL 2 minus eminus TD4θL( ) minus e

minus TW4θL( )1113874 1113875

(15)

where PW and DW are the power consumed byWiseMAC and the transmission delay for WiseMACrespectivelyAdvantages

(i) It achieves low-power transmissions in in-frastructure sensor networks through preamblesampling while checking for activity

(ii) It reduces overhearing due to increased trafficvolume

Disadvantages

(i) It has a problem of long wake-up preamble limitingthroughput and causing enormous power con-sumption overhead

2114 X-MAC It is a simple minimal transmission powerMAC protocol that uses a short preamble with decouplingfor the sending and receiving node sleep plans inWSNs-isleads to a significant reduction in energy usage at sendingand receiving nodes and decrease in per-hop latency andgives flexibility in adapting to bursty and periodic sensordata sources [32]

Protocol design X-MACrsquos design is aimed at increasingenergy efficiency reducing data latency and increasingthroughput -is enables application in different kindsof packetized and bit stream digitized radios as well assimplified low-overhead distributed application -edesign is as such in order to address the low-powerlistening problems including overhearing excessivepreamble and incompatibility with packetizing radiosas in [32]Asynchronous duty cycling a node with information totransmit will send the packets after sending a preamblewhereas the remaining nodes will preserve their sleepschedules that are not synchronized -e receiver willsample the link on waking up and stays awake if it detectsa preamble and if it is the target it will return to sleep if itis not the target after getting the whole preambleEmbedding the target ID in the preamble to avoidoverhearing the lengthy preamble is divided into manyshortened packets embedded with the receiver nodersquosID so as to eliminate overhearing One lengthy pre-amble is constituted by the shortened preamble packetstream Any node on receiving the short preamble afterwaking up will examine the target ID on the packet Itgoes back into the sleepmode right away resuming dutycycling if it is not the targeted recipient or stay awakefor successive packets if it is the one -is approachchecks energy wastage due to network density It canalso be used on several types of radiosReducing excessive preamble using strobing it ispossible to enable low-power transmission and alsosave energy by reducing time for preamble trans-mission through the use of enhanced preambles inaddition to preamble sampling X-MAC does not sendan endless preamble packet stream but puts slight

12 Journal of Computer Networks and Communications

pauses in between packets to allow the sending node topause and listen to the channel and allow the receivingnode to resend an early ACK packet during that pause-is makes the sending node begin sending datapackets and the receiving node cut excessive preamblesand thus decrease per-hop latency and energy wastageon unnecessary waiting and transmissionPacketizing radios low-power listening is limited incapability for supporting packetizing radios butX-MACrsquos short strobed preambles can support all typesof digital radiosAdaptation to traffic load while a number of WSNapplications are reliable in traffic production they stillmust adapt to changing traffic Nodes are going to haveseparate sleep schedules depending on traffic loadConsidering systems that have time-fluctuating trafficall predetermined static schedules are suboptimalOptimality Buettner et al [32] model the expectationof energy as follows

Es (preamble energy + energy perACK listen)

lowast(expected preamble minus listen iterations)

+(energy to send parket)

Es PTxSp + PRxSal1113872 1113873 Rl + Rs( 1113857

Rl minus Sp+ SdPTx

(16)

where Es expected energy to send a packet PTx andPRx are the transmission and receiver power re-spectively Sp Sal and Sd are the duration of thesenderrsquos preamble ACK listen and data transmissionperiods respectively and Rl and Rs are the receiverrsquoslisten and sleep periods respectively Expected energyto receive a packet is given as

Er (listen cycle energy + sleep cycle energy)

lowast(expected iterations for preamble arrival)

+(energy to sendACK) +(energy to receive packet)

Er PsRs + PRxRl( 1113857

1 minus 1 minus Pd(t)( 1113857 Rl + Rs( 1113857+ RaPTx + RdPRx

(17)

And the one-hop expected latency is given as

Lat (preamble duration + ACK listen)

lowast(expected number of iterations)

+ duration to send packet

Lat Sp + Sal1113872 1113873 Rl + Rs( 1113857

Rl minus Sp

(18)

Advantages

(i) It increases energy efficiency(ii) It reduces data latency(iii) It increases throughput

2115 A-MAC (Advanced MAC) A-MAC is a TDMA-based protocol whose design is aimed at enhancing thelifetime of the network for low rate and reliability of datatransmission [39] To assign time slots it does not depend ona central controller like most protocols but it gathers in-formation from its neighbors so as to allocate the slotsthrough a distributed algorithm Energy is controlled byscheduling power when there is not a single event Its designcomprises many frames with each having many time slotswhich at the beginning have got a beacon transmitted tosynchronize and exchange information with neighborsNodes to participate in the subsequent session are selectedby the controlled node so they do not go into the sleep modewith others [13] -is protocol operates under four time slotdivisions namely initial wait discover and active states Anode in the initial state begins by listening to the medium forthe beacon packet from other nodes for network synchro-nization which when received makes it change the timerthrough subtraction of beacon transmission time from re-ception time To avoid synchronization problems due todrift and allow for continuity the node has to get thestrongest beacon signal It then enters the wait state afterchoosing a number of waiting frames so as to reduce theprobability that nodes get into the discover state simulta-neously In case the synchronization beacon is lost while stillin the wait state it will return to the initial state for anotherone On expiry of the waiting counter it enters the discoverstate and collects information from neighbors by listening totheir beacon messages On successful selection of a time slotit enters the active state where it endlessly transmits beaconsat the initiation of the slot A node goes to sleep when abeacon has been transmitted and there are no more data fortransmission or in case the neighborrsquos beacon received in-dicates no data packets are coming in [39]

Advantages

(i) It can enable multicasting(ii) It enhances the lifetime of the network(iii) It offers reliable data transmission

2116 L-MAC (Lightweight MAC) Based on TDMA thisprotocol is aimed at minimizing the number of transceiverswitches adapting the nodal sleep interval to data trafficamount and making implementation less complex [40] Apacket is transmitted comprising a control and a data partControl packets have a fixed size and the time slot con-trollerrsquos ID -ey show the distance from the transmittingnode for simplified routing to the network gateway as well asaddressing target node and reporting data unit length andinternode synchronization Adjacent nodes are more in-terested in getting control packets from nearby nodes -enodersquos transceivers are switched off until the subsequenttime slot unless the node is also addressed or the message isomnicast If so it listens to the data unit which may not fillthe remaining time slot in totality-e transceivers on eitherside are turned off when the transfer of packets is successfulEnergy wastage through idle listening in uncontrolled timeslots is avoided via a short time-out interval Network nodes

Journal of Computer Networks and Communications 13

are able to communicate without colliding-is saves energyand increases the lifetime of the network A node in L-MACcan only transmit one message per frame [13] -e networkis set up in such a way that all nodes are not synchronized onbeing switched on-e gateway takes control of a timeslot soas to have them synchronized and its immediate neighborgets the control messages thereby synchronizing their timerswith the gatewayrsquos Neighboring nodes learn of multiplegatewaysrsquo time slots in their vicinity just after one frame-ejust synchronized nodes randomly choose time slots forcontrolling as long as they are not occupied -e nodes thuspreserve a table for the neighboring nodes which makes atimeslot reusable after a minimum of 3 hops and preventsmessage collision All nodes keep their slots before thebattery is down or learn of an impending collision All nodescan track their hop distances to a selected gateway node andthe information is communicated in control messages [40]

Advantages

(i) It minimizes the number of transceiver switches(ii) It adapts the nodal sleep interval to data traffic

amount(ii) It makes implementation less complex(iii) It offers improved network lifetime(iv) Energy wastage through idle listening in un-

controlled time slots is avoided via a short time-out interval

Disadvantages

(i) L-MAC is not very good for mobile SNs since slotsare calculated only once

(ii) It suffers idle listening because nodes keep listeningto slotsrsquo control parts so as to get data and enableother nodes to join the network all the time [13]

2117 C-MAC -is was designed for WSNs with 3 seg-ments namely (i) aggressive RTS having double channelcheck to assess the channel (ii) anycast to initialize the flowand (iii) convergent packet forwarding to stabilize the flowIt promises an improvement in energy efficiency lowerlatency and enhanced throughput while avoiding syn-chronization overhead though it operates only in low duty-cycling apps especially where there is less traffic But whenthere is more traffic the protocol employs anycast-basedpacket forwarding for waking the nodes up or finding aforwarder quickly and converges from route-suboptimal toroute-optimal unicast After this energy could be preservedby the nodes using a synchronized wake-up plan [13]

Advantages

(i) It improves energy efficiency(ii) It lowers latency(iii) It enhances throughput avoiding synchronization

overhead

Disadvantages

(i) It operates only in low duty-cycling applicationsespecially where there is less traffic

22 MAC Protocols for WMSNs

221 Diff-MAC -is is a CSMACA-based QoS-awarehybrid-priority-based MAC protocol for WMSNs It targetsenhancing the channel utilization while making use of ef-ficient service differentiation to do coordination for mediumaccess of each traffic class while giving fair quick datatransfer It offers QoS using the following features [14]

(i) It balances energy consumption and delay throughsensor node duty-cycling adaptation depending onthe dominant traffic class

(ii) It has a feature for fragmentation and messagepassing that breaks lengthy video frames intosmaller packets that are transmitted as a burst thusreducing retransmission costs if there are any MACfailures

(iii) -ere is fair data delivery amongst the nodes andtraffic classes respectively resulting from intranodeand intraqueue prioritization reducing unbearableperformance

(iv) -e number of collisions can be reduced and packetlatencies are kept within limits by adjusting the sizeof the congestion window in this protocol as per thetraffic needsAdvantages

(i) Diff-MAC is very suitable for WMSNs since itadapts quickly to changes in networkconditions

(ii) It is a fair protocol(iii) It is scalable

Disadvantages

(i) It is quite difficult to monitor the statistics ofthe network

(ii) Adapting dynamically to the network condi-tions is not an easy task

(iii) It exhibits some degree of latency in packetdelivery

-e implementation of Diff-MAC on Imote2 indicatesthat resource requirements can be met on the sensorhardware Results of extensive simulation runs showed thatDiff-MAC outperforms its competitors

222 Cluster-Based On-Demand Multichannel MAC Pro-tocol for WMSN (COM-MAC) -is is an energy-efficientprotocol in WMSNs with high throughput and reliable datatransmission It achieves energy efficiency by schedulingmultichannel media access to reduce channel contentionand remove collision idle listening and overhearing [41]

Design it has a three-phased operation with the (i)request phase (ii) scheduling phase and (iii) datatransmission phaseRequest phase here all nodes have two protocols eachfor sending requests to the cluster head (CH) ie thecontention-based and contention-free TDMAFDMA

14 Journal of Computer Networks and Communications

protocols -e contention-based protocol allocates achannel per node for sending the request message andfree channels are useable as control channels Forcontention-free protocols a control slot is allocatedwhen a network is deployed Here a slot per channel isallocated to a different node enabling requests to be sentwithout interfering with transmissions from othernodes -e contention-based protocol operates undertwo steps (i) Control channel assignment phase achannel is assigned per node for transmitting a request(REQ) message A channel is assigned to multiplenodes and nodes are evenly distributed to availablechannels for congestion control (ii) Request trans-mission phase nodes with data send a REQmessage viathe assigned control channel to notify the CH whichresponds with an ACK message -e contention-freeTDMAFDMA protocol too has two phases includingthe control slot assignment and request transmissionphases For unreliable channel conditions and lowtraffic load the contention-based protocol is fit for usebecause of less collisions and congestion whereas thecontention-free protocol is preferable in heavy trafficload and reliable channel applicationsScheduling phase from the above phase a schedule isgenerated by the CH for coordination and this enablesnodes to transmit and broadcast their data via controlchannels-e schedule for broadcasting only comprisesdata for nodes assigned to a channel -e schedule isrebroadcast for transmission reliability enhancementbut for energy efficiency the nodersquos transceiver isswitched off when all the schedule is completely de-livered till it is the time for transmitting again-e timeslot and data communication radio channel are allincluded in the schedule generated for each nodeData transmission phase on getting the schedule thesensor node sends its data as per the time slot and thechannel assigned -e time slot is divided into datatransmission andACK sections ACK supports link layererror control Using implicit selective repeat ARQ theCH sends an ACK message for all received packets Ifthere is no ACK the packet is assumed lost and hence aretransmission during the subsequent interval is done-is enhances reliability of transmissions though excessdelay is possible here which is also intolerable for real-time critical apps like video or audio -is gives rise to ahybrid MAC protocol to exploit the unused spectrumduring data transmission sessions [41]Advantages

(i) It achieves energy efficiency by scheduling mul-tichannel media access to reduce channel con-tention and remove collision idle listening andoverhearing

(ii) It maximizes network throughput by dynamicallyallocating time slots and channels for nodes byexecution of a traffic-adaptive QoS-aware sched-uling algorithm depending on needs for QoS andtraffic data information

(iii) It increases transmission reliability by in-corporating spectrum-aware ARQ to avoid spec-trum wastage

223 EQoSA -is protocol is developed for video andimage transfer across WMSNs in a hybrid mode withprovision of QoS Concerning the active sensor nodes andtheir traffic loads it employs dynamic session sizes andchanges bitmap-assisted (BMA) MACrsquos fixed session sizeNodes declare any data available for transmission duringcontention time -is forces the cluster head to make aschedule of how slots have been assigned It communicatesthis schedule to the nodes giving each node a given numberof slots per session and so enabling the busty traffic to beaccommodated [14] -e protocol does not perform wellwith DiffServe mechanisms but does well with schemessupporting busty traffic loads and strong cluster heads arenecessary in WMSNs that can assign and broadcast slots

224 QoS-Supported Energy-Efficient MAC (QE-MAC)Based on the IEEE 80211e standard this protocol ensuresfairness and has low latency and jitter and better efficiencyon energy usage than other QoS-aware MAC protocols forWMSNs It operates under a two-phased routine

-e first phase introduces the innovating prioritymechanism in IEEE 80211e QoS and several data types areassigned priorities Its coordination function is a combi-nation of distributed contention-based channel allocationand centralized polling-based channel access mechanisms

-e second phase preserves the nodesrsquo energy usingdynamic duty cycling -ere is an exchange of RTSCTSpackets to supplement CSMACA by muting nodes near thesenderreceiver during packet duration [42]

Advantages

(i) It ensures fairness(ii) It has low latency and jitter and better efficiency on

energy usage than other QoS-awareMAC protocolsfor WMSNs

225 Black-Burst (BB) Contention A contention-basedMAC scheme for enhancing instantaneous QoS access tocarrier-sense WSNs is proposed in [42] In this schemenodes contend for the channel until it is free Real-timepackets are given higher priority over non-real-time packetsand if a node is transmitting real-time packets it is prior-itized over other nodes [15] Voice and video are some of thereal-time apps in consideration here which periodicallyaccess the common radio channel when transmitting -eseneed constrained end-to-end delay of packets at the MAClayer -is protocol majorly does the following [42]

(i) Prioritization of real-time traffic(ii) Implementation of the round-robin mechanism in

real-time nodes(iii) Bounded access delays among real-time packets

Journal of Computer Networks and Communications 15

(iv) Implementation of real-time communication withvarying needs of bandwidth good for multimediatraffic

Amechanism for dynamically assigning packet prioritiesbased on their deadlines and the traversed hops is proposedin [43] -e ReAllocative Priority (ReAP) mechanism ismeant to provide QoS at the MAC layer to video traffic -isscheme introduces adaptive TXOP by modifying the TXOPdynamically based on the queue length [15] Other protocolsin this category include dynamic duty cycle and adaptivecontention window-based MAC protocol frame sharing(FRASH) MAC protocol and real-time independentchannels (RICH) MAC protocol

Advantages

(i) It can implement real-time communication withvarying needs of bandwidth

23 Challenges and Opportunities(a) One of the challenges in satellite transmission is

cross-talk leading to interference in different fre-quency bands It is necessary to increase the rise andfall times as well as protect sensitive nodes as wecircumvent floating ones if we are to regulate cross-talk as in satellite communication therefore Wefurther need to investigate the possibility of usingdifferential signaling -is constitutes an opportu-nity for further research

(b) Another challenge is that the hidden and exposedterminal problems are still noticeable despite the useof RTSCTS solutions in MAC since collisions arestill possible from transmissions of separate nodesAt the MAC layer ACKs might block efficientchannel usage by prohibiting exposed terminalsfrom reusing the channel and may be eliminatedleaving no proof of sender data packet receptionUsing a directional antenna is one of the proposedtechniques of high-speed WSNs like IEEE 80211-is constitutes an opportunity for further study intothese issues for better solutions though

(c) Another challenge is that since nodes in contention-based MAC protocols are mobile there are manycollisions by the packets and schedule in-consistencies in schedule-based protocols in the two-hop neighborhood information when nodes areentering or leaving [44] Deng et al [45] propose amobility-based clustering (MBC) protocol for mo-bile nodes to enhance the packet delivery rate -isprotocol in comparison with CBR has 25 chancesof decreasing collision-induced packet losses and50 chances in comparison with low-energyadaptive clustering hierarchy (LEACH)

(d) -e process of analyzing and evaluating routingperformance for WMSNs is quite complex becauseof the bursty multimedia traffic Because of the wayencoding scheme frames are structured as well as thechanges in between the scenes that occur naturally

the compressed video frames display some substantialburstiness on several time scales -ere will also befrequent route recalculation that is as well energy in-tensive for nodes frequently capturing and transmittingenergy -e solution here could as well possibly be theuse of the MBC protocol Route recalculation tooshould be dynamic to cater for the mobile nodes whichmight be in and out of the range at different instants oftime Recently the use of smart dumpsters for effectivewaste management in smart cities has also been pro-posed as a possible solution [46]

(e) For WSNs their routing protocols do not look atmultimedia applications which require enormousbandwidth and processing energy and should betransmitted in real time with utmost fidelity Extrachallenges in routing are imposed in WMSNs indeployment of heterogeneous sensors since therecould be audio video and still pictures in addition toscalar data during transmission Heterogeneous datahandled also have different QoS requirements andwith many different business needs different ser-vices with varying requirements give rise to a bigchallenge for the routing design -ere is a need todesign protocols that can support WMSNs such asEQoSA [14] for this purpose

Table 2 summarizes the reviewed MAC protocols forWSNs and the ones designed for suitability in WMSNs

3 QoS Strategies at the Application Layer

We discuss many protocols used to solve different needs ofcommunication between machines

31 ConstrainedApplication Protocol (CoAP) It is a requestresponse protocol designed by the Internet Engineering TaskForce (IETF) for constrained devices such as those havinginsufficient RAM or CPU and WPANs with low-powerconstraints resulting in poor transfer of packets andenormous overhead -e protocol was designed for ma-chine-to-machine (M2M) apps and system automation toreduce bandwidth requirements increase packet transferdecrease the overhead and make work simpler leading tolightweight implementation HTTP commands like GETPOST PUT and DELETE are used in a client-server ar-chitecture Using the publishsubscribe architecture CoAPcan support multiple users and hence yield better perfor-mance via multicasting enabling asynchronous messageexchange -is protocol can support both multicasting andunicasting via UDP and hence is a reliable and simpleprotocol -is is done via the messaging sublayer (for re-liability based on UDP) and the requestresponse sublayer(for communication and interaction) [16]

It uses the following messages

(i) Confirmable message this guarantees reliabilityusing ACK messages sent by the receiver node-ese could be synchronous or asynchronous ifthere is a need for more computational time

16 Journal of Computer Networks and Communications

(ii) Nonconfirmable message here no ACK messagesare sent back

(iii) Acknowledgment (ACK) message this confirmsreceipt of a confirmable message

(iv) Reset message this confirms receipt of an un-processed message in case any critical parts nec-essary for interpreting the message are missing

(v) Piggybacked response the receiver directly re-sponds while getting the ACK message

(vi) Separate response the reply by the receiver is dis-crete from the ACK message

CoAP has no in-built security features It is secured bythe datagram transport layer security (DTLS) that runs ontop of UDP though it was not designed for the IoT -ismakes its suitability debatable since DTLS offers no supportfor multicasting It also needs more packets for handshakingthereby increasing the traffic volume and computationalresources and so shortening the lifespan of mobile devicessince they use batteries [17]

Advantages

(i) It can support multiple users and hence yieldbetter performance via multicasting

Table 2 Comparison of MAC layer protocols used in QoS provisioning

Protocol Features QoS parameter Comments Access Priorityassignment Energy eff Designed for

S-MAC[22] Collision avoidance Energy latency Requires periodic

synchronization CSMA Hybrid Medium WSNs

B-MAC[23] Collision avoidance Energy delay

throughput Flexible interface CSMA Dynamic High WSNs

T-MAC[25] Collision avoidance Energy Complex nonscalable CSMA Dynamic High WSNs

C-MAC[13]

Dynamic channelallocation Energy Multichannel MAC CSMA Dynamic Medium WSNs

CU-MAC[24]

Multichannelmechanism

-roughputlatency High packet delivery ratio TDMA Dynamic mdash WSNs

A-MAC[39]

Per flow serviceguarantees Delay m-ary tree model

dynamic prioritiesTDMACSMA Dynamic Medium WSNs

ER-MAC[30]

Emergencyresponse

Energy delaydelivery ratio

Adapts to traffic andtopology changes

TDMACSMA Hybrid Higher than Z-

MAC WSNs

RI-MAC[31] Collision detection Delay packet

delivery ratioHigh packet delivery ratio

and throughput CSMA Dynamic Medium WSNs

SRI-MAC[33] Collision avoidance Energy delay Good data transfer and

energy consumption TDMA Dynamic High WSNs

H-MAC[34] Slotted sleep time -roughput

latencyData transmission only in

sleep timeCSMAAloha Dynamic Medium WSNs

Z-MAC[36] Collision avoidance Energy delay

throughputVery good channel usage

and error syncCSMATDMA Dynamic Medium WSNs

Q-MAC[37]

Intrainternodescheduling Energy High QoS improved

energy efficiency CSMA Hybrid High WSNs

X-MAC[32]

Overhearingavoidance

Energy delaythroughput

Reduces excessivepreamble by strobing CSMA Adaptive

static Medium WSNs

L-MAC[40] Collision avoidance Network

lifetime energyReduces the number oftransceiver switches TDMA Static Medium WSNs

WiseMAC[38]

No downlinkchannel collision Power delay Significantly low-power

consumptionAlohaCSMA Static Higher than

ZigbeeInfrastructure

WSNsAR-MAC[27] Collision avoidance Energy delay Uses data packets and

control packets TDMA Static Higher thanIEEE 802154 WBANs

BB-MAC[42]

Real-time trafficprioritization Delay Employs round robin in

real-time nodesCSMATDMA Dynamic High WMSNs

COM-MAC [41] Collision avoidance Packet delay

throughputBalances reliability and

retransmissionTDMAFDMA Dynamic High WMSNs

EQoSA[14]

Busty traffic loadaccommodation Energy Provides QoS support for

video and imagesTDMACDMA mdash High WMSNs

QE-MAC[42] Collision avoidance Energy jitter

latencyRuns central polling-like

channel access CSMA Dynamic High WMSNs

R-MAC[26] Collision avoidance Energy delay Reservation based mdash Static High Underwater

SNs

Journal of Computer Networks and Communications 17

(ii) It enables asynchronous message exchange(iii) It is a reliable and simple protocol

32 Message Queue Telemetry Transport (MQTT)Developed by IBM and later improved by OASIS with anintention of decreasing bandwidth MQTT is a simple open-source publishsubscribe protocol analogous to the client-server version It is appropriate for lightweight IoT appli-cations and M2M transmissions in constrained situationslike limited power computation capability and memory orbandwidth -e network bandwidth and the computationalresource usage are decreasing since clients are not updatedmaking publishsubscribe protocols outperform requestresponse protocols via IoT requirements [16] -is protocolcan run over TCPIP and so has the following advantages

(i) Delivering of packets with reliable guarantee(ii) Enabling multicasting(iii) Remote device session establishment(iv) Reduction of transport overhead and protocol ex-

change to minimize network traffic(v) Provision of the notification mechanism when

anomalies occur [47]

MQTTrsquos design targets economical use of bandwidthand battery so it has a major application in FacebookMessenger MQTT ensures QoS and reliability of messagedelivery by providing the following three QoS levels

(1) Fire and forget a message is sent once depending onthe networkrsquos best effort and there is no need forACK (lowest QoS level)

(2) Delivered at least once a message is sent at least onceand an ACK message is needed to avoid duplicates

(3) Delivered exactly once by a four-way handshake itensures a message is sent exactly one time (highestQoS level)-e architecture it has three key components (i)publishersmdashlightweight nodes sending data to thebroker before returning to sleep (ii) broker whichorganizes sensory data in topics and sends them tointerested subscribers and (iii) subscribersmdashIoTapps interested in data sent by sensors through thebroker as shown in Figure 3

MQTT is a simple lightweight protocol that is betterthan CoAP via low and high packet loss probability interms of throughput and exhibits lower latency in lowsampling rates [48] It supports a number of imple-mentations for embedded devices and mobile applicationsand offers web-socket support too Nevertheless it hassome shortcomings

(i) It is only employed in simple data types(ii) It exhibits high latency in high sampling rates(iii) It is unsuitable for real-time apps in such cases(iv) It does not provide automatic discovery andor

sufficient security at the protocol level [49]

33 Secure Message Queue Telemetry Transport (SMQTT)-is is an extension of MQTT that is based on lightweightattribute-based encryption using elliptic curve cryptogra-phy -e protocol has three entities as shown in Figure 4(i) publisher that publishes information under a certaintopic (ii) subscriber that obtains the same topic in-formation via a broker and (iii) public key generator(PKG) or broker that acts as the trusted third party -eprotocol has four phases namely (a) set-up phase (b)encrypt phase (c) publish phase and (d) decrypt phase

(a) Set-up phase publisher and subscriber provide aunique ID with attributes for registration with thePKG -is generates a secret master key set andpublic parameters that it publishes with set U A1A2 An For ciphertext policy-attribute-basedencryption (CP-ABE) the PKG gives a key set todevices with attributes [50]

(b) Encrypt phase based on the access tree the publishercreates an access policy An access tree is sent by thepublisher to the PKG for generating the key-gen-eration policy in the key policy- (KP-) based ABEalgorithm -e publishers know beforehand thetopics and subscribers to access as well as the rulesfor access Subscribers are issued with private keysfor all communication sessions Publishers in theCP-ABE algorithm come up with access trees andrules where the payload is encrypted by the pub-lisher who also issues more info needed for de-cryption and the rules With the help of public keycryptography data are encrypted giving rise to ci-phertext on the basis of KPCP-ABE and the receivergets SUBACK from the broker [51]

(c) Publish phase ciphertext is embedded in theSPublish command as payload and topic name isdefined in the header by the publisher -e brokerreceives the SPublish packet and replies withPUBACK prompting the publisher to respond withthe PUBREL packet -e broker then broadcasts totopic subscribers and deletes the data forwarding thePUBCOMP packet to the sender node

(d) Decrypt phase the encrypted information isdecrypted by the subscriber with the help of a secretkey Considering KP-ABE on verification that itsatisfies the access policy the subscriber invokes thePKG for a corresponding key which it sends afterverifying the request With CP-ABE subscribers usethe private key for decryption after authenticationthereby enabling offline interaction [51]Advantages

(i) -e information being encrypted makes it moresecure

34 MQTT for Sensor Networks (MQTT-SN) -is is a ver-sion of MQTT designed to operate under environments oflow bandwidth high link failures and short message lengthIt is a publishsubscribe (pubsub) protocol designed for

18 Journal of Computer Networks and Communications

UDP-based WSNs to visualize communication for low-power devices [5152] It comprises a publisher device actingas the MQTT-SN client to forward communications to agateway that also changes the MQTT-SN message to theMQTTmessage before delivering it to the broker that sendsit to MQTT-SN subscribers [51]

It has a gateway for changing protocols from MQTT toMQTT-SN (Figure 5) [52] -is protocol supports nodes in

the sleeping mode using an offline keep-alive procedure thatcan be used by battery-operated devices to get into thesleeping mode To support sleeping clients MQTT-SN has anew offline keep-alive procedure with which battery-oper-ated devices can go to the sleeping state in which theirmessages are buffered at the server until they are awake InMQTT-SN clients with no preconfigured gateway addresscan figure out the real address of their operational gateway

Publisher

Subscriber

Subscriber

Broker

Publisher

4G

Figure 3 Architecture of the MQTT protocol

Register by mentioning URI Register by mentioning URI

Access tree

Set-up

Publisher ReceiverPKG(broker)

Key management Key management

Encrypt data using public parametersand random key This is done using

CP-ABE Publish it as

Subscribe (topics)SUBACK

S-Publish (TopicEncrypt (data))S-Publish (TopicEncrypt (data))

Encryptand publish

PUBACK

PUBACK

PUBRELPUBRELPUBCOMP

PUBCOMP

Figure 4 SMQTT protocol

Journal of Computer Networks and Communications 19

within the network through a discovery procedure Simul-taneously we may have several gateways in a single wirelessnetwork that can be cooperative via load sharing

MQTT-SN architecture it is composed of MQTT-SN clients gateways and forwarders (Figure 5)rough the MQTT-SN gateway (GW) the MQTT-SN protocol is used by the MQTT-SN clients toconnect to the MQTT server Integration betweenthe MQTT-SN GW and the MQTT-SN server mightbe possible or not Translating between the MQTT-SN and the MQTT for a stand-alone GW is doneusing the MQTT-SN protocol If a GW is detachedfrom the network it can be accessed by the MQTT-SN clients through a forwarder is just encapsu-lates the received MQTT-SN frames that are on thewireless side before forwarding them to the GW un-altered It then decapsulates the received frames tooresending them to the client as sent by the GW [52]ere are 2 types of gateways (a) transparent gatewayand (b) aggregating gateway as shown in Figure 6

(i) Transparent gateway this launches and sustains adirect MQTT connection between the MQTT-SNclient and the MQTT server which is entirely re-served for end-to-end client-server messaging enumber of servers and that of MQTT-SN clientsconnected to the GW are the same is transparent

GW translates syntax between these protocols and iseasier to implement compared to the aggregatingGW But the MQTT server must support dierentsessions for all connected clients Some MQTTservers are congured to enable few parallel sessions

(ii) Aggregating gateway this maintains a single MQTTconnection to the server on behalf of all clients Allexchanged messages from the MQTT-SN client to theaggregating GW are directed to the GW which sortsthe information to forward to the server e aggre-gating GW decreases the number of MQTT sessionssimultaneously supported by the server in WSNs withmany battery-operated sensors and actuators (SAs)having limited storage and processing capabilities It ishowever more complex to implement compared to atransparent GWe general format of the MQTT-SNprotocol message has two or four message headeroctets and n octets of variable header that is optionaldepending on the message in question MQTT-SN has256 message types (MsgType) with those from0X1Eminus 0XFD 0X19 and 0XFF reserved for futureusage [51]

Advantages

(i) e protocol can operate under environmentsof low bandwidth high link failures and shortmessage length

MQTT-SN gateway

MQTT-SN

MQTT

MQTT-SN

MQTT-SN

MQTT-SN

MQTT-SN client

MQTT-SN clients

MQTT broker

MQTT-SN client

MQTT-SN client

MQTT-SN gateway

MQTT-SN forwarder

Figure 5 MQTT-SN architecture

20 Journal of Computer Networks and Communications

(ii) It is a pubsub protocol designed for UDP-basedWSNs to visualize communication for low-power devices

(iii) It can be optimal for low-cost battery-operatednodes having low processing power and lessstorage

(iv) It supports nodes in the sleeping mode using theobrvbarine keep-alive procedure used by battery-operated devices

35e Extensible Messaging and Presence Protocol (XMPP)e standardization of this protocol by the IETF made itpopular for usage over the Internet especially in instantmessaging and chatting platforms It is however susceptibleto spamming attacks and its failure to have worldwidesupport has made Google too withdraw its support of thestandard [17] e protocol is mainly useable in the client-server architecture but can also be applicable in client-clientXMPP communication [53] Among the application layerprotocols XMPP is the only one that supports both asyn-chronous (publishsubscribe) and synchronous (requestresponse) services while running over TCP but can still runover other protocols

Given the simplicity of its implementation and ease ofusability in XMPP instant messaging (IM) is suitable forXMPP and is implemented using the message stanzaltmessagegt ese stanzas are usable in one-to-one ormultiuser sessions or chatting or in error reporting andsending headlines Attribute type of the stanza denes themessage type e attribute type could be a chat group chatheadline normal or error An instant messaging network inXMPP can support many XMPP clients and servers with agateway congured to enable other IM network users toconnect to other XMPP users e XMPP server mainlycontrols sessions and XML streams to clients and servers inaddition to routing XML stanzas in the streams [53] eTCP ports 5269 and 5222 have been respectively registeredfor the XMPP server-to-server connections and XMPP clientconnections by the Internet Assigned Numbers Authority(IANA) XMPP enables specication of XMPP ExtensionProtocols (XEP) to enhance its usefulness [17]

XMPP security clients connect to the server usingtransport layer security (TLS) and simple authentication andsecurity layer (SASL) protocols which are at the same timeused by the servers for interdomain communications Fordata condentiality and integrity TLS will encrypt thestreams of XML for all sessions To conrm the identity of anyclient that needs to access the server the SASL protocol doesthe authentication of the XML stream For any communi-cating session with a server to start the client needs to resolvethe serverrsquos DNS hostname For protection of the server of theclient from third-party attacks the server maintains the au-thentication details secret ie IP address and access methodand only original connections of the server are required

In server-to-server communications the SASL protocolis employed for authentication and condentiality Server-to-server communications may as well be deactivated from aserver by an administrator depending on the organizationalpolicy To guard against domain-spoong attacks serverdialback can be a good solution since servers support itsecuring more the XML stanzas Nevertheless it cannot beused to authenticate secure or encrypt in-server streamsand the server identication resulting therein is not strongenough to be relied on It further cannot secure against DNSpoisoning attacks and IP session hijacking for remote do-mains [53] High-security domains can use TLS and SASL

Advantages

(i) It is popular for usage over the Internet especially ininstant messaging and chatting platforms

Disadvantages

(i) It is susceptible to spamming attacks(ii) It has failed to have worldwide support making

Google withdraw support of the standard(iii) It is not secure against DNS poisoning attacks and

IP session hijacking for remote domains

36 Advanced Message Queuing Protocol (AMQP) isprotocol can be utilized by a number of platforms formessage exchange since it is open source and written inmany languages It is a publishsubscribe model depending

MQTT-SNclients Broker

MQTT

(a)

MQTT-SNclients Broker

(b)

Figure 6 (a) Transparent gateway and (b) aggregating gateway

Journal of Computer Networks and Communications 21

on a reliable and efficient messaging queue with capacity forinstant messaging or e-mail for different applications es-pecially in IoT environments It differs from others in thesense that message types can be specified and have theirsource traceable and how to do trade-off between perfor-mance security and reliability For reliability of messagetransmission AMQPmakes use of some delivery guaranteeslike at most once at least once and exactly once It can alsohowever utilize the TCP transport layer for guaranteeingreliability -e AMQPrsquos publishsubscribe method has twocomponents namely (i) exchange queue meant for routingof messages to an appropriate order in the queue and (ii)message queue that acts as storage pending their delivery toreceivers [16] In this protocol model the broker can makerouting decisions contrary to othermessaging systems whereapps using the queue have in them embedded the decision-making logic for recipients or senders -e developer mayneed to check each of the affected apps so as to change thelogic for routing and delivery of the messages [54]

Advantages

(i) It is convenient(ii) It has capacity for interoperability between dif-

ferent vendors(iii) It can use open standards to enable connection

between business partners(iv) It can enable innovations based on AMQP foun-

dations [55](v) It can enable supporting of mission-critical ap-

plications in e-commerce(vi) It can guarantee reliable delivery of messages

Disadvantages

(i) It is not very appropriate for real-timeapplications

(ii) It may not offer automatic discovery(iii) Interoperability is not guaranteed(iv) It lacks open-source libraries for constrained

gadgets [49]

37 Data Distribution Service (DDS) A product of theObject Management Group (OMG) DDS is a pubsubprotocol meant for M2M communications with two sub-layers -ese are the data-centric pubsub sublayer re-sponsible for connecting anonymous data publishers tosubscribers and the data-local reconstruction sublayer that isoptional and integrates DDS into the application layer [56]-e DDS architecture connects system components asshown in Figure 7 Publishers and subscribers are discon-nected with respect to (i) time (ii) space (iii) flow andbehavior (iv) platforms and (v) programming languages[57] A DDS data publisher generates topics and publisher-subscriber coupling is represented as topic name data typeschema and publishers and subscribersrsquo QoS attributesDelivering of data is done by the publisher layer -e datawriter and publishers together make a decision about anynecessary alterations for subsequent forwarding to thesubscribers who need to send data to IoT applications Data

readers are responsible for delivering topics to the sub-scribers after reading through them [56]

Some of the major components of the DDS architectureinclude the following [57]

(i) Topic this is a logical path specifying publicationand subscriptionrsquos data type between DataWritersand DataReaders For a successful session to begintopic names types and DataWriters and Data-Readersrsquo QoS should correspond

(ii) Domain this is a logical setting for transmissionthat can be utilized for separating and optimizingnetwork transmission in a setup of similar appli-cations If DDS apps share a domain ID then theycan exchange data among themselves

(iii) DataWriter and DataReader these are terminalobjects for writing and reading messages re-spectively to and from a universal data space

(iv) Participant this is an object that stands for apublisher or subscriber of the DDS app in a do-main acting like a container for other objects

Advantages

(i) It is highly reliable and offers QoS support(ii) It is highly scalable(iii) It is appropriate for real-time apps(iv) It is fault tolerant(v) It has high level of interoperability(vi) It does not need brokers that may act as

bottlenecks(vii) It offers automatic discovery(viii) It has a user-defined data structure for topic(ix) It supports publisher and subscriber decoupling

Disadvantages

(i) It is originally developed for only stand-aloneLANs

(ii) A lot of memory is needed(iii) It has no open-source libraries to support con-

strained devices(iv) It is hard to design and configure compared to

other protocols [49]

38 RESTful Services REpresentational State Transfer (REST)-is is an architectural style based on a set of principles thatdefine and address networked resources It offers web servicesthat enable transmission from the HTTP device to the devicewithin the IoT architecture and enables clients to reach theserver [16]-is protocol utilizes HTTP commands to backupthe requestresponse messaging model HTTP is a popularand secure protocol in the WWW using TLSSSL [16] ARESTful application has the following characteristics

(i) State and functionality are separated into distrib-uted resources

(ii) All resources are distinctively named by the use ofunique HTTP commands (GET POST PUT orDELETE over the Internet)

22 Journal of Computer Networks and Communications

(iii) It is a clientserver stateless layered protocol thatsupports caching REST uses the same methods asHTTP for all requests or responses with POST andGET for creating and retrieving resources PUT forupdating and changing the state of resources andDELETE for removing resourcesAdvantages of REST over arbitrary web services(SOAP) in the IoT context

(i) It reduces overhead(ii) It reduces parsing complexity(iii) It is stateless(iii) It integrates more tightly with HTTP [58](iv) It is easy to implement and learnis together

with its natural t over HTTP makes it apreferred method for Web 20 applications toshowcase their data and secure M2M sessionsin the IoT

(v) Applications that support RESTful web ser-vices exhibit better performance than the re-source-constrained WSN nodes [59]

(vi) REST is important for the IoT since it supportscommercial M2M cloud platforms

(vii) Its implementation in smart phones and tabletapps is also easy since it needs only an HTTPlibrary that is accessible to all operating systemplatforms [56]

Disadvantages

(i) Although it enjoys wide usage in commercialM2M platforms it is not likely to dominatebecause it is hard to implement

(ii) Its HTTP usage makes it incompatible withconstrained M2M devices leaving its usage toend applications

39 WebSockets It is a web-based protocol operating on asingle TCP channel and providing full-duplex transmissions

It permits secure duplex communication between a clientexecuting unreliable code in a restricted environment and adistant node communicating from that code while using theorigin-based security model employed by web browsers overTCP [60] e protocol is not a requestresponse model andis not a publishsubscribe model So unlike other protocolsa web-socket session is established by the client initiating ahandshake to the server to launch the session is onceestablished a full-duplex client-to-server connection beginsasynchronously and runs until terminated by either party[56] It gives a model for browser-based apps utilizing bi-directional communication with no need of many HTTPconnections to servers

WebSocket comprises two parts ie (i) the handshakehaving a clientrsquos message and the serverrsquos handshake responseand (ii) the data transfer HTTP headers should be repeated byapplications in all client requests and server responses arisingfrom the endless polling which might increase the commu-nication overhead subject to the application [61] WebSocketcan be used to come up with scalable real-time web appli-cations since it can enable full-duplex transmission via onesocket across the Web Its security can be ensured with thehelp of TLSSSL is protocol is suitable for resource-con-strained devices and IoTapplications will not have a good userexperience with it since it operates a clientserver architectureNevertheless it can outcompete all TCP-based protocolsbecause of the following advantages

(i) It is t for real-time transmission(ii) It has improved security(iii) It minimizes transmission overhead(iv) It can provide ecient messaging systems using

WebSocket ApplicationMessaging Protocol (WAMP)[17]

310 Streaming (Simple) Text-Oriented Messaging Protocol(STOMP) is is a TCP-based protocol designed for textmessage-oriented middleware (MOM) with the use of

Topic

Datareader

Datareader

Datareader

Datawriter

Datawriter

Datawriter

Subscriber PublisherPublisher

Data bus (DDS domain)

Subscriber

Topic Topic

Figure 7 DDS architecture

Journal of Computer Networks and Communications 23

HTTP-like commands for interoperability among platformslanguages and brokers Data communication takes placebetween the client and the broker in line with commands(like CONNECT DISCONNECT ACK NACK SUB-SCRIBE UNSUBSCRIBE SEND BEGIN COMMIT orABORT) then headers in the form ltkeygt ltvaluegt (one perline) then a blank line and then body content terminatingin a null character Server-client communication takes placeover a MESSAGE RECEIPT or ERROR frame with headerand body content format being analogous MOM productssupporting STOMP include Apache ActiveMQ or FuseMessage Broker HornetQ Open Message Queue(OpenMQ) RabbitMQ and syslog-ng through its STOMPdestination plugin [62] In addition to these other appli-cation layer protocols include simple media control protocol(SMCP) lightweight local automation protocol (LLAP)simple sensor interface (SSI) lightweight M2M (LWM2M)XMPP-IOT and simple object access protocol (SOAP)

Advantages

(i) It is lightweight for constrained networks(ii) It has flexibility to choose quality of services with

the given functionality(iii) It is standardized by OASIS Technical Committee(iv) It is easy and quick to implement

Challenges and opportunities

Here security threats are imminent because of the open-endedness of the application layer [63] -ese could includethe following

(i) -ere is poor or lack of a security design for anapplicationrsquos function Coming up with a gooddesign to address this could be a good ideaopportunity

(ii) Some of the programs might have backdoorsevading otherwise secure controls to allow un-authorized access to network resources We requiremechanisms to patch up such loopholes

(iii) Applications that lack or have weak authenticationmechanisms can be easily targeted by unlicensedusers to abuse the network Such mechanisms needto be strengthened further

(iv) Excessively complicated access control rules maybecause of their misunderstanding or poor writinggive unauthorized access allowing dubious activityrather than protecting the network they purport toprotect -ese should be simple and clear to un-derstand by all users

(v) In booting workstations without disks and managingnetwork devices we usually use the TFTP protocolalthough access does not require authenticating theuser by usernamepassword -is makes an intruderable to gain easy access to configaccess info as longas he can guess the filenames -is needs to be fixed

A few steps can be used to secure the application layerTo protect network data transmitted across the network

applications need to strengthen their encryption and au-thentication mechanisms Apps too need to enable strictcontrols of privileges for data access using good mechanismsto balance between usability and effectiveness [63] Appli-cations handling sensitive data need detailed logging andaudit capability as well as testing and reviewing

Table 3 summarizes the application layer protocols dis-cussed with their weaknesses and differences among them

4 Cross-Layer QoS Strategies

41 Service-Differentiated Real-Time Communication Scheme(SDRCS) -is is an event-driven routing protocol that routesreal-time traffic by featuring a cross-layer packet-forwardingdesign in which it integrates the real-time routing functionalitywith a new priority-based MAC scheme -e protocol ap-proximates distributed packet traversal speed for traffic clas-sification and admission control on the basis of this design Itfurther localizes decision-making by prioritizing packet for-warding to maximize the speed of packets [64]

Advantages

(i) It improves bandwidth utilization bandwidthdegradation due to unschedulable packets can beavoided

(ii) It offers a lightweight packet schedulability estima-tion mechanism Here it uses received signalstrength and admission control and then earlymissed-deadline packet-dropping policies are made

(iii) It delivers high end-to-end throughput via highersource data rates alongside strict end-to-end latencyneeds

(iv) Nodes which meet real-time requirements of lesstraffic load and better channel quality are givenhighest priority when it comes to packet forwarding

-e SDRCS operates via 5 mechanisms namely (1) RSS-(Rich Site SummaryReally Simple Syndication-) basedgrouping (2) admission control (3) prioritized queuing(4) real-time MAC and (5) dynamic forwarding -isprotocol has a queuing policy that employs per-hopdeadline-based queues and the nodes schedule theirpackets using FIFO priority-based queues [1] -e prin-cipal features of this protocol are mentioned as follows

(i) It is event driven and is easily adaptable to anynetwork changes

(ii) It does not need additional hardware for locali-zation or multichannel transmission

(iii) -e cross-layer packet-forwarding strategy can beused when transmitting multimedia traffic

(iv) It is able to efficiently circumvent any voids therein

Early deadline miss (EDM) policy drops all unscheduledpackets -e SDRCS circumvents wrong packet dropsdue to EDM since it adjusts to network changes easily

42 Sensor Fuzzy-Based Image Transport (SUIT) -is pro-tocol regulates congestion using a scheme based on fuzzy

24 Journal of Computer Networks and Communications

logic It transmits packets with reduced quality to allowablelevels if there is congestion enabling it to transmit lower-quality packetsframes without being dropped-e numberof packets delivered per second thus rises and while thisenhances video streaming apps it is more appropriatelymade for those that quickly transmit non-real-time videosuch as in video surveillance SUIT can stream JPEG pic-tures other than the traditional predictive video codingmethods that are not suitable for sensor devices because oftheir constraints of energy memory and computationalspeed Progressive JPEG-PJPEG performs better via packettransmission rate and delay though it degrades the imagequality because of congestion [65] -is protocol employscross-layer methods for interlayer data exchange though itstill employs some layering making it still possible topreserve all advantages of the layered design since it hasapplication and transport layers but has noMAC or routinglayer

For congestion detection and control SUIT bases on thepercentage number of in-bound and out-bound frames perwindow number of contenders and buffer occupancy of thenext-hop node Since buffer management has an effect oneffectiveness and QoS inWMSNs it is useful in the transportlayer so the protocol prioritizes packets for better QoS SUITreorders packets by tracking the sequence numbers of thedelivered packets since they all have a source ID framenumber and sequence number that it uses to detect anymissing packets -e protocol however does not providepacket reliability [65]

Advantages

(i) -e protocol prioritizes packets for better QoS(ii) It can transmit lower-quality packetsframes

without being dropped

Disadvantages

(i) It does not provide packet reliability

43 Network Layer QoS Support Enforced by a Cross-LayerController (NLQS) -is scheme permits packet-level servicedifferentiation as a function of throughput end-to-endpacket error rate and delay [66] (see Figure 8) -is en-hances network layer QoS It comprises a cross-layer controlunit (XLCU) for configuring and controlling networkingfunctionalities at physical MAC and network layers [18]-is depends upon the unified logic which affects choices forapplication layer needs and status of functional blocks whichdo implementation of networking functions [66]

Advantages

(i) Cross-layer interactions can be controlled withoutweakening the upgradability modularity andsimplicity of designing the system

44 Cross-Layer Signaling Shortcuts (CLASS) CLASS has ahigh level of effectiveness flexibility and comprehensive-ness According to [67] it has the following features

(i) Direct signaling among remote layers leading tofaster transmission

(ii) Lightweighted internal message format standard-ized protocols are not suitable for use in internalsignaling since they are not usually lightweight forinstance transmitting against faults in the networkCLASS requires only three fields namely destina-tion address event type and event contents

(iii) -e setup of its exterior messages is standardconsidering external signaling ICMP and TCPIPare suitable for general messages and short notifi-cations respectively

(iv) Message control protocol should ensure optimizedand organized exchange of dense simultaneousmessages across layers for high efficiency andavoidance of possible conflicts -e protocol appliesto many cross-layer signaling situations

Table 3 Differences and problems with application layer protocols

Protocol Architecture Transport QoS options Security Weaknesses

CoAP Requestresponse UDP DTLS No in-built security features DTLS does not supportmulticast

MQTT Publishsubscribe TCP TLSSSL Insufficient security at the protocol level sincepayload values are not encrypted

SMQTT Publishsubscribe TCP CPKP-ABE Key revocation and group pubsub for distributedSMQTT are still a challenge

MQTT-SN Publishsubscribe UDP TLSSSL SSLTLS suffers from attacks like BEAST CRIMERC4 and Heartbleed

XMPP RequestresponsePublishsubscribe TCP X TLSSSL Susceptible to spamming attacks and lacks worldwide

supportAMQP Publishsubscribe TCP TLSSSL Inappropriate for real-time applications

DDS Publishsubscribe TCPUDP TLSSSL Memory intensive and no open-source libraries forconstrained devices

REST Requestresponse HTTP X HTTPS Hard to implement Uses HTTP so incompatiblewith constrained apps

WebSockets PublishsubscribeClientserver TCP X TLSSSL Bad user experience for IoT apps since it runs the

clientserver architectureSTOMP Clientserver TCP HTTP -e broker can act as a bottleneck

Journal of Computer Networks and Communications 25

Advantages

(i) e protocol is scalable and has a very lowpropagation delay

(ii) It is very ecient and exible since it uses directsignaling between any two layers

(iii) It is very comprehensive and eective

45 Cross-Layer and Multipath-Based Video Transmission(CMVT) CMVT is a hybrid of application and networklayers At the application layer it carries out the encoding ofvideo streams to video data frames (I- P- and B-frames) bymeans of the MPEG-4 encoding format At the networklayer route discovery and data transfer take place Twoalgorithms greedy forwarding and rollback are used todiscover a number of paths from the source to the sink nodevia the route discovery method [20] A specied node icalculates the evaluation of its adjacent node j using thefollowing equation

fij (1 minus α)d2(j D) minus d2min(i)d2max(i) minus d

2min(i)

+ αeinit(j) minus eres(j)

einit(j)

(19)

where fij evaluation value for the node i to j d2(j D) isthe distance from the node j to the destination node D

d2min(i) and d2max(i) are the minimum and maximum dis-

tances of neighbors of the node i to D respectively einit(j) isthe initial energy of the node j eres(j) is the current residualenergy of the node j and α is the energy coecient given asfollows

α emax(i) minus emin(i)

emax(i) (20)

where emax(i) and emin(i) are respectively the maximumand minimum energy left of all node irsquos neighbors

e network layer further transmits video streams withCMVT doing status evaluation choosing an appropriatecommunication path for any packets and the QoS guaranteelevel for the path i is calculated as

fi (1 minus ω)hisum hi

+ ωnisum ni

(21)

where fi is the evaluation value for the path i hi are the hopsfor the path i ni is the summation of packets sent via isum ni isthe summation of packets sent by sources and ω is theenergy consumption factor [18]

Advantages

(i) e CMVT protocol is very superior in mediatransmission particularly large-scale WMSNs

To network peers

Flow requirements

-

Atimes

Admissioncontrol

Capacityassignment

Next-hopselection

Coding rate

Data rate

QoScontracts

Cross-layer controlunit (XLCU)

Localization subsystem

Synchronizationsubsystem

Networkcoordinationunit (NCU)

UWB modulator

Channel coder

QoS scheduler

Georouting

Source coder

Rate control

Figure 8 Cross-layer controller architecture

26 Journal of Computer Networks and Communications

46 Cross-Layer Cooperative MAC (CoopMAC)CoopMAC [68] is based on the IEEE 80211 DCF approach-ere is a reduction in interference of neighboring cellsleading to uniform coverage in densely deployed networks[68] A connection is established by two-way authenticationbetween the sender and the receiver When a Clear-to-Send(CTS) packet and short interframe space (SIFS) are obtainedby the sender packets are transmitted right to the receiverwith no valuable cooperation If available the two nodes findout whether they can exchange data -is is done by the useof a helper identification signal to establish how feasiblecooperative communication is It is established in thepresence of a signal and if it is not available direct com-munication is activated For distributed systems helper-initiated cooperation is preferable because of the presence ofthe RTSCTS packets [69]

Advantages

(i) CoopMAC has an advantage over others due to thespatial diversity amongst the three nodes [70]

(ii) It has the potential to achieve significantthroughput and delay with no complexity in thedesign of the system

47 Cooperative MAC Protocol for Multihop Networks (M-CMAC) Just as CoopMAC M-CMAC is made in such away that high data rate stations help low data rate stations toforward traffic for broadcasting [71] -ere are also helperschosen in such a way that 2 fast-hop transmissions replace aslow hop transmission -e helper via whom there existsleast delay from the sender to the receiver is considered bestbecause of theminimum two-hop transmission rate and so isthe chosen neighbor to the duo Here the working as-sumption is that all nodes have their position coordinatesknown and thus the Euclidean distances between all pairscan be calculated and converted to the data rate for that linkAll nodes have cooperative tables (CT) of potential helperswith destination and helper MAC address Euclidean dis-tance and total distance through the helper Any sender withdata to transmit checks out for the existence of any helper inthe CT for the destination and if available forwards RTS forchannel reservation for single-hop duration M-CMAC hasgot an RTS format having a shape with five fields similar tothe one given as follows

Framecontrol Duration Source

addressDestination

addressHelperaddress

-e destination nodersquos address is kept by the source in thepart for the helper address that is in turn saved as the desti-nation address Nodes inspect helper and destination addressfields on receipt of RTS whereby the node acts as the othernodersquos helper in case the helper address field is not the sameand the helper transmits CTS back to the source if at all it wantsto forward data On delivery of the CTS packet the sourcetransmits to the helper which in turn sends the packets to thedestination that also sends ACK to the helper on receipt of thepacket In M-CMAC there exists a higher level of channelreuse (parallel transmissions) due to the increased number of

nodes therein that subsequently raises availability of helpers fordata forwarding -is results in increased throughput incomparison with CoopMAC and IEEE 80211 DCF [71]

Advantages

(i) It promises a higher throughput compared toCoopMAC and IEEE 80211 DCF

48 Cluster-Based Cooperative Routing (CBCR) Protocol-e CBCR protocol has a multihop data-forwarding func-tion realized at the link layer with cooperative links that useM-CMAC -is protocol encompasses two stages namelythe routing relay selection phase and data forwarding phase

(i) Routing relay selection phase all nodes announcetheir presence to their neighbors by broadcastingperiodical beacon messages that carry the MAC ad-dress of the node Each of these constructs a relaytable that contains all neighboring nodes with whichit is able to communicate -e node further broad-casts its neighbor list in case of any change to itsentries since the previous broadcast -e MAC ad-dresses of nodes next to the node X are contained inits relay tablersquos column one and the neighbor nodersquosrow has MAC addresses for neighbors of the adjacentnode Based on its relay table each node in-dependently chooses routing relays Choosing a nodeas a relay node depends on the number of nodes itconnectsmdashit should be the highest number [71]

(ii) Data forwarding phase a node with packets totransmit needs to first verify if the receiver is in thesame cluster and if that receiver has a helper to whichthe packet is thence forwarded but in whose absencethe packet is delivered right to the receiver In casethe intended receiver is in a different cluster therelay table will be checked to see whether that re-ceiver is reachable via other routing relays If it isreachable packets will be sent to the routing relaydirectly or via the helper if at all the relay has one Incase of destination unreachable via relays packetsare multicast to all relays by the node [71]Advantages

(i) It enables multicasting

49 MAC-PHY Cross-Layer Protocol A cooperative cross-layer standard for cooperation at the physical layer in next-generation WMSNs is developed in [70] It provides acomplete MAC layer algorithm that gives a supportive shellfor the PHY-MAC layer Like CoopMAC its scheme de-pends on the middle node for communication betweennodes-eMAC layer protocol has been changed to regulateinformation exchanges at the physical layer -e receivernode receives duplicate packets from source and helpernodes to decode the data [70]

410 MAC-Centric Approach -is cross-layer protocoltargets multimedia applications by using the MPEG-4

Journal of Computer Networks and Communications 27

scheme [21] It is characterized by 4 access categories(AC3ndashAC0) according to their priority in transmission It ismeant for supporting different QoS needs in upcomingvideo applications and enabling MAC layer differentiationfor H264 partitioning [18] Because of low bandwidth delayand other QoS challenges that cause inefficiency in trans-mittingmultimedia data inWMSNs a number of algorithmsdepending on IEEE 80211e have been proposed to supporttransmission of quality videos [72] Selecting an AC dependson the measures for QoS such as the loss rate and delay-us AC3 having the highest priority is mapped to theparameter set concept since the stream is sensitive to bit lossespecially in video transmission [21]

411 Adaptive Cross-Layer Forward Error Correction(ACFEC) In the ACFEC [73] model data packets are ex-changed among nodes through the access point (AP) whichoperates the infrastructure mode in which it adds FEC to videodata [72] -ese data are dealt by encapsulation using astreaming server to the receiver through thewireless AP as RTPpackets An adaptive FEC controller senses the category ofpackets out of the RTP header and recovers the header ofpackets from UDP -e encoder generates some error-cor-recting packets whose number is determined by the sourcepacket number of the block Multimedia transmissions aremonitored by the controller using MAC failure data and itscounter is increased by one if there is a failed transmission-econtroller uses the failure counter to change the packet numberproduced after transmitting a block [18] In case there are lostpackets it changes the redundancy rates and produces extrapackets to replace the lost ones and satisfy the requirements ofthe receiving node [72] -e FEC packet number is enhancedor reduced to satisfy the receiverrsquos requirements and stoppacket losses-is is done through accurate detection of packetlosses and adjustment of redundancy rates If all video datapackets are well received there will not be generation of FECpackets [73]

Advantages

(i) It promises a good QoS through packet loss re-duction and redundancy rate adjustment

412 Balanced Cross-Layer Fuzzy Logic (BCFL) DesignRouting Algorithm in WSNs A new fuzzy logic-basedrouting algorithm (BCFL) [74] was designed using disper-sion of the cross-layer parameter as the fuzzy logic inferencesystem input Each cross-layer parameter has a dynamicweight depending on the value of dispersion -e designcomes with some innovations as follows (i) for fuzzy logicinference system input the absolute parameter value issubstituted by parameter dispersion thereby significantlyreducing algorithmic complexity (ii) dispersion does notchange with the order of magnitude according to the dis-persion formula and (iii) the weight of the parameter de-pends on the size of its dispersion and the two are inverselyproportional to each other -is enables BCFL to have somenobility unlike other algorithms as per the following dis-tinguishing properties making it advantageous

(i) It has got simple if-then rules which remain con-stant even when the constraints increase

(ii) It is capable of dealing with many constraints withno increment in complexity

(iii) It has capacity to yield a more balanced solutionthan other algorithms

(iv) It is easily adaptable to changes in network con-ditions and topology even when the changes arefrequent like in underwater WSNs -e algorithm isuseable in choosing the CH in cluster-based routingprotocols [74]

413 Minimum Hop Disjoint Multipath Routing Algorithmwith Time Slice Load-Balancing Congestion Control SchemeMHDMwTS is a two-phased routing protocol comprisingpath build-up and path acknowledgment phases It consists ofmultiple sources each of which has three build-up disjointpaths namely primary alternate and back-up paths Onactivation the source node requests to build up a route to thenearest hop neighbor which is the path build-up phase [19] Inthis phase step one has source activation in which therequested node for path building adds its number and time-stamp and sends to the least hop-count neighboring node-isgoes on until the least time latency sink having the informationneeded to construct the primary route is reached [75] In steptwo there is path extraction when the new package fromanother path arrives whereby the extracted path is comparedwith the primary path If the node is shared the package will berejected or an alternative path will be searched for to get abackup through comparison of the preceding two paths Phasetwo is path acknowledgment where we have the third step inwhich the sink returns the ACK packet to the sender with pathinformation having nodes and their time information after ithas computed it using the timestamp [1819]

Advantages

(i) -e protocol minimizes end-to-end latency(ii) It enhances congestion control

414 Cross-Layer Optimal Design (CLOD) Authors in [76]come up with a CLOD for scheduling at the data link layerrouting at the network layer and controlling congestion atthe transport layer with an assumption of fixed link capacity-rough congestion control energy efficiency is improvedTransport layer congestion at the nodes is minimized bycompressed sensing (CS) in which transmitted bits are re-duced whereas optimally allocating resources decreasescongestion on the links at the data link layer CLODpromises minimized computational complexity and betterperformance in light traffic scenarios [77] Generating andstoring CLOD is more efficient compared to Gaussianrandom matrices It prolongs network lifetime and savesenergy

Advantages

(i) It does congestion control subsequently increasingenergy efficiency

28 Journal of Computer Networks and Communications

(ii) It reduces computational complexity(iii) It improves performance in light traffic scenarios

415 Challenges and Opportunities of the Cross-Layer Design(CLD)

(i) Different layers in a real network assume differentfunctions andor services A layer only communi-cates with its neighboring layers -e layered modeldegrades system performance because of many

features in wireless transmission QoS parameters inthe layers of WMSNs are shared by routing pro-tocols to optimize the performance Neverthelessthere is a need to design cross-layer models forincreased efficiency in routing [1]

(ii) -e physical layer plays a highly vital role in CLDRate adaptation and channel allocation take place atthe physical layer via signal processing to enhanceQoS End-to-end performance in wireless media isaffected by the changes therein given the effect on

Table 4 A summary of some of the reviewed cross-layer models

Protocol Layers Aims Comments QoS parameters

SDRCS MACPHY Routing real-time traffic Transmits multimedia traffic viacross-layer packet forwarding

-roughputlatency

SUIT Applicationtransport Transmission of non-real-time video Uses fuzzy logic to regulate

congestionTransmission rate

delay

CLASS Any twoDesign serves as a framework fordifferent implementations ofdifferent application scenarios

Scalable very efficient flexible andhas very low propagation delay

Propagation delayjitter

NLQS PHYMACnetwork

Permits packet-level servicedifferentiation as a function of

throughput packet error rate anddelay

Network layer QoS is enhanced -roughput delaypacket error rate

CMVT Applicationnetwork

Encodes video streams to video dataframes via MPEG-4 encoding anddoes route discovery and data

transfer

Greedy forwarding and rollback areused to find source-to-sink paths Energy

CoopMAC PHYMAC Offers spatial diversity among thethree nodes

Helper ID signal finds out howfeasible cooperative communication

is

-roughput ratedelay

M-CMAC PHYMAC Increases end-to-end throughputand packet delivery ratio

Euclidean distances between nodesare calculated and converted to the

data rate for a link

-roughputpacket delivery

ratio

CBCR PHYMACMinimizes control overhead andtime consumed in establishing thecooperative paths than M-CMAC

Energy consumption is moreuniformly distributed in a network

enhancing network lifetime

-roughputpacket deliveryratio energy

MAC-PHY MACPHYMeant to enable PHY layer

cooperation and maximize gains ofcooperation at the MAC layer

Leverages both spatial diversity andcoding gain -roughput delay

MAC-centric MACAPPL

Meant to support QoS needs in newvideo apps and enable MAC layer

differentiation for H264partitioning

Targets multimedia applications byusing the MPEG-4 scheme

Delay packet lossrate

ACFEC MACnetwork(UDP)

Meant to enhance the quality ofvideo streaming over 80211WLANs

and overcome packet losses

Adjusts redundancy rates toovercome channel fluctuations and

detect and reduce packet lossPacket loss rate

MHDMwTS mdash

Meant to provide reliable datatransfer with the multipath routingand load-balancing congestioncontrol method in WMSNs

More reliable than basic routingschemes for transport multimedia

data

Latency packagetransmit rate

BCFL mdash

Introduces dispersion into fuzzylogic-based routing and sets every

cross-layer parameter with adynamic weight

Can be used to select a CH in cluster-based routing protocols andproposes a dispersion formula

Node utilitydispersion

CLODDatalinknetworktransport

Prolongs network lifetime andachieves congestion control anddesigned for lightly loaded WSNs

Assumes fixed link capacity andintegrates compressed sensing

technology

-roughputaverage energy CS

error ratio

Journal of Computer Networks and Communications 29

the purposefulness of the protocols at the networklayer CLD offers solutions to conserving powerminimizing energy and controlling both the flowand congestion in the network as well as fault tol-erance so it is an opportunity for designers con-sidering other layers We further desire to developCLDs that shall make an impact on network oper-ation and get close attention [18]

(iii) CLD interface standardization the architecturemust deliver module functionality although thereexist some queries amongst modules about possibleinterfaces predicted by the necessity to share in-formation among remote protocol layers Technicalchallenges include developing designing andstandardizing of cross-layer interfaces and algo-rithms that satisfy the cross-layer optimization re-quirements among protocol layers

Table 4 summarizes these cross-layer models with a fewcomments

5 Future Research Direction

51 At theMACLayer In WMSNs achieving duty cycling istricky since video traffic is volatile Listening to adjacentnodes wastes energy We need to switch states from awake tosleep -is requires further research

Self-organizing networks and green communicationalthough most CRWSN energy-saving algorithms aredesigned to reduce node transmission power energyconsumption is inevitable during the operational mode ofBSs because of internal processing Scholars are trying withpolicy-makers to encourage a shift to greenWSNs to reduceoperational costs and carbon footprints With self-orga-nizing CRWSNs for automatic switch-off when idle energyis saved which calls for cognitive switching algorithms Sointegrating cognitive switching with radio resource allo-cation in CRWSNs is a promising research area [78]

52 At the Application Layer Considering a WMSN ar-chitecture end users access the network via the BS to andfrom which they send their sensed data We desire to getguaranteed QoS and energy efficiency by designing a newarchitecture in the form of hardware and software whichneeds to be investigated further for a fix of the same [1]

QoS and energy efficiency are very crucial in WMSNsand mostly real-time apps that require guaranteed band-width and throughput in their network lifetime Mostprotocols ignore base station (BS) mobility and WMSNnodes Traffic management telemedicine and battlefieldsurveillance apps require mobile nodes or BSs so designingdynamic routing protocols to be adaptable in these cir-cumstances is necessary

53 Cross-Layer Security of data transmitted on WMSNseg in military surveillance and e-commerce is key withQoS and energy efficiency WMSNs are vulnerable to attacks

like worms and sinkhole Moreover their computationpower is not that high complicating implementation ofstrong secure protocols -is needs further attention

For fairness and priority issues in CRWSNs nodes areassigned diverse priorities depending on importance andurgency of their data For better network performance thereis a need to develop fair resource allocation mechanismswith priority-based fairness in sensors-ere is still potentialin this research area [78]

We must develop multiobjective adaptive protocols tooptimize QoS metrics involved in routing for best trans-mission results to get a trade-off between different opti-mization metrics We need to design standardized cross-layer algorithms which are able to meet cross-layer opti-mization standards

6 Conclusion

We have reviewed the different QoS strategies for WSNs inthe context of IoT from the MAC layer and application layeras well as the cross-layer paradigm For the MAC layer wehave reviewed protocols for WSNs such as the hybrid of thecontention-free and contention-based MAC protocols andthose for WMSNs We further reviewed a number of ap-plication layer protocols including many machine-to-ma-chine requestresponse and publishsubscribe protocols Forsystem optimization cross-layer QoS strategies are veryimportant in wireless communication We have reviewed anumber of cross-layer strategies For all categories reviewedchallenges and opportunities are discussed Finally somepossible future directions are discussed for research andapplication showing a promising potential for future re-search in this relatively new area especially as more WMSNapplications emerge recently

Conflicts of Interest

-e authors hereby declare no conflicts of interest regardingthe publication of this work

Acknowledgments

-is research was funded by the National Natural ScienceFoundation of China (Nos 61602398 61672447 and61711540306)

References

[1] V Bhandary A Malik and S Kumar ldquoRouting in wirelessmultimedia sensor networks a survey of existing protocolsand open research issuesrdquo Journal of Engineering vol 2016Article ID 9608757 27 pages 2016

[2] I F Akyildiz T Melodia and K R Chowdhury ldquoWirelessmultimedia sensor networks applications and testbedsrdquoProceedings of the IEEE vol 96 no 10 pp 1588ndash1605 2008

[3] M Iqbal M Naeem A Ahmed M Awais A Anpalagan andA Ahmad ldquoSwarm intelligence based resource managementfor cooperative cognitive radio network in smart hospitalsrdquoWireless Personal Communications vol 98 no 1 pp 571ndash5922018

30 Journal of Computer Networks and Communications

[4] S A Alvi B Afzal G A Shah L Atzori and W MahmoodldquoInternet of multimedia things vision and challengesrdquo AdHoc Networks vol 33 pp 87ndash111 2015

[5] Q F Hassan A U R Khan and S A Madani Internet ofings Challenges Advances and Applications CRC PressTaylor amp Francis Group A Chapman amp Hall Group BocaRaton FL USA 2018

[6] Q Wang Y Zhao W Wang et al ldquoMultimedia IoT systemsand applicationsrdquo in Proceedings of the 2017 Global Internet ofings Summit (GIoTS) vol 2 Geneva Switzerland June 2017

[7] A Aslam and E Curry ldquoTowards a generalized approach fordeep neural network based event processing for the internet ofmultimedia thingsrdquo IEEE Access vol 6 pp 25573ndash25587 2018

[8] M Jridi T Chapel V Dorez L Bougeant and A Le BotlanldquoSoC-based edge computing gateway in the context of theinternet of multimedia things experimental platformrdquoJournal of Low Power Electronics and Applications vol 8no 1 p 1 2018

[9] H Noura A Chehab L Sleem andM N Rapha ldquoOne roundcipher algorithm for multimedia IoT devicesrdquo MultimediaTools and Applications vol 77 no 14 pp 18383ndash18413 2018

[10] T Balan D Robu and F Sandu ldquoMultihoming for mobileinternet of multimedia thingsrdquo Mobile Information Systemsvol 2017 Article ID 6965028 16 pages 2017

[11] H Shen and G Bai ldquoRouting in wireless multimedia sensornetworks a survey and challenges aheadrdquo Journal of Networkand Computer Applications vol 71 pp 30ndash49 2016

[12] A Ahmad M H Rehmani H Tembine O A Mohammedand A Jamalipour ldquoIEEE access special section Editorial optimization for emerging wireless Networks IoT 5G andsmart grid communication networksrdquo IEEE Access vol 5pp 2096ndash2100 2017

[13] B A Muzakkari M A Mohamed M F A Kadir andZ Mohamad ldquoRecent advances in energy efficient-QoS awareMAC protocols for wireless sensor networksrdquo InternationalJournal of Advanced Computer Research vol 8 no 38pp 212ndash228 2018

[14] M A Yigitel O D Incel and C Ersoy ldquoQoS-aware MACprotocols for wireless sensor networks a surveyrdquo ComputerNetworks vol 55 no 8 pp 1982ndash2004 2011

[15] A M Abbas and O Kure ldquoQuality of Service in mobile ad hocnetworks a surveyrdquo International Journal of Ad Hoc andUbiquitous Computing vol 6 no 2 p 75 2010

[16] M Q Shatnawi ldquoApplication layer protocols for the internetof things a surveyrdquo in Proceedings of the InternationalConference on Engineering amp MIS (ICEMIS) Agadir Mo-rocco September 2016

[17] V Karagiannis P Chatzimisios F Vazquez-gallego andJ Alonso-zarate ldquoA survey on application layer protocols forthe internet of things research motivationrdquo InternationalJournal of Advanced Research in Computer Science vol 8no 3 pp 1ndash10 2015

[18] M S Bernard T Pei Z Li and K Li ldquoQoS strategies forwireless multimedia sensor networks in the context of IoTrdquoLecture Notes of the Institute for Computer Sciences SocialInformatics and Telecommunications Engineering vol 4pp 228ndash253 2019

[19] A AlAmri and M Abdullah ldquoCross-layer quality of serviceprotocols for wireless multimedia sensor networksrdquo in Pro-ceedings of the International Conference on CommunicationManagement and Information Technology pp 649ndash658Cosenza Italy April 2017

[20] J Guo L Sun and R Wang ldquoA cross-layer and multipathbased video transmission scheme for wireless multimedia

sensor networksrdquo Journal of Networks vol 7 no 9pp 1334ndash1340 2012

[21] A Ksentini A Naimi and M Gueroui ldquoToward an im-provement of H264 video transmission over IEEE 80211ethrough a cross-layer architecturerdquo IEEE CommunicationsMagazine vol 44 no 1 pp 107ndash114 2006

[22] W Ye J Heidemann and D Estrin ldquoAn energy-efficientMAC protocol for wireless sensor networksrdquo in Proceedings ofthe 21st International Annual Joint Conference of the IEEEComputer and Communications Societies (INFOCOM 2002)New York NY USA June 2002

[23] J Polastre J Hill and D Culler ldquoVersatile low power mediaaccess for wireless sensor networks categories and subjectdescriptorsrdquo in Proceedings of the Second InternationalConference on Embedded Networked Sensor Systems Sen-Sysrsquo04 pp 95ndash107 Baltimore MA USA November 2004

[24] T Danmanee K N Nakorn and K Rojviboonchai ldquoCU-MAC a duty-cycle MAC protocol for internet of things inwireless sensor networksrdquo ECTI Transactions on ElectricalEngineering Electronics and Communications vol 16 no 2pp 30ndash43 2018

[25] T Van Dam and K Langendoen ldquoAn adaptive energy-efficientMAC protocol for wireless sensor networksrdquo in Proceedings ofthe 1st International Conference on Embedded NetworkedSensor Systems Los Angeles CA USA November 2003

[26] P Xie and J Cui ldquoR-MAC an energy-efficient MAC protocolfor underwater sensor networksrdquo in International Conferenceon Wireless Algorithms Systems and Applications (WASA2007) Chicago IL USA August 2007

[27] A Rahim N Javaid M Aslam U Qasim and Z A KhanldquoAdaptive-reliable medium access control protocol forwireless body area networksrdquo in Proceedings of the AnnualIEEE Communications Society Conference on Sensor Meshand Ad Hoc Communications and Networks (SECON) SeoulKorea June 2012

[28] H S Kwak S Ullah D H Kwak and C H Lee ldquoA powerefficient MAC protocol for wireless body area networksrdquoJournal of the Korean Institute of Intelligent Systems vol 8no 6 pp 131ndash140 2009

[29] C Omeni A J Burdett and M Park ldquoEnergy efficientmedium access protocol for wireless medical body area sensornetworksrdquo in Proceedings of the 4th IEEEEMBS InternationalSummer School and Symposium on Medical Devices andBiosensors pp 29ndash32 Cambridge UK August 2007

[30] L Sitanayah C J Sreenan and K N Brown ldquoER-MAC ahybrid MAC protocol for emergency response wireless sensornetworksrdquo in Proceedings of the Fourth International Con-ference on Sensor Technologies and Applications pp 244ndash249Mestre Italy July 2010

[31] Y Sun and D B Johnson ldquoRI-MAC a receiver-initiatedasynchronous duty cycle MAC protocol for dynamic trafficloads in wireless sensor networksrdquo in Proceedings of the 6thACM conference on Embedded network sensorsystemsmdashSenSysrsquo08 pp 1ndash14 Raleigh NC USA November2008

[32] M Buettner G V Yee E Anderson and R Han ldquoX-MAC ashort preamble MAC protocol for duty-cycled wireless sensornetworksrdquo in Proceedings of the 4th International Conferenceon Embedded Networked Sensor Systems SenSys 2006pp 307ndash320 Boulder Colorado USA October 2006

[33] S Boulfekhar and M Benmohammed ldquoSynchronous receiverinitiated MAC protocol for long-lived sensor networksrdquoComputers amp Electrical Engineering vol 40 no 2 pp 504ndash516 2014

Journal of Computer Networks and Communications 31

[34] S Mehta and K S Kwak ldquoH-MAC a hybrid Mac protocol forwireless sensor networksrdquo International Journal of ComputerNetworks amp Communications vol 2 no 2 pp 108ndash117 2010

[35] I Rhee A Warrier M Aia J Min and M L Sichitiu ldquoZ-MAC a hybrid MAC for wireless sensor networksrdquo IEEEACM Transactions on Networking vol 16 no 3 pp 511ndash5242008

[36] S Arshad A Al-sadi and A Barnawi ldquoZ-MAC performanceevaluation and enhancementsrdquo Procedia Computer Sciencevol 21 pp 485ndash490 2013

[37] Y Liu I Elhanany and H Qi ldquoAn energy-efficient QoS-aware media access control protocol for wireless sensornetworksrdquo in Proceedings of the IEEE International Confer-ence on Mobile ad hoc and Sensor Systems pp 9ndash11 Wash-ington DC USA December 2005

[38] J Decotignie ldquoWiseMAC an ultra low power MAC protocolfor the downlink of infrastructure wireless sensor networksrdquoin Proceedings of the Ninth International Symposium onComputers And Communications pp 244ndash251 AlexandriaEgypt June 2004

[39] R A Rashid ldquoDevelopment of energy aware TDMA-basedMAC protocol for wireless sensor network systemrdquo EuropeanJournal of Scientific Research vol 30 no 4 pp 571ndash578 2009

[40] H P Van Hoesel ldquoA lightweight medium access protocol(LMAC) for wireless sensor networksrdquo in Proceedings of theInternational Workshop on Networked Sensing Systems Bal-timore MD USA November 2004

[41] C Li P Wang H Chen andM Guizani ldquoA cluster based on-demand multi-channel MAC protocol for wireless multi-media sensor networksrdquo in Proceedings of the IEEE In-ternational Conference on Communications pp 2371ndash2376Beijing China May 2008

[42] J L Sobrinho and A S Krishnakumar ldquoQuality-of-service inad hoc carrier sense multiple access wireless networksrdquo IEEEJournal on Selected Areas in Communications vol 17 no 8pp 1353ndash1368 1999

[43] J P John ldquoProviding MAC QoS for multimedia traffic in80211e based multi-hop ad hoc wireless networksrdquo ComputerNetworks vol 51 no 1 pp 153ndash176 2007

[44] S Lofty and N Padmavati ldquoA survey on mobility basedprotocols in WSNsrdquo in Proceedings of the InternationalConference Communication and Manufacturing pp 12ndash15Hangzhou China July 2014

[45] S Deng L Shen and J Li ldquoMobility-based clustering pro-tocol for wireless sensor networks with mobile nodesrdquo IETWireless Sensor Systems vol 1 no 1 pp 39ndash47 2011

[46] M Khalil A Khalid F U Khan and A Shabbir ldquoA review ofrouting protocol selection for wireless sensor networks insmart citiesrdquo in Proceedings of the 24th Asia-Pacific Confer-ence on Communications (APCC) Ningbo China November2014

[47] O Message Q Telemetry and T Mqtt Committee Specifi-cation Draft 02 April 2014

[48] M Collina M Bartolucci A Vanelli-coralli andG E Corazza ldquoInternet of things application layer protocolanalysis over error and delay prone linksrdquo in Proceedings ofthe 7th Advanced Satellite Multimedia Systems Conference andthe 13th Signal Processing for Space CommunicationsWorkshop Livorno Italy September 2014

[49] A Talaminos-barroso M A Estudillo-valderrama L M RoaJ Reina-tosina and F Ortega-ruiz ldquoA machine-to-machineprotocol benchmark for e health applicationsmdashuse case re-spiratory rehabilitationrdquo Computer Methods and Programs inBiomedicine vol 129 pp 1ndash11 2016

[50] V Goyal O Pandey A Sahai and B Waters ldquoAttribute-based encryption for fine-grained access control of encrypteddatardquo in Proceedings of the 13th ACM Conference on Com-puter and Communications Security pp 89ndash98 AlexandriaVA USA October 2006

[51] M Singh M A Rajan V L Shivraj and P BalamuralidharldquoSecure MQTT for internet of things (IoT)rdquo in Proceedings ofthe Fifth International Conference on Communication Systemsand Network Technologies Gwalior India April 2015

[52] A Stanford-Clark and H L Truong MQTT for Sensor Net-works (MQTT-SN) Protocol Specification httpmqttorgnewwp-contentuploads200906MQTT-SN_spec_v12pdf 2013

[53] RFC ldquoExtensible messaging and presence protocol (XMPP)rdquoin Proceedings of the IETF pp 1ndash9 San Diego CA USAAugust 2005

[54] S Vinoski and I Technologies ldquoQueuing protocolrdquo IEEEInternet Computing vol 10 no 6 pp 87ndash89 2006

[55] J L Fernandes I C Lopes J J P C Rodrigues and S UllahldquoPerformance evaluation of restful web services and AMQPprotocolrdquo in Proceedings of the Fifth International Conferenceon Ubiquitous and Future Networks (ICUFN) pp 2ndash7 DaNang Vietnam July 2013

[56] M Asim ldquoA survey on application layer protocols for internetof things (IoT)rdquo International Journal of Advanced Researchin Computer Science vol 8 no 3 pp 996ndash1000 2017

[57] K An A Gokhale D Schmidt S Tambe P Pazandak andG Pardo-castellote ldquoContent-based filtering discovery pro-tocol (CFDP) scalable and efficient OMG DDS discoveryprotocolrdquo in Proceedings of the 8th ACM InternationalConference on Distributed Event-Based Systems New YorkNY USA July 2014

[58] E Web What is a web service pp 52ndash57 2010[59] M Laine RESTful Web Services for the Internet of ings

pp 2ndash4 2012 httpmediaTkk[60] IETF -e WebSocket protocol pp 1ndash71 2011[61] V Pimentel ldquoCommunicating and displaying real-time data

with websocketrdquo IEEE Internet Computing vol 16 no 4pp 45ndash53 2012

[62] Wikipedia 2019 httpsenwikipediaorgwikiStreaming_Text_Oriented_Messaging_Protocol2019

[63] K Raghavendra and S Nireshwalya ldquoApplication layer se-curity issues and its solutionsrdquo IJCSET vol 2 no 6pp 1266ndash1269 2012

[64] Y Xue B Ramamurthy andM C Vuran ldquoSDRCS a service-differentiated real-time communication scheme for eventsensing in wireless sensor networksrdquo Computer Networksvol 55 no 15 pp 3287ndash3302 2011

[65] C Sonmez S Isik M Y Donmez O D Incel and C ErsoyldquoSUIT a cross layer image transport protocol with fuzzy logicbased congestion control for wireless multimedia sensornetworksrdquo in Proceedings of the 5th International Conferenceon New Technologies Mobility and Security (NTMS) pp 1ndash6IEEE Istanbul Turkey May 2012

[66] T Melodia and I Akyildiz ldquoCross-layer QoS-aware com-munication for ultra wide band wireless multimedia sensornetworksrdquo IEEE Journal on Selected Areas in Communica-tions vol 28 no 5 pp 653ndash663 2010

[67] Q Wang and M A Abu-Rgheff ldquoCross-layer signalling fornext-generation wireless systemsrdquo in Proceedings of the 2003IEEE Wireless Communications and Networking 2003WCNC 2003 vol 2 pp 1084ndash1089 IEEE New Orleans LAUSA March 2003

[68] P Liu Z Tao S Narayanan T Korakis and S PanwarldquoCoopMAC a cooperative MAC for wireless LANsrdquo IEEE

32 Journal of Computer Networks and Communications

Journal on Selected Areas in Communications vol 25 no 2pp 340ndash354 2007

[69] H Shan H T Cheng and W Zhuang ldquoCross-layer co-operative MAC protocol in distributed wireless networksrdquoIEEE Transactions onWireless Communications vol 10 no 8pp 2603ndash2615 2011

[70] F Liu T Korakis Z Tao and S Panwar ldquoAMAC-PHY cross-layer protocol for wireless Ad-Hoc networksrdquo in Proceedingsof the 2008 IEEE Wireless Communications and NetworkingConference pp 1792ndash1797 IEEE Las Vegas NV USA March2008

[71] L Jacob and H R Shamna ldquoEfficient cooperative MAC androuting in wireless networksrdquo Transactions on Networks andCommunications vol 3 no 5 2015

[72] S Rao and K Shama ldquoCross layer protocols for multimediatransmission in wireless networksrdquo International Journal ofComputer Science amp Engineering Survey vol 3 no 3pp 15ndash28 2012

[73] L Han S Park S Kang and H Peter ldquoAn adaptive cross-layer FEC mechanism for video transmission over 80211WLANsrdquo in Proceedings of the 2009 International Conferenceon Internet pp 209ndash215 Las Vegas NV USA December2009

[74] N Li J Martınez and V H Dıaz ldquo-e balanced cross-layerdesign routing algorithm in wireless sensor networks usingfuzzy logicrdquo Sensors vol 15 no 8 pp 19541ndash19559 2015

[75] G Sun J Qi Z Zang and Q Xu ldquoA reliable multipathrouting algorithm with related congestion control scheme inwireless multimedia sensor networksrdquo in Proceedings of the2011 3rd International Conference on Computer Research andDevelopment vol 4 pp 229ndash233 IEEE Shanghai ChinaMarch 2011

[76] M Li Y Jing and C Li ldquoA robust and efficient cross-layeroptimal design in wireless sensor networksrdquoWireless PersonalCommunications vol 72 no 4 pp 1889ndash1902 2013

[77] J Yan M Zhou and Z Ding ldquoRecent advances in energy-efficient routing protocols for wireless sensor networks areviewrdquo IEEE Access vol 4 pp 5673ndash5686 2016

[78] A Ahmad S Ahmad M H Rehmani and N U Hassan ldquoAsurvey on radio resource allocation in cognitive radio sensornetworksrdquo IEEE Communications Surveys amp Tutorials vol 17no 2 pp 888ndash917 2015

Journal of Computer Networks and Communications 33

International Journal of

AerospaceEngineeringHindawiwwwhindawicom Volume 2018

RoboticsJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Active and Passive Electronic Components

VLSI Design

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Shock and Vibration

Hindawiwwwhindawicom Volume 2018

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawiwwwhindawicom

Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Control Scienceand Engineering

Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Journal ofEngineeringVolume 2018

SensorsJournal of

Hindawiwwwhindawicom Volume 2018

International Journal of

RotatingMachinery

Hindawiwwwhindawicom Volume 2018

Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Navigation and Observation

International Journal of

Hindawi

wwwhindawicom Volume 2018

Advances in

Multimedia

Submit your manuscripts atwwwhindawicom

Page 6: ReviewArticle - downloads.hindawi.com · ReviewArticle QoSStrategiesforWirelessMultimediaSensorNetworksinthe ContextofIoTattheMACLayer,ApplicationLayer,and Cross-LayerAlgorithms MuwongeSsajjabbiBernard

consumption and self-configuration are controlled usingthese three techniques Nodes periodically go into the sleepmode to control the energy consumed in listening to idlechannels and virtual clusters are formed by neighbor nodesfor sleep schedule autosynchronization -e protocol em-ploys in-channel signaling when the radio is in the sleepmode and message passing is applied to decrease contentionlatency for store-and-forwardWSN applications during datatransmission [22] For adjacent nodes to solve their clockdrift S-MAC requires periodic synchronization To reducesynchronization errors exchanged timestamps are relativenot absolute

Advantages

(i) S-MAC has a longer listening period than the clockerror

(ii) It has more capacity to conserve energy than IEEE80211

(iii) It can do trade-off between energy and latencydepending on traffic conditions [22]

212 B-MAC (Berkeley MAC) -is is a carrier-sense MACprotocol designed for WSNs featuring a simplistic scalableimplementation that does network variations and is effectivein channel valuation [23] -e protocol has an interface withsome flexibility leading to very low-power usage and con-trols collisions through collision avoidance mechanismsleading to effective use of the frequency band It uses asampling scheme that is adjustable to minimize duty cyclingand idle listening so as to reduce power wastage Nodes arescheduled with their duty cycle each within which they sendthe preamble to the channel in case of availability of data fortransmission In different duty-cycle schedules channelusage is checked by the destination nodes that is if thepreamble found is long it will remain powered on until thedestination address is obtained and in case it is the target itwill either await senderrsquos data or go to sleep-e protocol haslower latency than synchronous MAC protocols but has aproblem of overhearing where the neighbor node getslengthy preambles though not the intended receiver hencewasting a lot of power in doing so [24] It outperforms anumber of protocols via reconfiguration feedback andbidirectional interfaces for upper layer services It can run at

low duty cycles and its apps cannot experience synchro-nization overhead [23]

Advantages

(i) -e protocol features a simplistic scalableimplementation that does network variations andis effective in channel valuation

(ii) It has an interface with flexibility leading to verylow-power usage

(iii) It controls collisions through collision avoidancemechanisms leading to effective use of the fre-quency band

(iv) It uses an adjustable sampling scheme to minimizeduty cycling and idle listening reducing powerwastage

Disadvantages

(i) It has a problem of overhearing where the neighbornode gets lengthy preambles though not theintended receiver hence wasting a lot of power indoing so

213 T-MAC -is is a contention-based MAC protocol forWSNs that dynamically with little difficulty via fine-grainedtimeouts adapts a listensleep duty cycle It uses this dutycycle by ending its active part so as to vary load in time andposition thereby decreasing the energy wastage on idlelistening [25] When starting an active period it uses a shortlistening window thereby improving the use of energy byS-MAC -rough adaptive duty cycling T-MAC savespower but at a cost of less throughput and enhanced delayBoth S-MAC and T-MAC under similar conditions performin the same way though in variable workloads S-MAC uses 5times more power than T-MAC But T-MAC is complex andnot scalable If its active window is decreased its snoopingcapability on nearby traffic and adaptation to variableconditions of the network is reduced [23] All nodes in thisprotocol are configured to wake up periodically listen toadjacent nodes and go back into the sleep mode till the nextframe as more messages come into the queue Collisionavoidance and reliable communication are ensured throughRequest-to-Send (RTS) Clear-to-Send (CTS) and Ac-knowledgment (ACK) messages among nodes [25]

Table 1 Comparison of this survey with existing surveys

Author(s) Year CharacteristicsBhandary et al [1] 2016 Routing strategies in WMSNs properties and limitationsMuzakkari et al [13] 2018 WSN contention-based scheduling-based and hybridMAC protocolsAhmad et al [12] 2011 QoS-aware MAC protocols for WSNsMuzakkari et al [13] 2010 QoS provisioning at the levels of routing MAC layer and cross-layerShatnawi [14] 2016 IoT application layer protocolsKaragiannis et al [15] 2015 Application layer protocols for the Internet of thingsBernard et al [16] 2019 Cross-layer QoS strategies for the IoT in WMSNsAlAmri and Abdullah [17] 2017 Cross-layer QoS protocols for WMSNsGuo et al [20] 2012 Cross-layer and multipath-based video transmission for WMSNs

Ksentini et al [21] 2006 Toward an improvement of H264 video transmission over IEEE80211e through a cross-layer architecture for WMSNs

6 Journal of Computer Networks and Communications

T-MAC design the major forms of energy consump-tion in T-MAC are through (i) idle listening (ii) col-lisions (iii) protocol overhead and (iv) overhearing Incontrast in the active period the node remains lis-tening and sending data If no activation event occursfor a certain time t then the active period ends -eactivation event might be data reception on a radiocommunication sensing on a radio and end oftransmission for the data packet of the node or ACKknowing that the neighbor has finished transmitting

Clustering and synchronization on waking up a nodebegins by listening to the channel If after a certain timethere is nothing heard it chooses a schedule fortransmitting Synchronization (SYNC) packetsappended with the time of the next frame startup Ifduring this time it hears another nodersquos SYNC packetit follows its schedule subsequently transmitting itsown SYNC Following both schedules enables it to havean activation incident at the beginning of both frames

RTS and t choice if a node does not get a CTS messageafter sending an RTS then either (i) there is a collisionpreventing the destination node from hearing the RTS(ii) RTS or CTS is overheard and destination node barredfrom responding or (iii) the destination node is in thesleep mode So if after time t there is no answer thesource node switches to the sleepmode It should only doso if none of its neighbors is still communicating A newinterval t can be triggered by a neighborrsquos RTS or CTSreception A node may be out of range to hear theinitiating RTS-e length t of the interval should be longenough to get the CTS packetrsquos beginning [25] and t isthus given by the comparison in the following equation

tgtC + R + T (1)

where C contention interval length RRTS packet lengthand T turn-around time

Advantages

(i) It saves power through adaptive duty cycling

Disadvantages

(i) It is unscalable and more complex(ii) It offers less throughput and enhanced delay

214 CU-MAC -is protocol supports IoT standards re-quiring request-response transmissions It employs a multi-channel mechanism for constant bidirectional transmissionof packets at low duty cycle to resolve hidden-terminal issuesAccording to Danmanee et al [24] CU-MAC is an SI-MACprotocol that is aimed at improving bandwidth of the channelby initiating connections and transmitting data throughdifferent channels using the multichannel approach

Protocol design the protocol has the following threeproperties that make it perform more efficiently withWSNs in the IoT

(i) It enables multichannel transmission using 9useable channels in IEEE 802154

(ii) It can enable linked data transmission using oneconnection which reduces additional overheadand delay

(iii) It can support continuous transmission of packets-is is enabled depending on how the bufferoverflow protection policy is set in the buffer stack-is stack has 2 parts (a) bidirectional buffer slotsfor enabling sensor nodes to support linked datatransmission and (b) normal buffer slots fornormal data queuing Buffer stack policies are setas follows

(a) A packet is only dropped in the buffer when itexpires

(b) Unless the packet is from a target node to which areceiver is sending the data the receiver deniespacket reception when buffer slots fill up

(c) -e bidirectional buffer slot is used for bi-directional buffer transfer

CU-MAC can reduce power wastage by choosing toemploy only a single channel for transmitting both controlpackets and data in case there is only one sender in the sameregion of coverage It can further enable multiple trans-missions through another usable channel if there are othersending nodes in this area [24] -e protocol has two phases(i) advertising phase in which a source node announces toneighbors before transmission and (ii) data transfer phasewhich commences when the source node receives a Ready-to-Receive Information packet from the destination node

Advantages

(i) It has a higher packet delivery ratio compared toother MAC protocols

(ii) It performs well in high traffic situations since itsduty-cycle is lower

(iii) It reduces power wastage by employing only asingle channel for transmitting both controlpackets and data

(iv) It can enable multiple transmissions through an-other usable channel if there are other sendingnodes

215 R-MAC -is is a reservation-based MAC protocoldesigned majorly with an aim to be fair and efficient in termsof energy use To eliminate collision of data packets R-MAC[26] schedules control and data packet transmissions at thesource and destination nodes rather than using RTSCTSmessage exchanges Channel utilization in R-MAC is en-hanced by an ARQ method called burst-based acknowl-edgment which together with other scheduling algorithmsaddresses the exposed terminal problem besides enhancingthe network throughput

R-MAC design all nodes in this protocol intermittentlyfunction in listen and sleep states so as to avoid wastingenergy in overhearing and idle modes -e time takento listensleep is the same for all nodes each of which

Journal of Computer Networks and Communications 7

chooses a schedule for itself rendering central sched-uling and synchronization irrelevant So in the absenceof traffic in the locality the node simply does periodiclistening and sleeping and to eliminate collisionstransmissions are distributively synchronized using areservation-based model in case there is a sender totransmit data R-MAC differs from S-MAC in that ittransmits data to a number of nodes using the sleepmode [24] -ere are 3 stages in this protocol (i) la-tency detection (ii) period announcement and (iii)periodic operation of which (i) and (ii) are for nodesynchronization and (iii) is for listensleep procedures

(i) Latency detection phase here all nodes broadcast aNeighbor Discovery control packet (ND) at an ar-bitrarily chosen time since they are powered onWhen the ND is received from the neighbor nodethe receiver notes its arrival time and chooses arandom time to send an acknowledgment (ACK-ND)mdashthe same as the receivedND in sizemdashinwhichthe duration from ND arrival to ACK-ND trans-mission I2 is specified and the time from ND packettransmission to ACK-ND arrival I1 is computed

(a) Propagation latency between the 2 nodesL (I1 minus I2)2

(b) -us propagation latency can be defined as ldquotheinterval from the time the first node sends the firstbit of a packet to the time the second node receivesthe last bit of the packetrdquo [24]

(ii) Period announcement phase here nodes choosetheir own random listensleep time and start timeand broadcast these and when packets are re-ceived the node turns the received schedule intoits own All nodes will record their neighborsrsquoschedules in relation to theirs -ese two phasesmake a number of rounds to ensure all nodes areinformed about their neighboring nodes

(iii) Periodic operation phase here there is datatransmission [26] and nodes intermittently go intothe sleep mode and wake up One period is a listensleep cycle and nodes possess similar periods Twoparts make a period T0 listen window TL and sleepwindow TS as shown in the following equation

T0 TL + TS (2)

In this protocol the nodes make use of control packetsfor communication-ese include REV (reservation packet)ACKREV (acknowledgment packet for REV) and DATAACK-DATA (acknowledgment packet for data) messageexchanges all of which are similar in size

Advantages

(i) It is fair and efficient in terms of energy use(ii) It eliminates collision of data packets

216 AR-MAC (Adaptive-Reliable Media Access Control)-is is a TDMA-based MAC protocol for wireless body areanetworks (WBANs) designed to reduce energy consumption

as proposed by Rahim et al [27] -e protocol allocates aguaranteed time slot (GTS) to all nodes for transmissiondepending on their needs It utilizes periodic sleep andwakeup to minimize overhearing and idle listening based onthe needs of the node It has a central node (CN) having bigbatteries with higher computational power and single ordouble transceivers For double transceivers the sum of thetime frame TFrame is assigned for node transmission But inAR-MAC a single transceiver CN is considered with TFramehaving 3 subdivisions contention-free period (CFP) forsensor connection contention access period (CAP) foremergencies and time TMS to communicate with themonitoring station (MS)

Channel selection the CN first scans for redundantchannels from which it chooses one for transmissionand broadcasts its address and other information to thenodes -e destination nodes search for channels byscanning the radio frequency (RF) channels If free itswitches to the next and if busy it will wait for time TCPlistening to packets and again switch to the subsequentchannel if it does not get the channel packets When thenode finally succeeds in getting a channel packettransmission commences and an ACK packet is sent tothe CNTime slot assignment a Time Slot Request (TSR) packetis sent to the CN by the sensor node after it has selectedthe RF channel -e TSR packet has data rate and timeslot info for the node Static-size time slots with staticguard-band time TGB are proposed in [28] -isprotocol employs the adaptive algorithm of time slot(TS) and GB time and the CN allocates TS whilesending Time Slot Request Reply (TSRR) based on thenode traffic pattern -e size of the time slots variesdepending on the node requirements Based on thetransmissionmodel the data packet transmission ACKpacket reception and some amount of delay can betolerated by allocated TS To prevent interference be-tween adjacent time slots resulting from the node andCN clock drift TGB is inserted between them and itsvalue is given as follows

TGB1

FxTS1100

TGBn

FxTSn

100

TGBnn+1

F

100X12

TSn + TSn+11113858 1113859

(3)

where FGB factor depending on the mean driftvalue Nodes go to sleep after slots have been suc-cessfully assigned and wake up to transmit in assignedslots thus saving energy wasted in idle listening sinceeven allocated slots are collision freeSynchronization for periodic synchronization TDMAneeds additional energy [29] It needs a lot of energy tosynchronize nodes after several cycles To reduce en-ergy wastage and control collisions AR-MAC employs

8 Journal of Computer Networks and Communications

a new synchronization scheme When a data packetarrives the current and expected packet arrival timeswith allowable delay (D) are compared by the CN and adrift value (DV) is computed based on the differenceΔT -e DV is sent to the node in the ACK packet forchanging the slot in subsequent transmissions thoughthe value is based on allowable delay and F In caseΔTgtD the CN will send the DV in the SYNC-ACKpacket for subsequent SYNC to nodes or else a mereACK packet is sent by the CN for the received packetFor any forthcoming data transmissions the node willchange its wake-up time plan based on the DV Withsuch a sync approach a sleeping node will not lose syncfor all cycles N Acceptable delay (D) is given as

D min TS1 TSn( 1113857XF

100 (4)

where F is the guard-band factor

ΔT expected arrival time minus current arrival time

DV 0 if |ΔT|ltD

ΔT if |ΔT|gtD1113896

(5)

Frame format the AR-MAC protocol employs twoforms of packets namely (i) data packets in which anode transmits in its assigned time slot and employsCAP for emergency data and (ii) control packets thatinclude channel packet Time Slot Request (TSR)packet Time Slot Request Reply (TSRR) packet Syn-chronization-Acknowledgment (SYNC-ACK) packetData Request (DR) packet and Acknowledgment(ACK) packet [27]Energy consumption if N is the number of cyclesenergy consumption for these is measured as follows

ETotal 1113944N

k1ESleepk

+ 1113944N

k1EActivek

ESleep TSleep x ISleep x V

TSleep Tframe minus Tactive

(6)

where ISleep current from the voltage source V whenthe node is asleep If Esw is the switching energy Etransthe transmission energy Erec the receiving energy andETout the time-out energy then

EActive 2Esw + Etrans + Erec + ETout (7)

Advantages

(i) It reduces energy consumption(ii) It minimizes overhearing and idle listening based

on the needs of the node via periodic sleep andwakeup

217 ER-MAC It is a combination of TDMA and CSMAmethods making it flexible and adaptable to topological andtraffic changes and hence applicable in rapid response

situations inWSNs Collision-free slots can be planned usingTDMA [30] ER-MAC is designed in such a way that thefollowing are met

(a) It has a higher ratio of packet delivery and low delayunder lower energy consumption levels out-performing Z-MAC

(b) It keeps two priority queues in order to distinguishhigh from least priority packets

(c) It permits contention in TDMA slots so as to handleenormous traffic volumes

(d) It has a harmonized slot shape and the nodes areable to locally change their plans enabling themconnect to or exit the network with easeProtocol design it discovers the topology in such away that it uses the simple flooding mechanism tobuild the tree as initiated by the base station (BS) -etopology discovery aims at neighbor discovery andchange tracking in addition to setting up a routingtree A TOPOLOGY_DISCOVERY message com-prising hop_count new_parent_id and old_-parent_id is generated by the BS A node broadcasts itto discover if it has any potential children and aresponse to the parent In case of the desire to changethe parent the former parent is notified For all nodesthe hop count to the BS parent ID children and one-hop neighbor are registered with the BS at this stageTDMA slot assignment when a BS sends a SYNCmessage it switches to the TDMAmode which whenreceived by the child uses the broadcast slot for itschildrenrsquos synchronization Neighboring nodescannot share a slot since slots are assigned andschedules are exchanged accordingly Slot assign-ment begins with a leaf node in a bottom-up modesuch that the flow of messages to the BS may besmooth Before getting the child-node schedules anonleaf node will not transmit its data via a unicastslot descendantrsquos data via many unicast slots andchild synchronization via a broadcast slot Childnodes send SCHEDULE_NOTIFICATION mes-sages to the BS to end this phase

Local time synchronization is done via parent-childbroadcast sync since they both need to operate the same clockso that when one is transmitting the other is ready to receivePacket prioritization happens in such a way that a highpriority queue is first emptied before a lower priority queue isallowed to transmit In case a queue is filled up the shortestslack packet is dropped and then fairness over source isconsidered as queue modification is done to enable the BSbalance info from different nodes Energy conservation is bysender turning off the radio in absence of any info to transmitand the receiver will switch to the sleep mode in case nopackets are received after a certain timeout [30]

Advantages

(i) It is flexible and adaptable to topological and trafficchanges

Journal of Computer Networks and Communications 9

(ii) It is applicable in rapid response situations inWSNs

(iii) It has a higher ratio of packet delivery and lowdelay under lower energy consumption levels

218 RI-MAC (Receiver-Initiated MAC) -is is an asyn-chronous duty-cycle protocol that tries to reduce timeoccupied by the communicating nodes on the medium toagree on time for data exchange-e sending node waits forthe receiving node to signify the beginning of the com-munication session through a short beacon frame that itsends -e channel is only occupied by the beacon and datatransmission reducing channel occupancy and thus en-abling data exchange for other nodes too -is implies animprovement in channel utilization that subsequently leadsto increment in throughput packet delivery ratio andpower efficiency over a wide-ranging traffic load [31] Beinga receiver-initiated model it ensures substantial reductionin overhearing and lowers the probability of colliding andthe cost of recovery compared to B-MAC [23] or X-MAC[32] RI-MAC is further capable of using a beacon withACK and Ready-to-Receive packets though it has aproblem if a sending node with Ready packets needs to havethe radio on till a ready-to-receive node is awake [13] whichthen sends it a beacon and transmission begins immediatelyand later is acknowledged by another beacon Nodes in-termittently wake up to listen to any possible frames meantfor them depending on their schedules A node announcesits readiness to receive a frame through a broadcast beaconon waking up when it switches on the radio When a dataframe has been received the receiver will spend some moretime known as ldquodwell timerdquo in the active state so as toreceive queued packets which is the time in RI-MAC thatdepends on the number of contending senders If there areseveral contending senders that might not have beenpredicted there is a challenge in trying to reduce the re-ceiverrsquos active time to optimize power efficiency and reducethe cost of collision detection and lost data recovery even inthe case of hidden senders For this protocol this can beaddressed by a beacon frame from the receiving node forcoordinating data frame transmissions of contendingsending nodes-e protocol also beats B-MAC and X-MACin minimizing the cost of collision detection and datarecovery through coordination of the receiversrsquo dataframes -is is because the receiving node is aware of themaximum delay prior to the arrival of the frame-e abilityto broadcast is supported by RI-MAC when it sends data asunicast to each senderrsquos neighboring node or continuouslytransmits the frame in a back-to-back mode for as long asthe sleep interval [31]

Advantages

(i) It improves channel utilization leading to in-crement in throughput packet delivery ratio andpower efficiency over a wide-ranging traffic load

(ii) It ensures substantial reduction in overhearing andlowers the probability of colliding and the cost ofrecovery

219 SRI-MAC (Synchronous Receiver-Initiated MAC)-is is a synchronous duty-cycle protocol that implementsthe method of receiver-initiated data transmission with anaim of enhancing network lifetime through energy wastageavoidance via collision overhearing and idle listening To dothis and also minimize duty cycle it uses a sequence ofadaptive beacons and RTSCTS packets In this protocolsending nodes do not go to sleep till they have got CTSpackets from receiving nodes to begin transmitting exceptfor nonsending nodes that return into the sleep mode rightaway as if the channel was idle [33] Since sending nodeshave separate schedules it reduces the contention proba-bility leading to more energy conservation A time frame isdivided into the information period the allocation periodand the communication period which are discussed asfollows

Information period at frame initiation a beacon is sentby the receiving node declaring its return from sleepand readiness to accept data packets in case the channelis not busy -e beacon is a small packet alerting otherlistening nodes of the transmitterrsquos readiness to get arequest for channel It consists of the receiverrsquos ID andthe duration allocation period (DAP) that relies onnumber of the receiving nodersquos neighbors [13]Allocation period here the receiving node stays awake forsome little time after broadcasting its beacon to confirm ifthere is any node having anRTS packet to transmitWhenthe beacon has been received an RTS packet (with 3fieldsmdashsender ID receiver ID and data size) is sent by thesender prompting the receiver to send a CTS for SYNC asthere is neighbor information exchangeCommunication period SRI-MAC is capable of or-ganizing and scheduling transmissions by means ofsensor node IDs -e receiver assigns order andtransmitting time for every sender such that they switchoff their radios till their time to transmit is dueHowever the receiving node remains in the wake-upmode to get all sending nodersquos data SRI-MAC canenable sensors have lengthy sleeping times duringtransmission such that energy can be saved as long as itis neither sending nor receiving data -e protocolrsquoscommunication period is premised on the TDMAtechnique which conserves energy since the radiorsquosduty cycle is lowered and collision is eliminated [33]

Advantages

(i) Since it employs the TDMA principle it eliminatescollisions while maintaining energy conservationtoo

(ii) Noncommunicating nodes switch off their radioand get into the sleep mode which eliminatesoverhearing

(iii) Since receiving nodes broadcast SYNC signals toall nodes in the form of CTS packets overemittingis avoided

(iv) It decreases duty cycling as sleep time increasesand lowers idle listening

10 Journal of Computer Networks and Communications

2110 H-MAC -is protocol is built on the power savingmechanism (PSM) of IEEE 80211 and slotted Aloha pro-tocols to enhance performance using multiple slots Timehere comprises large frames each having an active and asleeping part [34] If any node has got data to send it will askfor slots from the receiver node during active time and latertransmit the packets in the prenegotiated slots Nodes gointo the sleep mode during sleep-time slots if they do nothave data to transmit

Latency in IEEE 80211 if N is the number of hops nthe value for the current hop tcsn the back-off time andttx the transmission delay the total latency is calculatedusing the following equation

D(N) 1113944N

n1tcsn + ttx1113872 1113873 (8)

and the average latency is given by the followingequation

E[D(N)] N tcsn + ttx1113872 1113873 (9)

-roughput in H-MAC is given by the followingequation

Th npnm

tact + Cts (10)

where np is the number of transmitted packets nm is themaximum number of nodes connected in a time frameC is the number of slots with the same length tact is thelistening time and ts is the sleep delayAdvantages

(i) It enhances performance using multiple slots

2111 Z-MAC (ZebraMAC) -is rides on the strengths andoffsets weaknesses of TDMA and CSMA It yields very goodchannel usage and low delay under low contention and lowerscollision among two-hop neighbors at a reduced cost Itsperformance is strong via error synchronization failure inassigning slots and channel conditions that change with timeIn the set-up phase the protocol operates through certainsteps including neighbor discovery slot assignment localframe exchange and global time synchronization whichunless the topology of the network changes significantly donot run again When the data are being transmitted there isimprovement in network throughput and energy efficiencywhich covers up for the initial upfront operational costs [35]

Disadvantages

(i) Initial slot assignment is quite a cumbersome taskhere since it is done at the beginning yet this maynot be at the same time in reality It further as-sumes static links until setup happens again whichtoo is impractical

(ii) It also suffers from the hidden-terminal problemsince slot owners have a small contention windowand priority for slots allocated to them due to theECN messages

(iii) Synchronizing time also poses a challenge in thisprotocol due to a large clock drift between nodes

(iv) Source and destination nodes too are not wellcoordinated in Z-MAC [36]

2112 Q-MAC (QoS-Aware Media Access Control) -isprotocol offers high QoS with improved energy efficiencyand accesses the wireless channel using MACAW as theunderlying protocol [37] Traffic from different nodes isprioritized for QoS satisfaction depending on how criticalthe data are It mainly has two scheduling algorithms inWSNs (i) intranode and (ii) internode scheduling -eintranode scheduling scheme adopts a first-in first-out-(FIFO-) based queuing algorithm based on application andMAC layer abstraction for data packet classification In-ternode scheduling reduces idle listening and collision on thechannel for improved energy consumption

Advantages

(i) It gives a high QoS with improved energy efficiency

2113 WiseMAC (Wireless Sensor MAC) WiseMAC isbased on the Aloha protocol and employs preamble sam-pling in achieving low-power transmissions in infrastructuresensor networks by occasionally sampling the channel tocheck for activity [3238] -e protocol employs the samemethod as B-MAC except that the sender shortens theextended preamblersquos length by scheduling its transmissionsaccordingly after learning the receiverrsquos awake schedules-is is done by the receiver node putting its subsequentawake time on the ACK packet such that when the senderneeds to send to it again it starts the preamble shortly beforeit wakes up so as to avoid energy wastage in transmitting thepreamble [32] -e data frame is repeated instead of theextended preamble for less traffic transmissions that have alengthier preamble than the data frame in WiseMAC Afterprocessing the frame a node that is not the target recipientwill go back into the sleep mode while the recipient staysawake till the session ends thereby sending an ACK packetFor an idle channel the power consumption is rather lowwith this method Nevertheless there is a problem of longwake-up preamble limiting throughput and causing anenormous power consumption overhead in reception [38]All nodes overhearing a transmission bear the overheadtogether with the recipient node -e protocol can addressissues concerning low-power transmission but has nomechanism of nodes adapting to varying traffic patterns[32] To compensate for drift between clock at the accesspoint and on the sensor node there is a need to compute thewake-up preamble duration -e drift varies directly withtime since previous resynchronization and is given by thefollowing equation

TP min 4θl TW( 1113857 (11)

where θ is the time-base quartz frequency tolerance and l isthe interval between two transmissions Increase in trafficvolume naturally reduces overhearing

Journal of Computer Networks and Communications 11

Radiomodel for protocols with lower power transitiondelays between transceiver states and their powerconsumption are modeled Some of the states includeDOZE where the transceiver neither transmits norreceives but is prepared to power on to the receivetransmit state very fast RXwhere the transceiver listensto the channel receives data or demodulates data fromthe noisyidle channel and TX which is the trans-mission state by the transceiver Let TS be the set-uptime needed to turn the transceiver from the DOZE tothe RX or TX state TT the turn-around time needed toswitch between RX and TX states and B the trans-ceiverrsquos bit rate PZ PR and PT are the values of powerconsumed in the DOZE RX and TX statesrespectively

1113954PR PR minus PZ

1113954PT PT minus PZ(12)

where 1113954PR and 1113954PT are the changes in power as a result ofbeing in RX and TX states respectivelyTraffic model let N be the number of sensor nodesconnected to an access point (AP) and λ be the global rateat which the downlink Poisson traffic arrives at the APfrom the fixed network Traffic toward each sensor node isthe same and is given by λN Average packet interarrivaltime Lwith which data packets arrive at a node is given byL Nλ If TC and TD are the control packets and datapackets respectively assuming a small traffic volume theglobal interarrival 1λ far exceeds the summation ofdurations for the data packet and control packet and theturn-around time Mathematically it is given as

1λ≫TD + TC + TT (13)

-e average power consumed by WiseMAC andtransmission delay incurred are given by equations (14)and (15) [38]

PW PZ +1113954PR TS +(1B)( 1113857

TW+

1113954PR X + TD + TT( 1113857

L

+ 1113954PR(N minus 1)Y

L

(14)

where

X 2θL 1 minus eminus TD4θL( )1113874 1113875

Y T2D + 12TDθL

2TW1 minus e

minus TW4θL( )1113874 1113875

DW TD +TW

21 minus e

minus TW4θL( )1113874 1113875

+ 2θL 2 minus eminus TD4θL( ) minus e

minus TW4θL( )1113874 1113875

(15)

where PW and DW are the power consumed byWiseMAC and the transmission delay for WiseMACrespectivelyAdvantages

(i) It achieves low-power transmissions in in-frastructure sensor networks through preamblesampling while checking for activity

(ii) It reduces overhearing due to increased trafficvolume

Disadvantages

(i) It has a problem of long wake-up preamble limitingthroughput and causing enormous power con-sumption overhead

2114 X-MAC It is a simple minimal transmission powerMAC protocol that uses a short preamble with decouplingfor the sending and receiving node sleep plans inWSNs-isleads to a significant reduction in energy usage at sendingand receiving nodes and decrease in per-hop latency andgives flexibility in adapting to bursty and periodic sensordata sources [32]

Protocol design X-MACrsquos design is aimed at increasingenergy efficiency reducing data latency and increasingthroughput -is enables application in different kindsof packetized and bit stream digitized radios as well assimplified low-overhead distributed application -edesign is as such in order to address the low-powerlistening problems including overhearing excessivepreamble and incompatibility with packetizing radiosas in [32]Asynchronous duty cycling a node with information totransmit will send the packets after sending a preamblewhereas the remaining nodes will preserve their sleepschedules that are not synchronized -e receiver willsample the link on waking up and stays awake if it detectsa preamble and if it is the target it will return to sleep if itis not the target after getting the whole preambleEmbedding the target ID in the preamble to avoidoverhearing the lengthy preamble is divided into manyshortened packets embedded with the receiver nodersquosID so as to eliminate overhearing One lengthy pre-amble is constituted by the shortened preamble packetstream Any node on receiving the short preamble afterwaking up will examine the target ID on the packet Itgoes back into the sleepmode right away resuming dutycycling if it is not the targeted recipient or stay awakefor successive packets if it is the one -is approachchecks energy wastage due to network density It canalso be used on several types of radiosReducing excessive preamble using strobing it ispossible to enable low-power transmission and alsosave energy by reducing time for preamble trans-mission through the use of enhanced preambles inaddition to preamble sampling X-MAC does not sendan endless preamble packet stream but puts slight

12 Journal of Computer Networks and Communications

pauses in between packets to allow the sending node topause and listen to the channel and allow the receivingnode to resend an early ACK packet during that pause-is makes the sending node begin sending datapackets and the receiving node cut excessive preamblesand thus decrease per-hop latency and energy wastageon unnecessary waiting and transmissionPacketizing radios low-power listening is limited incapability for supporting packetizing radios butX-MACrsquos short strobed preambles can support all typesof digital radiosAdaptation to traffic load while a number of WSNapplications are reliable in traffic production they stillmust adapt to changing traffic Nodes are going to haveseparate sleep schedules depending on traffic loadConsidering systems that have time-fluctuating trafficall predetermined static schedules are suboptimalOptimality Buettner et al [32] model the expectationof energy as follows

Es (preamble energy + energy perACK listen)

lowast(expected preamble minus listen iterations)

+(energy to send parket)

Es PTxSp + PRxSal1113872 1113873 Rl + Rs( 1113857

Rl minus Sp+ SdPTx

(16)

where Es expected energy to send a packet PTx andPRx are the transmission and receiver power re-spectively Sp Sal and Sd are the duration of thesenderrsquos preamble ACK listen and data transmissionperiods respectively and Rl and Rs are the receiverrsquoslisten and sleep periods respectively Expected energyto receive a packet is given as

Er (listen cycle energy + sleep cycle energy)

lowast(expected iterations for preamble arrival)

+(energy to sendACK) +(energy to receive packet)

Er PsRs + PRxRl( 1113857

1 minus 1 minus Pd(t)( 1113857 Rl + Rs( 1113857+ RaPTx + RdPRx

(17)

And the one-hop expected latency is given as

Lat (preamble duration + ACK listen)

lowast(expected number of iterations)

+ duration to send packet

Lat Sp + Sal1113872 1113873 Rl + Rs( 1113857

Rl minus Sp

(18)

Advantages

(i) It increases energy efficiency(ii) It reduces data latency(iii) It increases throughput

2115 A-MAC (Advanced MAC) A-MAC is a TDMA-based protocol whose design is aimed at enhancing thelifetime of the network for low rate and reliability of datatransmission [39] To assign time slots it does not depend ona central controller like most protocols but it gathers in-formation from its neighbors so as to allocate the slotsthrough a distributed algorithm Energy is controlled byscheduling power when there is not a single event Its designcomprises many frames with each having many time slotswhich at the beginning have got a beacon transmitted tosynchronize and exchange information with neighborsNodes to participate in the subsequent session are selectedby the controlled node so they do not go into the sleep modewith others [13] -is protocol operates under four time slotdivisions namely initial wait discover and active states Anode in the initial state begins by listening to the medium forthe beacon packet from other nodes for network synchro-nization which when received makes it change the timerthrough subtraction of beacon transmission time from re-ception time To avoid synchronization problems due todrift and allow for continuity the node has to get thestrongest beacon signal It then enters the wait state afterchoosing a number of waiting frames so as to reduce theprobability that nodes get into the discover state simulta-neously In case the synchronization beacon is lost while stillin the wait state it will return to the initial state for anotherone On expiry of the waiting counter it enters the discoverstate and collects information from neighbors by listening totheir beacon messages On successful selection of a time slotit enters the active state where it endlessly transmits beaconsat the initiation of the slot A node goes to sleep when abeacon has been transmitted and there are no more data fortransmission or in case the neighborrsquos beacon received in-dicates no data packets are coming in [39]

Advantages

(i) It can enable multicasting(ii) It enhances the lifetime of the network(iii) It offers reliable data transmission

2116 L-MAC (Lightweight MAC) Based on TDMA thisprotocol is aimed at minimizing the number of transceiverswitches adapting the nodal sleep interval to data trafficamount and making implementation less complex [40] Apacket is transmitted comprising a control and a data partControl packets have a fixed size and the time slot con-trollerrsquos ID -ey show the distance from the transmittingnode for simplified routing to the network gateway as well asaddressing target node and reporting data unit length andinternode synchronization Adjacent nodes are more in-terested in getting control packets from nearby nodes -enodersquos transceivers are switched off until the subsequenttime slot unless the node is also addressed or the message isomnicast If so it listens to the data unit which may not fillthe remaining time slot in totality-e transceivers on eitherside are turned off when the transfer of packets is successfulEnergy wastage through idle listening in uncontrolled timeslots is avoided via a short time-out interval Network nodes

Journal of Computer Networks and Communications 13

are able to communicate without colliding-is saves energyand increases the lifetime of the network A node in L-MACcan only transmit one message per frame [13] -e networkis set up in such a way that all nodes are not synchronized onbeing switched on-e gateway takes control of a timeslot soas to have them synchronized and its immediate neighborgets the control messages thereby synchronizing their timerswith the gatewayrsquos Neighboring nodes learn of multiplegatewaysrsquo time slots in their vicinity just after one frame-ejust synchronized nodes randomly choose time slots forcontrolling as long as they are not occupied -e nodes thuspreserve a table for the neighboring nodes which makes atimeslot reusable after a minimum of 3 hops and preventsmessage collision All nodes keep their slots before thebattery is down or learn of an impending collision All nodescan track their hop distances to a selected gateway node andthe information is communicated in control messages [40]

Advantages

(i) It minimizes the number of transceiver switches(ii) It adapts the nodal sleep interval to data traffic

amount(ii) It makes implementation less complex(iii) It offers improved network lifetime(iv) Energy wastage through idle listening in un-

controlled time slots is avoided via a short time-out interval

Disadvantages

(i) L-MAC is not very good for mobile SNs since slotsare calculated only once

(ii) It suffers idle listening because nodes keep listeningto slotsrsquo control parts so as to get data and enableother nodes to join the network all the time [13]

2117 C-MAC -is was designed for WSNs with 3 seg-ments namely (i) aggressive RTS having double channelcheck to assess the channel (ii) anycast to initialize the flowand (iii) convergent packet forwarding to stabilize the flowIt promises an improvement in energy efficiency lowerlatency and enhanced throughput while avoiding syn-chronization overhead though it operates only in low duty-cycling apps especially where there is less traffic But whenthere is more traffic the protocol employs anycast-basedpacket forwarding for waking the nodes up or finding aforwarder quickly and converges from route-suboptimal toroute-optimal unicast After this energy could be preservedby the nodes using a synchronized wake-up plan [13]

Advantages

(i) It improves energy efficiency(ii) It lowers latency(iii) It enhances throughput avoiding synchronization

overhead

Disadvantages

(i) It operates only in low duty-cycling applicationsespecially where there is less traffic

22 MAC Protocols for WMSNs

221 Diff-MAC -is is a CSMACA-based QoS-awarehybrid-priority-based MAC protocol for WMSNs It targetsenhancing the channel utilization while making use of ef-ficient service differentiation to do coordination for mediumaccess of each traffic class while giving fair quick datatransfer It offers QoS using the following features [14]

(i) It balances energy consumption and delay throughsensor node duty-cycling adaptation depending onthe dominant traffic class

(ii) It has a feature for fragmentation and messagepassing that breaks lengthy video frames intosmaller packets that are transmitted as a burst thusreducing retransmission costs if there are any MACfailures

(iii) -ere is fair data delivery amongst the nodes andtraffic classes respectively resulting from intranodeand intraqueue prioritization reducing unbearableperformance

(iv) -e number of collisions can be reduced and packetlatencies are kept within limits by adjusting the sizeof the congestion window in this protocol as per thetraffic needsAdvantages

(i) Diff-MAC is very suitable for WMSNs since itadapts quickly to changes in networkconditions

(ii) It is a fair protocol(iii) It is scalable

Disadvantages

(i) It is quite difficult to monitor the statistics ofthe network

(ii) Adapting dynamically to the network condi-tions is not an easy task

(iii) It exhibits some degree of latency in packetdelivery

-e implementation of Diff-MAC on Imote2 indicatesthat resource requirements can be met on the sensorhardware Results of extensive simulation runs showed thatDiff-MAC outperforms its competitors

222 Cluster-Based On-Demand Multichannel MAC Pro-tocol for WMSN (COM-MAC) -is is an energy-efficientprotocol in WMSNs with high throughput and reliable datatransmission It achieves energy efficiency by schedulingmultichannel media access to reduce channel contentionand remove collision idle listening and overhearing [41]

Design it has a three-phased operation with the (i)request phase (ii) scheduling phase and (iii) datatransmission phaseRequest phase here all nodes have two protocols eachfor sending requests to the cluster head (CH) ie thecontention-based and contention-free TDMAFDMA

14 Journal of Computer Networks and Communications

protocols -e contention-based protocol allocates achannel per node for sending the request message andfree channels are useable as control channels Forcontention-free protocols a control slot is allocatedwhen a network is deployed Here a slot per channel isallocated to a different node enabling requests to be sentwithout interfering with transmissions from othernodes -e contention-based protocol operates undertwo steps (i) Control channel assignment phase achannel is assigned per node for transmitting a request(REQ) message A channel is assigned to multiplenodes and nodes are evenly distributed to availablechannels for congestion control (ii) Request trans-mission phase nodes with data send a REQmessage viathe assigned control channel to notify the CH whichresponds with an ACK message -e contention-freeTDMAFDMA protocol too has two phases includingthe control slot assignment and request transmissionphases For unreliable channel conditions and lowtraffic load the contention-based protocol is fit for usebecause of less collisions and congestion whereas thecontention-free protocol is preferable in heavy trafficload and reliable channel applicationsScheduling phase from the above phase a schedule isgenerated by the CH for coordination and this enablesnodes to transmit and broadcast their data via controlchannels-e schedule for broadcasting only comprisesdata for nodes assigned to a channel -e schedule isrebroadcast for transmission reliability enhancementbut for energy efficiency the nodersquos transceiver isswitched off when all the schedule is completely de-livered till it is the time for transmitting again-e timeslot and data communication radio channel are allincluded in the schedule generated for each nodeData transmission phase on getting the schedule thesensor node sends its data as per the time slot and thechannel assigned -e time slot is divided into datatransmission andACK sections ACK supports link layererror control Using implicit selective repeat ARQ theCH sends an ACK message for all received packets Ifthere is no ACK the packet is assumed lost and hence aretransmission during the subsequent interval is done-is enhances reliability of transmissions though excessdelay is possible here which is also intolerable for real-time critical apps like video or audio -is gives rise to ahybrid MAC protocol to exploit the unused spectrumduring data transmission sessions [41]Advantages

(i) It achieves energy efficiency by scheduling mul-tichannel media access to reduce channel con-tention and remove collision idle listening andoverhearing

(ii) It maximizes network throughput by dynamicallyallocating time slots and channels for nodes byexecution of a traffic-adaptive QoS-aware sched-uling algorithm depending on needs for QoS andtraffic data information

(iii) It increases transmission reliability by in-corporating spectrum-aware ARQ to avoid spec-trum wastage

223 EQoSA -is protocol is developed for video andimage transfer across WMSNs in a hybrid mode withprovision of QoS Concerning the active sensor nodes andtheir traffic loads it employs dynamic session sizes andchanges bitmap-assisted (BMA) MACrsquos fixed session sizeNodes declare any data available for transmission duringcontention time -is forces the cluster head to make aschedule of how slots have been assigned It communicatesthis schedule to the nodes giving each node a given numberof slots per session and so enabling the busty traffic to beaccommodated [14] -e protocol does not perform wellwith DiffServe mechanisms but does well with schemessupporting busty traffic loads and strong cluster heads arenecessary in WMSNs that can assign and broadcast slots

224 QoS-Supported Energy-Efficient MAC (QE-MAC)Based on the IEEE 80211e standard this protocol ensuresfairness and has low latency and jitter and better efficiencyon energy usage than other QoS-aware MAC protocols forWMSNs It operates under a two-phased routine

-e first phase introduces the innovating prioritymechanism in IEEE 80211e QoS and several data types areassigned priorities Its coordination function is a combi-nation of distributed contention-based channel allocationand centralized polling-based channel access mechanisms

-e second phase preserves the nodesrsquo energy usingdynamic duty cycling -ere is an exchange of RTSCTSpackets to supplement CSMACA by muting nodes near thesenderreceiver during packet duration [42]

Advantages

(i) It ensures fairness(ii) It has low latency and jitter and better efficiency on

energy usage than other QoS-awareMAC protocolsfor WMSNs

225 Black-Burst (BB) Contention A contention-basedMAC scheme for enhancing instantaneous QoS access tocarrier-sense WSNs is proposed in [42] In this schemenodes contend for the channel until it is free Real-timepackets are given higher priority over non-real-time packetsand if a node is transmitting real-time packets it is prior-itized over other nodes [15] Voice and video are some of thereal-time apps in consideration here which periodicallyaccess the common radio channel when transmitting -eseneed constrained end-to-end delay of packets at the MAClayer -is protocol majorly does the following [42]

(i) Prioritization of real-time traffic(ii) Implementation of the round-robin mechanism in

real-time nodes(iii) Bounded access delays among real-time packets

Journal of Computer Networks and Communications 15

(iv) Implementation of real-time communication withvarying needs of bandwidth good for multimediatraffic

Amechanism for dynamically assigning packet prioritiesbased on their deadlines and the traversed hops is proposedin [43] -e ReAllocative Priority (ReAP) mechanism ismeant to provide QoS at the MAC layer to video traffic -isscheme introduces adaptive TXOP by modifying the TXOPdynamically based on the queue length [15] Other protocolsin this category include dynamic duty cycle and adaptivecontention window-based MAC protocol frame sharing(FRASH) MAC protocol and real-time independentchannels (RICH) MAC protocol

Advantages

(i) It can implement real-time communication withvarying needs of bandwidth

23 Challenges and Opportunities(a) One of the challenges in satellite transmission is

cross-talk leading to interference in different fre-quency bands It is necessary to increase the rise andfall times as well as protect sensitive nodes as wecircumvent floating ones if we are to regulate cross-talk as in satellite communication therefore Wefurther need to investigate the possibility of usingdifferential signaling -is constitutes an opportu-nity for further research

(b) Another challenge is that the hidden and exposedterminal problems are still noticeable despite the useof RTSCTS solutions in MAC since collisions arestill possible from transmissions of separate nodesAt the MAC layer ACKs might block efficientchannel usage by prohibiting exposed terminalsfrom reusing the channel and may be eliminatedleaving no proof of sender data packet receptionUsing a directional antenna is one of the proposedtechniques of high-speed WSNs like IEEE 80211-is constitutes an opportunity for further study intothese issues for better solutions though

(c) Another challenge is that since nodes in contention-based MAC protocols are mobile there are manycollisions by the packets and schedule in-consistencies in schedule-based protocols in the two-hop neighborhood information when nodes areentering or leaving [44] Deng et al [45] propose amobility-based clustering (MBC) protocol for mo-bile nodes to enhance the packet delivery rate -isprotocol in comparison with CBR has 25 chancesof decreasing collision-induced packet losses and50 chances in comparison with low-energyadaptive clustering hierarchy (LEACH)

(d) -e process of analyzing and evaluating routingperformance for WMSNs is quite complex becauseof the bursty multimedia traffic Because of the wayencoding scheme frames are structured as well as thechanges in between the scenes that occur naturally

the compressed video frames display some substantialburstiness on several time scales -ere will also befrequent route recalculation that is as well energy in-tensive for nodes frequently capturing and transmittingenergy -e solution here could as well possibly be theuse of the MBC protocol Route recalculation tooshould be dynamic to cater for the mobile nodes whichmight be in and out of the range at different instants oftime Recently the use of smart dumpsters for effectivewaste management in smart cities has also been pro-posed as a possible solution [46]

(e) For WSNs their routing protocols do not look atmultimedia applications which require enormousbandwidth and processing energy and should betransmitted in real time with utmost fidelity Extrachallenges in routing are imposed in WMSNs indeployment of heterogeneous sensors since therecould be audio video and still pictures in addition toscalar data during transmission Heterogeneous datahandled also have different QoS requirements andwith many different business needs different ser-vices with varying requirements give rise to a bigchallenge for the routing design -ere is a need todesign protocols that can support WMSNs such asEQoSA [14] for this purpose

Table 2 summarizes the reviewed MAC protocols forWSNs and the ones designed for suitability in WMSNs

3 QoS Strategies at the Application Layer

We discuss many protocols used to solve different needs ofcommunication between machines

31 ConstrainedApplication Protocol (CoAP) It is a requestresponse protocol designed by the Internet Engineering TaskForce (IETF) for constrained devices such as those havinginsufficient RAM or CPU and WPANs with low-powerconstraints resulting in poor transfer of packets andenormous overhead -e protocol was designed for ma-chine-to-machine (M2M) apps and system automation toreduce bandwidth requirements increase packet transferdecrease the overhead and make work simpler leading tolightweight implementation HTTP commands like GETPOST PUT and DELETE are used in a client-server ar-chitecture Using the publishsubscribe architecture CoAPcan support multiple users and hence yield better perfor-mance via multicasting enabling asynchronous messageexchange -is protocol can support both multicasting andunicasting via UDP and hence is a reliable and simpleprotocol -is is done via the messaging sublayer (for re-liability based on UDP) and the requestresponse sublayer(for communication and interaction) [16]

It uses the following messages

(i) Confirmable message this guarantees reliabilityusing ACK messages sent by the receiver node-ese could be synchronous or asynchronous ifthere is a need for more computational time

16 Journal of Computer Networks and Communications

(ii) Nonconfirmable message here no ACK messagesare sent back

(iii) Acknowledgment (ACK) message this confirmsreceipt of a confirmable message

(iv) Reset message this confirms receipt of an un-processed message in case any critical parts nec-essary for interpreting the message are missing

(v) Piggybacked response the receiver directly re-sponds while getting the ACK message

(vi) Separate response the reply by the receiver is dis-crete from the ACK message

CoAP has no in-built security features It is secured bythe datagram transport layer security (DTLS) that runs ontop of UDP though it was not designed for the IoT -ismakes its suitability debatable since DTLS offers no supportfor multicasting It also needs more packets for handshakingthereby increasing the traffic volume and computationalresources and so shortening the lifespan of mobile devicessince they use batteries [17]

Advantages

(i) It can support multiple users and hence yieldbetter performance via multicasting

Table 2 Comparison of MAC layer protocols used in QoS provisioning

Protocol Features QoS parameter Comments Access Priorityassignment Energy eff Designed for

S-MAC[22] Collision avoidance Energy latency Requires periodic

synchronization CSMA Hybrid Medium WSNs

B-MAC[23] Collision avoidance Energy delay

throughput Flexible interface CSMA Dynamic High WSNs

T-MAC[25] Collision avoidance Energy Complex nonscalable CSMA Dynamic High WSNs

C-MAC[13]

Dynamic channelallocation Energy Multichannel MAC CSMA Dynamic Medium WSNs

CU-MAC[24]

Multichannelmechanism

-roughputlatency High packet delivery ratio TDMA Dynamic mdash WSNs

A-MAC[39]

Per flow serviceguarantees Delay m-ary tree model

dynamic prioritiesTDMACSMA Dynamic Medium WSNs

ER-MAC[30]

Emergencyresponse

Energy delaydelivery ratio

Adapts to traffic andtopology changes

TDMACSMA Hybrid Higher than Z-

MAC WSNs

RI-MAC[31] Collision detection Delay packet

delivery ratioHigh packet delivery ratio

and throughput CSMA Dynamic Medium WSNs

SRI-MAC[33] Collision avoidance Energy delay Good data transfer and

energy consumption TDMA Dynamic High WSNs

H-MAC[34] Slotted sleep time -roughput

latencyData transmission only in

sleep timeCSMAAloha Dynamic Medium WSNs

Z-MAC[36] Collision avoidance Energy delay

throughputVery good channel usage

and error syncCSMATDMA Dynamic Medium WSNs

Q-MAC[37]

Intrainternodescheduling Energy High QoS improved

energy efficiency CSMA Hybrid High WSNs

X-MAC[32]

Overhearingavoidance

Energy delaythroughput

Reduces excessivepreamble by strobing CSMA Adaptive

static Medium WSNs

L-MAC[40] Collision avoidance Network

lifetime energyReduces the number oftransceiver switches TDMA Static Medium WSNs

WiseMAC[38]

No downlinkchannel collision Power delay Significantly low-power

consumptionAlohaCSMA Static Higher than

ZigbeeInfrastructure

WSNsAR-MAC[27] Collision avoidance Energy delay Uses data packets and

control packets TDMA Static Higher thanIEEE 802154 WBANs

BB-MAC[42]

Real-time trafficprioritization Delay Employs round robin in

real-time nodesCSMATDMA Dynamic High WMSNs

COM-MAC [41] Collision avoidance Packet delay

throughputBalances reliability and

retransmissionTDMAFDMA Dynamic High WMSNs

EQoSA[14]

Busty traffic loadaccommodation Energy Provides QoS support for

video and imagesTDMACDMA mdash High WMSNs

QE-MAC[42] Collision avoidance Energy jitter

latencyRuns central polling-like

channel access CSMA Dynamic High WMSNs

R-MAC[26] Collision avoidance Energy delay Reservation based mdash Static High Underwater

SNs

Journal of Computer Networks and Communications 17

(ii) It enables asynchronous message exchange(iii) It is a reliable and simple protocol

32 Message Queue Telemetry Transport (MQTT)Developed by IBM and later improved by OASIS with anintention of decreasing bandwidth MQTT is a simple open-source publishsubscribe protocol analogous to the client-server version It is appropriate for lightweight IoT appli-cations and M2M transmissions in constrained situationslike limited power computation capability and memory orbandwidth -e network bandwidth and the computationalresource usage are decreasing since clients are not updatedmaking publishsubscribe protocols outperform requestresponse protocols via IoT requirements [16] -is protocolcan run over TCPIP and so has the following advantages

(i) Delivering of packets with reliable guarantee(ii) Enabling multicasting(iii) Remote device session establishment(iv) Reduction of transport overhead and protocol ex-

change to minimize network traffic(v) Provision of the notification mechanism when

anomalies occur [47]

MQTTrsquos design targets economical use of bandwidthand battery so it has a major application in FacebookMessenger MQTT ensures QoS and reliability of messagedelivery by providing the following three QoS levels

(1) Fire and forget a message is sent once depending onthe networkrsquos best effort and there is no need forACK (lowest QoS level)

(2) Delivered at least once a message is sent at least onceand an ACK message is needed to avoid duplicates

(3) Delivered exactly once by a four-way handshake itensures a message is sent exactly one time (highestQoS level)-e architecture it has three key components (i)publishersmdashlightweight nodes sending data to thebroker before returning to sleep (ii) broker whichorganizes sensory data in topics and sends them tointerested subscribers and (iii) subscribersmdashIoTapps interested in data sent by sensors through thebroker as shown in Figure 3

MQTT is a simple lightweight protocol that is betterthan CoAP via low and high packet loss probability interms of throughput and exhibits lower latency in lowsampling rates [48] It supports a number of imple-mentations for embedded devices and mobile applicationsand offers web-socket support too Nevertheless it hassome shortcomings

(i) It is only employed in simple data types(ii) It exhibits high latency in high sampling rates(iii) It is unsuitable for real-time apps in such cases(iv) It does not provide automatic discovery andor

sufficient security at the protocol level [49]

33 Secure Message Queue Telemetry Transport (SMQTT)-is is an extension of MQTT that is based on lightweightattribute-based encryption using elliptic curve cryptogra-phy -e protocol has three entities as shown in Figure 4(i) publisher that publishes information under a certaintopic (ii) subscriber that obtains the same topic in-formation via a broker and (iii) public key generator(PKG) or broker that acts as the trusted third party -eprotocol has four phases namely (a) set-up phase (b)encrypt phase (c) publish phase and (d) decrypt phase

(a) Set-up phase publisher and subscriber provide aunique ID with attributes for registration with thePKG -is generates a secret master key set andpublic parameters that it publishes with set U A1A2 An For ciphertext policy-attribute-basedencryption (CP-ABE) the PKG gives a key set todevices with attributes [50]

(b) Encrypt phase based on the access tree the publishercreates an access policy An access tree is sent by thepublisher to the PKG for generating the key-gen-eration policy in the key policy- (KP-) based ABEalgorithm -e publishers know beforehand thetopics and subscribers to access as well as the rulesfor access Subscribers are issued with private keysfor all communication sessions Publishers in theCP-ABE algorithm come up with access trees andrules where the payload is encrypted by the pub-lisher who also issues more info needed for de-cryption and the rules With the help of public keycryptography data are encrypted giving rise to ci-phertext on the basis of KPCP-ABE and the receivergets SUBACK from the broker [51]

(c) Publish phase ciphertext is embedded in theSPublish command as payload and topic name isdefined in the header by the publisher -e brokerreceives the SPublish packet and replies withPUBACK prompting the publisher to respond withthe PUBREL packet -e broker then broadcasts totopic subscribers and deletes the data forwarding thePUBCOMP packet to the sender node

(d) Decrypt phase the encrypted information isdecrypted by the subscriber with the help of a secretkey Considering KP-ABE on verification that itsatisfies the access policy the subscriber invokes thePKG for a corresponding key which it sends afterverifying the request With CP-ABE subscribers usethe private key for decryption after authenticationthereby enabling offline interaction [51]Advantages

(i) -e information being encrypted makes it moresecure

34 MQTT for Sensor Networks (MQTT-SN) -is is a ver-sion of MQTT designed to operate under environments oflow bandwidth high link failures and short message lengthIt is a publishsubscribe (pubsub) protocol designed for

18 Journal of Computer Networks and Communications

UDP-based WSNs to visualize communication for low-power devices [5152] It comprises a publisher device actingas the MQTT-SN client to forward communications to agateway that also changes the MQTT-SN message to theMQTTmessage before delivering it to the broker that sendsit to MQTT-SN subscribers [51]

It has a gateway for changing protocols from MQTT toMQTT-SN (Figure 5) [52] -is protocol supports nodes in

the sleeping mode using an offline keep-alive procedure thatcan be used by battery-operated devices to get into thesleeping mode To support sleeping clients MQTT-SN has anew offline keep-alive procedure with which battery-oper-ated devices can go to the sleeping state in which theirmessages are buffered at the server until they are awake InMQTT-SN clients with no preconfigured gateway addresscan figure out the real address of their operational gateway

Publisher

Subscriber

Subscriber

Broker

Publisher

4G

Figure 3 Architecture of the MQTT protocol

Register by mentioning URI Register by mentioning URI

Access tree

Set-up

Publisher ReceiverPKG(broker)

Key management Key management

Encrypt data using public parametersand random key This is done using

CP-ABE Publish it as

Subscribe (topics)SUBACK

S-Publish (TopicEncrypt (data))S-Publish (TopicEncrypt (data))

Encryptand publish

PUBACK

PUBACK

PUBRELPUBRELPUBCOMP

PUBCOMP

Figure 4 SMQTT protocol

Journal of Computer Networks and Communications 19

within the network through a discovery procedure Simul-taneously we may have several gateways in a single wirelessnetwork that can be cooperative via load sharing

MQTT-SN architecture it is composed of MQTT-SN clients gateways and forwarders (Figure 5)rough the MQTT-SN gateway (GW) the MQTT-SN protocol is used by the MQTT-SN clients toconnect to the MQTT server Integration betweenthe MQTT-SN GW and the MQTT-SN server mightbe possible or not Translating between the MQTT-SN and the MQTT for a stand-alone GW is doneusing the MQTT-SN protocol If a GW is detachedfrom the network it can be accessed by the MQTT-SN clients through a forwarder is just encapsu-lates the received MQTT-SN frames that are on thewireless side before forwarding them to the GW un-altered It then decapsulates the received frames tooresending them to the client as sent by the GW [52]ere are 2 types of gateways (a) transparent gatewayand (b) aggregating gateway as shown in Figure 6

(i) Transparent gateway this launches and sustains adirect MQTT connection between the MQTT-SNclient and the MQTT server which is entirely re-served for end-to-end client-server messaging enumber of servers and that of MQTT-SN clientsconnected to the GW are the same is transparent

GW translates syntax between these protocols and iseasier to implement compared to the aggregatingGW But the MQTT server must support dierentsessions for all connected clients Some MQTTservers are congured to enable few parallel sessions

(ii) Aggregating gateway this maintains a single MQTTconnection to the server on behalf of all clients Allexchanged messages from the MQTT-SN client to theaggregating GW are directed to the GW which sortsthe information to forward to the server e aggre-gating GW decreases the number of MQTT sessionssimultaneously supported by the server in WSNs withmany battery-operated sensors and actuators (SAs)having limited storage and processing capabilities It ishowever more complex to implement compared to atransparent GWe general format of the MQTT-SNprotocol message has two or four message headeroctets and n octets of variable header that is optionaldepending on the message in question MQTT-SN has256 message types (MsgType) with those from0X1Eminus 0XFD 0X19 and 0XFF reserved for futureusage [51]

Advantages

(i) e protocol can operate under environmentsof low bandwidth high link failures and shortmessage length

MQTT-SN gateway

MQTT-SN

MQTT

MQTT-SN

MQTT-SN

MQTT-SN

MQTT-SN client

MQTT-SN clients

MQTT broker

MQTT-SN client

MQTT-SN client

MQTT-SN gateway

MQTT-SN forwarder

Figure 5 MQTT-SN architecture

20 Journal of Computer Networks and Communications

(ii) It is a pubsub protocol designed for UDP-basedWSNs to visualize communication for low-power devices

(iii) It can be optimal for low-cost battery-operatednodes having low processing power and lessstorage

(iv) It supports nodes in the sleeping mode using theobrvbarine keep-alive procedure used by battery-operated devices

35e Extensible Messaging and Presence Protocol (XMPP)e standardization of this protocol by the IETF made itpopular for usage over the Internet especially in instantmessaging and chatting platforms It is however susceptibleto spamming attacks and its failure to have worldwidesupport has made Google too withdraw its support of thestandard [17] e protocol is mainly useable in the client-server architecture but can also be applicable in client-clientXMPP communication [53] Among the application layerprotocols XMPP is the only one that supports both asyn-chronous (publishsubscribe) and synchronous (requestresponse) services while running over TCP but can still runover other protocols

Given the simplicity of its implementation and ease ofusability in XMPP instant messaging (IM) is suitable forXMPP and is implemented using the message stanzaltmessagegt ese stanzas are usable in one-to-one ormultiuser sessions or chatting or in error reporting andsending headlines Attribute type of the stanza denes themessage type e attribute type could be a chat group chatheadline normal or error An instant messaging network inXMPP can support many XMPP clients and servers with agateway congured to enable other IM network users toconnect to other XMPP users e XMPP server mainlycontrols sessions and XML streams to clients and servers inaddition to routing XML stanzas in the streams [53] eTCP ports 5269 and 5222 have been respectively registeredfor the XMPP server-to-server connections and XMPP clientconnections by the Internet Assigned Numbers Authority(IANA) XMPP enables specication of XMPP ExtensionProtocols (XEP) to enhance its usefulness [17]

XMPP security clients connect to the server usingtransport layer security (TLS) and simple authentication andsecurity layer (SASL) protocols which are at the same timeused by the servers for interdomain communications Fordata condentiality and integrity TLS will encrypt thestreams of XML for all sessions To conrm the identity of anyclient that needs to access the server the SASL protocol doesthe authentication of the XML stream For any communi-cating session with a server to start the client needs to resolvethe serverrsquos DNS hostname For protection of the server of theclient from third-party attacks the server maintains the au-thentication details secret ie IP address and access methodand only original connections of the server are required

In server-to-server communications the SASL protocolis employed for authentication and condentiality Server-to-server communications may as well be deactivated from aserver by an administrator depending on the organizationalpolicy To guard against domain-spoong attacks serverdialback can be a good solution since servers support itsecuring more the XML stanzas Nevertheless it cannot beused to authenticate secure or encrypt in-server streamsand the server identication resulting therein is not strongenough to be relied on It further cannot secure against DNSpoisoning attacks and IP session hijacking for remote do-mains [53] High-security domains can use TLS and SASL

Advantages

(i) It is popular for usage over the Internet especially ininstant messaging and chatting platforms

Disadvantages

(i) It is susceptible to spamming attacks(ii) It has failed to have worldwide support making

Google withdraw support of the standard(iii) It is not secure against DNS poisoning attacks and

IP session hijacking for remote domains

36 Advanced Message Queuing Protocol (AMQP) isprotocol can be utilized by a number of platforms formessage exchange since it is open source and written inmany languages It is a publishsubscribe model depending

MQTT-SNclients Broker

MQTT

(a)

MQTT-SNclients Broker

(b)

Figure 6 (a) Transparent gateway and (b) aggregating gateway

Journal of Computer Networks and Communications 21

on a reliable and efficient messaging queue with capacity forinstant messaging or e-mail for different applications es-pecially in IoT environments It differs from others in thesense that message types can be specified and have theirsource traceable and how to do trade-off between perfor-mance security and reliability For reliability of messagetransmission AMQPmakes use of some delivery guaranteeslike at most once at least once and exactly once It can alsohowever utilize the TCP transport layer for guaranteeingreliability -e AMQPrsquos publishsubscribe method has twocomponents namely (i) exchange queue meant for routingof messages to an appropriate order in the queue and (ii)message queue that acts as storage pending their delivery toreceivers [16] In this protocol model the broker can makerouting decisions contrary to othermessaging systems whereapps using the queue have in them embedded the decision-making logic for recipients or senders -e developer mayneed to check each of the affected apps so as to change thelogic for routing and delivery of the messages [54]

Advantages

(i) It is convenient(ii) It has capacity for interoperability between dif-

ferent vendors(iii) It can use open standards to enable connection

between business partners(iv) It can enable innovations based on AMQP foun-

dations [55](v) It can enable supporting of mission-critical ap-

plications in e-commerce(vi) It can guarantee reliable delivery of messages

Disadvantages

(i) It is not very appropriate for real-timeapplications

(ii) It may not offer automatic discovery(iii) Interoperability is not guaranteed(iv) It lacks open-source libraries for constrained

gadgets [49]

37 Data Distribution Service (DDS) A product of theObject Management Group (OMG) DDS is a pubsubprotocol meant for M2M communications with two sub-layers -ese are the data-centric pubsub sublayer re-sponsible for connecting anonymous data publishers tosubscribers and the data-local reconstruction sublayer that isoptional and integrates DDS into the application layer [56]-e DDS architecture connects system components asshown in Figure 7 Publishers and subscribers are discon-nected with respect to (i) time (ii) space (iii) flow andbehavior (iv) platforms and (v) programming languages[57] A DDS data publisher generates topics and publisher-subscriber coupling is represented as topic name data typeschema and publishers and subscribersrsquo QoS attributesDelivering of data is done by the publisher layer -e datawriter and publishers together make a decision about anynecessary alterations for subsequent forwarding to thesubscribers who need to send data to IoT applications Data

readers are responsible for delivering topics to the sub-scribers after reading through them [56]

Some of the major components of the DDS architectureinclude the following [57]

(i) Topic this is a logical path specifying publicationand subscriptionrsquos data type between DataWritersand DataReaders For a successful session to begintopic names types and DataWriters and Data-Readersrsquo QoS should correspond

(ii) Domain this is a logical setting for transmissionthat can be utilized for separating and optimizingnetwork transmission in a setup of similar appli-cations If DDS apps share a domain ID then theycan exchange data among themselves

(iii) DataWriter and DataReader these are terminalobjects for writing and reading messages re-spectively to and from a universal data space

(iv) Participant this is an object that stands for apublisher or subscriber of the DDS app in a do-main acting like a container for other objects

Advantages

(i) It is highly reliable and offers QoS support(ii) It is highly scalable(iii) It is appropriate for real-time apps(iv) It is fault tolerant(v) It has high level of interoperability(vi) It does not need brokers that may act as

bottlenecks(vii) It offers automatic discovery(viii) It has a user-defined data structure for topic(ix) It supports publisher and subscriber decoupling

Disadvantages

(i) It is originally developed for only stand-aloneLANs

(ii) A lot of memory is needed(iii) It has no open-source libraries to support con-

strained devices(iv) It is hard to design and configure compared to

other protocols [49]

38 RESTful Services REpresentational State Transfer (REST)-is is an architectural style based on a set of principles thatdefine and address networked resources It offers web servicesthat enable transmission from the HTTP device to the devicewithin the IoT architecture and enables clients to reach theserver [16]-is protocol utilizes HTTP commands to backupthe requestresponse messaging model HTTP is a popularand secure protocol in the WWW using TLSSSL [16] ARESTful application has the following characteristics

(i) State and functionality are separated into distrib-uted resources

(ii) All resources are distinctively named by the use ofunique HTTP commands (GET POST PUT orDELETE over the Internet)

22 Journal of Computer Networks and Communications

(iii) It is a clientserver stateless layered protocol thatsupports caching REST uses the same methods asHTTP for all requests or responses with POST andGET for creating and retrieving resources PUT forupdating and changing the state of resources andDELETE for removing resourcesAdvantages of REST over arbitrary web services(SOAP) in the IoT context

(i) It reduces overhead(ii) It reduces parsing complexity(iii) It is stateless(iii) It integrates more tightly with HTTP [58](iv) It is easy to implement and learnis together

with its natural t over HTTP makes it apreferred method for Web 20 applications toshowcase their data and secure M2M sessionsin the IoT

(v) Applications that support RESTful web ser-vices exhibit better performance than the re-source-constrained WSN nodes [59]

(vi) REST is important for the IoT since it supportscommercial M2M cloud platforms

(vii) Its implementation in smart phones and tabletapps is also easy since it needs only an HTTPlibrary that is accessible to all operating systemplatforms [56]

Disadvantages

(i) Although it enjoys wide usage in commercialM2M platforms it is not likely to dominatebecause it is hard to implement

(ii) Its HTTP usage makes it incompatible withconstrained M2M devices leaving its usage toend applications

39 WebSockets It is a web-based protocol operating on asingle TCP channel and providing full-duplex transmissions

It permits secure duplex communication between a clientexecuting unreliable code in a restricted environment and adistant node communicating from that code while using theorigin-based security model employed by web browsers overTCP [60] e protocol is not a requestresponse model andis not a publishsubscribe model So unlike other protocolsa web-socket session is established by the client initiating ahandshake to the server to launch the session is onceestablished a full-duplex client-to-server connection beginsasynchronously and runs until terminated by either party[56] It gives a model for browser-based apps utilizing bi-directional communication with no need of many HTTPconnections to servers

WebSocket comprises two parts ie (i) the handshakehaving a clientrsquos message and the serverrsquos handshake responseand (ii) the data transfer HTTP headers should be repeated byapplications in all client requests and server responses arisingfrom the endless polling which might increase the commu-nication overhead subject to the application [61] WebSocketcan be used to come up with scalable real-time web appli-cations since it can enable full-duplex transmission via onesocket across the Web Its security can be ensured with thehelp of TLSSSL is protocol is suitable for resource-con-strained devices and IoTapplications will not have a good userexperience with it since it operates a clientserver architectureNevertheless it can outcompete all TCP-based protocolsbecause of the following advantages

(i) It is t for real-time transmission(ii) It has improved security(iii) It minimizes transmission overhead(iv) It can provide ecient messaging systems using

WebSocket ApplicationMessaging Protocol (WAMP)[17]

310 Streaming (Simple) Text-Oriented Messaging Protocol(STOMP) is is a TCP-based protocol designed for textmessage-oriented middleware (MOM) with the use of

Topic

Datareader

Datareader

Datareader

Datawriter

Datawriter

Datawriter

Subscriber PublisherPublisher

Data bus (DDS domain)

Subscriber

Topic Topic

Figure 7 DDS architecture

Journal of Computer Networks and Communications 23

HTTP-like commands for interoperability among platformslanguages and brokers Data communication takes placebetween the client and the broker in line with commands(like CONNECT DISCONNECT ACK NACK SUB-SCRIBE UNSUBSCRIBE SEND BEGIN COMMIT orABORT) then headers in the form ltkeygt ltvaluegt (one perline) then a blank line and then body content terminatingin a null character Server-client communication takes placeover a MESSAGE RECEIPT or ERROR frame with headerand body content format being analogous MOM productssupporting STOMP include Apache ActiveMQ or FuseMessage Broker HornetQ Open Message Queue(OpenMQ) RabbitMQ and syslog-ng through its STOMPdestination plugin [62] In addition to these other appli-cation layer protocols include simple media control protocol(SMCP) lightweight local automation protocol (LLAP)simple sensor interface (SSI) lightweight M2M (LWM2M)XMPP-IOT and simple object access protocol (SOAP)

Advantages

(i) It is lightweight for constrained networks(ii) It has flexibility to choose quality of services with

the given functionality(iii) It is standardized by OASIS Technical Committee(iv) It is easy and quick to implement

Challenges and opportunities

Here security threats are imminent because of the open-endedness of the application layer [63] -ese could includethe following

(i) -ere is poor or lack of a security design for anapplicationrsquos function Coming up with a gooddesign to address this could be a good ideaopportunity

(ii) Some of the programs might have backdoorsevading otherwise secure controls to allow un-authorized access to network resources We requiremechanisms to patch up such loopholes

(iii) Applications that lack or have weak authenticationmechanisms can be easily targeted by unlicensedusers to abuse the network Such mechanisms needto be strengthened further

(iv) Excessively complicated access control rules maybecause of their misunderstanding or poor writinggive unauthorized access allowing dubious activityrather than protecting the network they purport toprotect -ese should be simple and clear to un-derstand by all users

(v) In booting workstations without disks and managingnetwork devices we usually use the TFTP protocolalthough access does not require authenticating theuser by usernamepassword -is makes an intruderable to gain easy access to configaccess info as longas he can guess the filenames -is needs to be fixed

A few steps can be used to secure the application layerTo protect network data transmitted across the network

applications need to strengthen their encryption and au-thentication mechanisms Apps too need to enable strictcontrols of privileges for data access using good mechanismsto balance between usability and effectiveness [63] Appli-cations handling sensitive data need detailed logging andaudit capability as well as testing and reviewing

Table 3 summarizes the application layer protocols dis-cussed with their weaknesses and differences among them

4 Cross-Layer QoS Strategies

41 Service-Differentiated Real-Time Communication Scheme(SDRCS) -is is an event-driven routing protocol that routesreal-time traffic by featuring a cross-layer packet-forwardingdesign in which it integrates the real-time routing functionalitywith a new priority-based MAC scheme -e protocol ap-proximates distributed packet traversal speed for traffic clas-sification and admission control on the basis of this design Itfurther localizes decision-making by prioritizing packet for-warding to maximize the speed of packets [64]

Advantages

(i) It improves bandwidth utilization bandwidthdegradation due to unschedulable packets can beavoided

(ii) It offers a lightweight packet schedulability estima-tion mechanism Here it uses received signalstrength and admission control and then earlymissed-deadline packet-dropping policies are made

(iii) It delivers high end-to-end throughput via highersource data rates alongside strict end-to-end latencyneeds

(iv) Nodes which meet real-time requirements of lesstraffic load and better channel quality are givenhighest priority when it comes to packet forwarding

-e SDRCS operates via 5 mechanisms namely (1) RSS-(Rich Site SummaryReally Simple Syndication-) basedgrouping (2) admission control (3) prioritized queuing(4) real-time MAC and (5) dynamic forwarding -isprotocol has a queuing policy that employs per-hopdeadline-based queues and the nodes schedule theirpackets using FIFO priority-based queues [1] -e prin-cipal features of this protocol are mentioned as follows

(i) It is event driven and is easily adaptable to anynetwork changes

(ii) It does not need additional hardware for locali-zation or multichannel transmission

(iii) -e cross-layer packet-forwarding strategy can beused when transmitting multimedia traffic

(iv) It is able to efficiently circumvent any voids therein

Early deadline miss (EDM) policy drops all unscheduledpackets -e SDRCS circumvents wrong packet dropsdue to EDM since it adjusts to network changes easily

42 Sensor Fuzzy-Based Image Transport (SUIT) -is pro-tocol regulates congestion using a scheme based on fuzzy

24 Journal of Computer Networks and Communications

logic It transmits packets with reduced quality to allowablelevels if there is congestion enabling it to transmit lower-quality packetsframes without being dropped-e numberof packets delivered per second thus rises and while thisenhances video streaming apps it is more appropriatelymade for those that quickly transmit non-real-time videosuch as in video surveillance SUIT can stream JPEG pic-tures other than the traditional predictive video codingmethods that are not suitable for sensor devices because oftheir constraints of energy memory and computationalspeed Progressive JPEG-PJPEG performs better via packettransmission rate and delay though it degrades the imagequality because of congestion [65] -is protocol employscross-layer methods for interlayer data exchange though itstill employs some layering making it still possible topreserve all advantages of the layered design since it hasapplication and transport layers but has noMAC or routinglayer

For congestion detection and control SUIT bases on thepercentage number of in-bound and out-bound frames perwindow number of contenders and buffer occupancy of thenext-hop node Since buffer management has an effect oneffectiveness and QoS inWMSNs it is useful in the transportlayer so the protocol prioritizes packets for better QoS SUITreorders packets by tracking the sequence numbers of thedelivered packets since they all have a source ID framenumber and sequence number that it uses to detect anymissing packets -e protocol however does not providepacket reliability [65]

Advantages

(i) -e protocol prioritizes packets for better QoS(ii) It can transmit lower-quality packetsframes

without being dropped

Disadvantages

(i) It does not provide packet reliability

43 Network Layer QoS Support Enforced by a Cross-LayerController (NLQS) -is scheme permits packet-level servicedifferentiation as a function of throughput end-to-endpacket error rate and delay [66] (see Figure 8) -is en-hances network layer QoS It comprises a cross-layer controlunit (XLCU) for configuring and controlling networkingfunctionalities at physical MAC and network layers [18]-is depends upon the unified logic which affects choices forapplication layer needs and status of functional blocks whichdo implementation of networking functions [66]

Advantages

(i) Cross-layer interactions can be controlled withoutweakening the upgradability modularity andsimplicity of designing the system

44 Cross-Layer Signaling Shortcuts (CLASS) CLASS has ahigh level of effectiveness flexibility and comprehensive-ness According to [67] it has the following features

(i) Direct signaling among remote layers leading tofaster transmission

(ii) Lightweighted internal message format standard-ized protocols are not suitable for use in internalsignaling since they are not usually lightweight forinstance transmitting against faults in the networkCLASS requires only three fields namely destina-tion address event type and event contents

(iii) -e setup of its exterior messages is standardconsidering external signaling ICMP and TCPIPare suitable for general messages and short notifi-cations respectively

(iv) Message control protocol should ensure optimizedand organized exchange of dense simultaneousmessages across layers for high efficiency andavoidance of possible conflicts -e protocol appliesto many cross-layer signaling situations

Table 3 Differences and problems with application layer protocols

Protocol Architecture Transport QoS options Security Weaknesses

CoAP Requestresponse UDP DTLS No in-built security features DTLS does not supportmulticast

MQTT Publishsubscribe TCP TLSSSL Insufficient security at the protocol level sincepayload values are not encrypted

SMQTT Publishsubscribe TCP CPKP-ABE Key revocation and group pubsub for distributedSMQTT are still a challenge

MQTT-SN Publishsubscribe UDP TLSSSL SSLTLS suffers from attacks like BEAST CRIMERC4 and Heartbleed

XMPP RequestresponsePublishsubscribe TCP X TLSSSL Susceptible to spamming attacks and lacks worldwide

supportAMQP Publishsubscribe TCP TLSSSL Inappropriate for real-time applications

DDS Publishsubscribe TCPUDP TLSSSL Memory intensive and no open-source libraries forconstrained devices

REST Requestresponse HTTP X HTTPS Hard to implement Uses HTTP so incompatiblewith constrained apps

WebSockets PublishsubscribeClientserver TCP X TLSSSL Bad user experience for IoT apps since it runs the

clientserver architectureSTOMP Clientserver TCP HTTP -e broker can act as a bottleneck

Journal of Computer Networks and Communications 25

Advantages

(i) e protocol is scalable and has a very lowpropagation delay

(ii) It is very ecient and exible since it uses directsignaling between any two layers

(iii) It is very comprehensive and eective

45 Cross-Layer and Multipath-Based Video Transmission(CMVT) CMVT is a hybrid of application and networklayers At the application layer it carries out the encoding ofvideo streams to video data frames (I- P- and B-frames) bymeans of the MPEG-4 encoding format At the networklayer route discovery and data transfer take place Twoalgorithms greedy forwarding and rollback are used todiscover a number of paths from the source to the sink nodevia the route discovery method [20] A specied node icalculates the evaluation of its adjacent node j using thefollowing equation

fij (1 minus α)d2(j D) minus d2min(i)d2max(i) minus d

2min(i)

+ αeinit(j) minus eres(j)

einit(j)

(19)

where fij evaluation value for the node i to j d2(j D) isthe distance from the node j to the destination node D

d2min(i) and d2max(i) are the minimum and maximum dis-

tances of neighbors of the node i to D respectively einit(j) isthe initial energy of the node j eres(j) is the current residualenergy of the node j and α is the energy coecient given asfollows

α emax(i) minus emin(i)

emax(i) (20)

where emax(i) and emin(i) are respectively the maximumand minimum energy left of all node irsquos neighbors

e network layer further transmits video streams withCMVT doing status evaluation choosing an appropriatecommunication path for any packets and the QoS guaranteelevel for the path i is calculated as

fi (1 minus ω)hisum hi

+ ωnisum ni

(21)

where fi is the evaluation value for the path i hi are the hopsfor the path i ni is the summation of packets sent via isum ni isthe summation of packets sent by sources and ω is theenergy consumption factor [18]

Advantages

(i) e CMVT protocol is very superior in mediatransmission particularly large-scale WMSNs

To network peers

Flow requirements

-

Atimes

Admissioncontrol

Capacityassignment

Next-hopselection

Coding rate

Data rate

QoScontracts

Cross-layer controlunit (XLCU)

Localization subsystem

Synchronizationsubsystem

Networkcoordinationunit (NCU)

UWB modulator

Channel coder

QoS scheduler

Georouting

Source coder

Rate control

Figure 8 Cross-layer controller architecture

26 Journal of Computer Networks and Communications

46 Cross-Layer Cooperative MAC (CoopMAC)CoopMAC [68] is based on the IEEE 80211 DCF approach-ere is a reduction in interference of neighboring cellsleading to uniform coverage in densely deployed networks[68] A connection is established by two-way authenticationbetween the sender and the receiver When a Clear-to-Send(CTS) packet and short interframe space (SIFS) are obtainedby the sender packets are transmitted right to the receiverwith no valuable cooperation If available the two nodes findout whether they can exchange data -is is done by the useof a helper identification signal to establish how feasiblecooperative communication is It is established in thepresence of a signal and if it is not available direct com-munication is activated For distributed systems helper-initiated cooperation is preferable because of the presence ofthe RTSCTS packets [69]

Advantages

(i) CoopMAC has an advantage over others due to thespatial diversity amongst the three nodes [70]

(ii) It has the potential to achieve significantthroughput and delay with no complexity in thedesign of the system

47 Cooperative MAC Protocol for Multihop Networks (M-CMAC) Just as CoopMAC M-CMAC is made in such away that high data rate stations help low data rate stations toforward traffic for broadcasting [71] -ere are also helperschosen in such a way that 2 fast-hop transmissions replace aslow hop transmission -e helper via whom there existsleast delay from the sender to the receiver is considered bestbecause of theminimum two-hop transmission rate and so isthe chosen neighbor to the duo Here the working as-sumption is that all nodes have their position coordinatesknown and thus the Euclidean distances between all pairscan be calculated and converted to the data rate for that linkAll nodes have cooperative tables (CT) of potential helperswith destination and helper MAC address Euclidean dis-tance and total distance through the helper Any sender withdata to transmit checks out for the existence of any helper inthe CT for the destination and if available forwards RTS forchannel reservation for single-hop duration M-CMAC hasgot an RTS format having a shape with five fields similar tothe one given as follows

Framecontrol Duration Source

addressDestination

addressHelperaddress

-e destination nodersquos address is kept by the source in thepart for the helper address that is in turn saved as the desti-nation address Nodes inspect helper and destination addressfields on receipt of RTS whereby the node acts as the othernodersquos helper in case the helper address field is not the sameand the helper transmits CTS back to the source if at all it wantsto forward data On delivery of the CTS packet the sourcetransmits to the helper which in turn sends the packets to thedestination that also sends ACK to the helper on receipt of thepacket In M-CMAC there exists a higher level of channelreuse (parallel transmissions) due to the increased number of

nodes therein that subsequently raises availability of helpers fordata forwarding -is results in increased throughput incomparison with CoopMAC and IEEE 80211 DCF [71]

Advantages

(i) It promises a higher throughput compared toCoopMAC and IEEE 80211 DCF

48 Cluster-Based Cooperative Routing (CBCR) Protocol-e CBCR protocol has a multihop data-forwarding func-tion realized at the link layer with cooperative links that useM-CMAC -is protocol encompasses two stages namelythe routing relay selection phase and data forwarding phase

(i) Routing relay selection phase all nodes announcetheir presence to their neighbors by broadcastingperiodical beacon messages that carry the MAC ad-dress of the node Each of these constructs a relaytable that contains all neighboring nodes with whichit is able to communicate -e node further broad-casts its neighbor list in case of any change to itsentries since the previous broadcast -e MAC ad-dresses of nodes next to the node X are contained inits relay tablersquos column one and the neighbor nodersquosrow has MAC addresses for neighbors of the adjacentnode Based on its relay table each node in-dependently chooses routing relays Choosing a nodeas a relay node depends on the number of nodes itconnectsmdashit should be the highest number [71]

(ii) Data forwarding phase a node with packets totransmit needs to first verify if the receiver is in thesame cluster and if that receiver has a helper to whichthe packet is thence forwarded but in whose absencethe packet is delivered right to the receiver In casethe intended receiver is in a different cluster therelay table will be checked to see whether that re-ceiver is reachable via other routing relays If it isreachable packets will be sent to the routing relaydirectly or via the helper if at all the relay has one Incase of destination unreachable via relays packetsare multicast to all relays by the node [71]Advantages

(i) It enables multicasting

49 MAC-PHY Cross-Layer Protocol A cooperative cross-layer standard for cooperation at the physical layer in next-generation WMSNs is developed in [70] It provides acomplete MAC layer algorithm that gives a supportive shellfor the PHY-MAC layer Like CoopMAC its scheme de-pends on the middle node for communication betweennodes-eMAC layer protocol has been changed to regulateinformation exchanges at the physical layer -e receivernode receives duplicate packets from source and helpernodes to decode the data [70]

410 MAC-Centric Approach -is cross-layer protocoltargets multimedia applications by using the MPEG-4

Journal of Computer Networks and Communications 27

scheme [21] It is characterized by 4 access categories(AC3ndashAC0) according to their priority in transmission It ismeant for supporting different QoS needs in upcomingvideo applications and enabling MAC layer differentiationfor H264 partitioning [18] Because of low bandwidth delayand other QoS challenges that cause inefficiency in trans-mittingmultimedia data inWMSNs a number of algorithmsdepending on IEEE 80211e have been proposed to supporttransmission of quality videos [72] Selecting an AC dependson the measures for QoS such as the loss rate and delay-us AC3 having the highest priority is mapped to theparameter set concept since the stream is sensitive to bit lossespecially in video transmission [21]

411 Adaptive Cross-Layer Forward Error Correction(ACFEC) In the ACFEC [73] model data packets are ex-changed among nodes through the access point (AP) whichoperates the infrastructure mode in which it adds FEC to videodata [72] -ese data are dealt by encapsulation using astreaming server to the receiver through thewireless AP as RTPpackets An adaptive FEC controller senses the category ofpackets out of the RTP header and recovers the header ofpackets from UDP -e encoder generates some error-cor-recting packets whose number is determined by the sourcepacket number of the block Multimedia transmissions aremonitored by the controller using MAC failure data and itscounter is increased by one if there is a failed transmission-econtroller uses the failure counter to change the packet numberproduced after transmitting a block [18] In case there are lostpackets it changes the redundancy rates and produces extrapackets to replace the lost ones and satisfy the requirements ofthe receiving node [72] -e FEC packet number is enhancedor reduced to satisfy the receiverrsquos requirements and stoppacket losses-is is done through accurate detection of packetlosses and adjustment of redundancy rates If all video datapackets are well received there will not be generation of FECpackets [73]

Advantages

(i) It promises a good QoS through packet loss re-duction and redundancy rate adjustment

412 Balanced Cross-Layer Fuzzy Logic (BCFL) DesignRouting Algorithm in WSNs A new fuzzy logic-basedrouting algorithm (BCFL) [74] was designed using disper-sion of the cross-layer parameter as the fuzzy logic inferencesystem input Each cross-layer parameter has a dynamicweight depending on the value of dispersion -e designcomes with some innovations as follows (i) for fuzzy logicinference system input the absolute parameter value issubstituted by parameter dispersion thereby significantlyreducing algorithmic complexity (ii) dispersion does notchange with the order of magnitude according to the dis-persion formula and (iii) the weight of the parameter de-pends on the size of its dispersion and the two are inverselyproportional to each other -is enables BCFL to have somenobility unlike other algorithms as per the following dis-tinguishing properties making it advantageous

(i) It has got simple if-then rules which remain con-stant even when the constraints increase

(ii) It is capable of dealing with many constraints withno increment in complexity

(iii) It has capacity to yield a more balanced solutionthan other algorithms

(iv) It is easily adaptable to changes in network con-ditions and topology even when the changes arefrequent like in underwater WSNs -e algorithm isuseable in choosing the CH in cluster-based routingprotocols [74]

413 Minimum Hop Disjoint Multipath Routing Algorithmwith Time Slice Load-Balancing Congestion Control SchemeMHDMwTS is a two-phased routing protocol comprisingpath build-up and path acknowledgment phases It consists ofmultiple sources each of which has three build-up disjointpaths namely primary alternate and back-up paths Onactivation the source node requests to build up a route to thenearest hop neighbor which is the path build-up phase [19] Inthis phase step one has source activation in which therequested node for path building adds its number and time-stamp and sends to the least hop-count neighboring node-isgoes on until the least time latency sink having the informationneeded to construct the primary route is reached [75] In steptwo there is path extraction when the new package fromanother path arrives whereby the extracted path is comparedwith the primary path If the node is shared the package will berejected or an alternative path will be searched for to get abackup through comparison of the preceding two paths Phasetwo is path acknowledgment where we have the third step inwhich the sink returns the ACK packet to the sender with pathinformation having nodes and their time information after ithas computed it using the timestamp [1819]

Advantages

(i) -e protocol minimizes end-to-end latency(ii) It enhances congestion control

414 Cross-Layer Optimal Design (CLOD) Authors in [76]come up with a CLOD for scheduling at the data link layerrouting at the network layer and controlling congestion atthe transport layer with an assumption of fixed link capacity-rough congestion control energy efficiency is improvedTransport layer congestion at the nodes is minimized bycompressed sensing (CS) in which transmitted bits are re-duced whereas optimally allocating resources decreasescongestion on the links at the data link layer CLODpromises minimized computational complexity and betterperformance in light traffic scenarios [77] Generating andstoring CLOD is more efficient compared to Gaussianrandom matrices It prolongs network lifetime and savesenergy

Advantages

(i) It does congestion control subsequently increasingenergy efficiency

28 Journal of Computer Networks and Communications

(ii) It reduces computational complexity(iii) It improves performance in light traffic scenarios

415 Challenges and Opportunities of the Cross-Layer Design(CLD)

(i) Different layers in a real network assume differentfunctions andor services A layer only communi-cates with its neighboring layers -e layered modeldegrades system performance because of many

features in wireless transmission QoS parameters inthe layers of WMSNs are shared by routing pro-tocols to optimize the performance Neverthelessthere is a need to design cross-layer models forincreased efficiency in routing [1]

(ii) -e physical layer plays a highly vital role in CLDRate adaptation and channel allocation take place atthe physical layer via signal processing to enhanceQoS End-to-end performance in wireless media isaffected by the changes therein given the effect on

Table 4 A summary of some of the reviewed cross-layer models

Protocol Layers Aims Comments QoS parameters

SDRCS MACPHY Routing real-time traffic Transmits multimedia traffic viacross-layer packet forwarding

-roughputlatency

SUIT Applicationtransport Transmission of non-real-time video Uses fuzzy logic to regulate

congestionTransmission rate

delay

CLASS Any twoDesign serves as a framework fordifferent implementations ofdifferent application scenarios

Scalable very efficient flexible andhas very low propagation delay

Propagation delayjitter

NLQS PHYMACnetwork

Permits packet-level servicedifferentiation as a function of

throughput packet error rate anddelay

Network layer QoS is enhanced -roughput delaypacket error rate

CMVT Applicationnetwork

Encodes video streams to video dataframes via MPEG-4 encoding anddoes route discovery and data

transfer

Greedy forwarding and rollback areused to find source-to-sink paths Energy

CoopMAC PHYMAC Offers spatial diversity among thethree nodes

Helper ID signal finds out howfeasible cooperative communication

is

-roughput ratedelay

M-CMAC PHYMAC Increases end-to-end throughputand packet delivery ratio

Euclidean distances between nodesare calculated and converted to the

data rate for a link

-roughputpacket delivery

ratio

CBCR PHYMACMinimizes control overhead andtime consumed in establishing thecooperative paths than M-CMAC

Energy consumption is moreuniformly distributed in a network

enhancing network lifetime

-roughputpacket deliveryratio energy

MAC-PHY MACPHYMeant to enable PHY layer

cooperation and maximize gains ofcooperation at the MAC layer

Leverages both spatial diversity andcoding gain -roughput delay

MAC-centric MACAPPL

Meant to support QoS needs in newvideo apps and enable MAC layer

differentiation for H264partitioning

Targets multimedia applications byusing the MPEG-4 scheme

Delay packet lossrate

ACFEC MACnetwork(UDP)

Meant to enhance the quality ofvideo streaming over 80211WLANs

and overcome packet losses

Adjusts redundancy rates toovercome channel fluctuations and

detect and reduce packet lossPacket loss rate

MHDMwTS mdash

Meant to provide reliable datatransfer with the multipath routingand load-balancing congestioncontrol method in WMSNs

More reliable than basic routingschemes for transport multimedia

data

Latency packagetransmit rate

BCFL mdash

Introduces dispersion into fuzzylogic-based routing and sets every

cross-layer parameter with adynamic weight

Can be used to select a CH in cluster-based routing protocols andproposes a dispersion formula

Node utilitydispersion

CLODDatalinknetworktransport

Prolongs network lifetime andachieves congestion control anddesigned for lightly loaded WSNs

Assumes fixed link capacity andintegrates compressed sensing

technology

-roughputaverage energy CS

error ratio

Journal of Computer Networks and Communications 29

the purposefulness of the protocols at the networklayer CLD offers solutions to conserving powerminimizing energy and controlling both the flowand congestion in the network as well as fault tol-erance so it is an opportunity for designers con-sidering other layers We further desire to developCLDs that shall make an impact on network oper-ation and get close attention [18]

(iii) CLD interface standardization the architecturemust deliver module functionality although thereexist some queries amongst modules about possibleinterfaces predicted by the necessity to share in-formation among remote protocol layers Technicalchallenges include developing designing andstandardizing of cross-layer interfaces and algo-rithms that satisfy the cross-layer optimization re-quirements among protocol layers

Table 4 summarizes these cross-layer models with a fewcomments

5 Future Research Direction

51 At theMACLayer In WMSNs achieving duty cycling istricky since video traffic is volatile Listening to adjacentnodes wastes energy We need to switch states from awake tosleep -is requires further research

Self-organizing networks and green communicationalthough most CRWSN energy-saving algorithms aredesigned to reduce node transmission power energyconsumption is inevitable during the operational mode ofBSs because of internal processing Scholars are trying withpolicy-makers to encourage a shift to greenWSNs to reduceoperational costs and carbon footprints With self-orga-nizing CRWSNs for automatic switch-off when idle energyis saved which calls for cognitive switching algorithms Sointegrating cognitive switching with radio resource allo-cation in CRWSNs is a promising research area [78]

52 At the Application Layer Considering a WMSN ar-chitecture end users access the network via the BS to andfrom which they send their sensed data We desire to getguaranteed QoS and energy efficiency by designing a newarchitecture in the form of hardware and software whichneeds to be investigated further for a fix of the same [1]

QoS and energy efficiency are very crucial in WMSNsand mostly real-time apps that require guaranteed band-width and throughput in their network lifetime Mostprotocols ignore base station (BS) mobility and WMSNnodes Traffic management telemedicine and battlefieldsurveillance apps require mobile nodes or BSs so designingdynamic routing protocols to be adaptable in these cir-cumstances is necessary

53 Cross-Layer Security of data transmitted on WMSNseg in military surveillance and e-commerce is key withQoS and energy efficiency WMSNs are vulnerable to attacks

like worms and sinkhole Moreover their computationpower is not that high complicating implementation ofstrong secure protocols -is needs further attention

For fairness and priority issues in CRWSNs nodes areassigned diverse priorities depending on importance andurgency of their data For better network performance thereis a need to develop fair resource allocation mechanismswith priority-based fairness in sensors-ere is still potentialin this research area [78]

We must develop multiobjective adaptive protocols tooptimize QoS metrics involved in routing for best trans-mission results to get a trade-off between different opti-mization metrics We need to design standardized cross-layer algorithms which are able to meet cross-layer opti-mization standards

6 Conclusion

We have reviewed the different QoS strategies for WSNs inthe context of IoT from the MAC layer and application layeras well as the cross-layer paradigm For the MAC layer wehave reviewed protocols for WSNs such as the hybrid of thecontention-free and contention-based MAC protocols andthose for WMSNs We further reviewed a number of ap-plication layer protocols including many machine-to-ma-chine requestresponse and publishsubscribe protocols Forsystem optimization cross-layer QoS strategies are veryimportant in wireless communication We have reviewed anumber of cross-layer strategies For all categories reviewedchallenges and opportunities are discussed Finally somepossible future directions are discussed for research andapplication showing a promising potential for future re-search in this relatively new area especially as more WMSNapplications emerge recently

Conflicts of Interest

-e authors hereby declare no conflicts of interest regardingthe publication of this work

Acknowledgments

-is research was funded by the National Natural ScienceFoundation of China (Nos 61602398 61672447 and61711540306)

References

[1] V Bhandary A Malik and S Kumar ldquoRouting in wirelessmultimedia sensor networks a survey of existing protocolsand open research issuesrdquo Journal of Engineering vol 2016Article ID 9608757 27 pages 2016

[2] I F Akyildiz T Melodia and K R Chowdhury ldquoWirelessmultimedia sensor networks applications and testbedsrdquoProceedings of the IEEE vol 96 no 10 pp 1588ndash1605 2008

[3] M Iqbal M Naeem A Ahmed M Awais A Anpalagan andA Ahmad ldquoSwarm intelligence based resource managementfor cooperative cognitive radio network in smart hospitalsrdquoWireless Personal Communications vol 98 no 1 pp 571ndash5922018

30 Journal of Computer Networks and Communications

[4] S A Alvi B Afzal G A Shah L Atzori and W MahmoodldquoInternet of multimedia things vision and challengesrdquo AdHoc Networks vol 33 pp 87ndash111 2015

[5] Q F Hassan A U R Khan and S A Madani Internet ofings Challenges Advances and Applications CRC PressTaylor amp Francis Group A Chapman amp Hall Group BocaRaton FL USA 2018

[6] Q Wang Y Zhao W Wang et al ldquoMultimedia IoT systemsand applicationsrdquo in Proceedings of the 2017 Global Internet ofings Summit (GIoTS) vol 2 Geneva Switzerland June 2017

[7] A Aslam and E Curry ldquoTowards a generalized approach fordeep neural network based event processing for the internet ofmultimedia thingsrdquo IEEE Access vol 6 pp 25573ndash25587 2018

[8] M Jridi T Chapel V Dorez L Bougeant and A Le BotlanldquoSoC-based edge computing gateway in the context of theinternet of multimedia things experimental platformrdquoJournal of Low Power Electronics and Applications vol 8no 1 p 1 2018

[9] H Noura A Chehab L Sleem andM N Rapha ldquoOne roundcipher algorithm for multimedia IoT devicesrdquo MultimediaTools and Applications vol 77 no 14 pp 18383ndash18413 2018

[10] T Balan D Robu and F Sandu ldquoMultihoming for mobileinternet of multimedia thingsrdquo Mobile Information Systemsvol 2017 Article ID 6965028 16 pages 2017

[11] H Shen and G Bai ldquoRouting in wireless multimedia sensornetworks a survey and challenges aheadrdquo Journal of Networkand Computer Applications vol 71 pp 30ndash49 2016

[12] A Ahmad M H Rehmani H Tembine O A Mohammedand A Jamalipour ldquoIEEE access special section Editorial optimization for emerging wireless Networks IoT 5G andsmart grid communication networksrdquo IEEE Access vol 5pp 2096ndash2100 2017

[13] B A Muzakkari M A Mohamed M F A Kadir andZ Mohamad ldquoRecent advances in energy efficient-QoS awareMAC protocols for wireless sensor networksrdquo InternationalJournal of Advanced Computer Research vol 8 no 38pp 212ndash228 2018

[14] M A Yigitel O D Incel and C Ersoy ldquoQoS-aware MACprotocols for wireless sensor networks a surveyrdquo ComputerNetworks vol 55 no 8 pp 1982ndash2004 2011

[15] A M Abbas and O Kure ldquoQuality of Service in mobile ad hocnetworks a surveyrdquo International Journal of Ad Hoc andUbiquitous Computing vol 6 no 2 p 75 2010

[16] M Q Shatnawi ldquoApplication layer protocols for the internetof things a surveyrdquo in Proceedings of the InternationalConference on Engineering amp MIS (ICEMIS) Agadir Mo-rocco September 2016

[17] V Karagiannis P Chatzimisios F Vazquez-gallego andJ Alonso-zarate ldquoA survey on application layer protocols forthe internet of things research motivationrdquo InternationalJournal of Advanced Research in Computer Science vol 8no 3 pp 1ndash10 2015

[18] M S Bernard T Pei Z Li and K Li ldquoQoS strategies forwireless multimedia sensor networks in the context of IoTrdquoLecture Notes of the Institute for Computer Sciences SocialInformatics and Telecommunications Engineering vol 4pp 228ndash253 2019

[19] A AlAmri and M Abdullah ldquoCross-layer quality of serviceprotocols for wireless multimedia sensor networksrdquo in Pro-ceedings of the International Conference on CommunicationManagement and Information Technology pp 649ndash658Cosenza Italy April 2017

[20] J Guo L Sun and R Wang ldquoA cross-layer and multipathbased video transmission scheme for wireless multimedia

sensor networksrdquo Journal of Networks vol 7 no 9pp 1334ndash1340 2012

[21] A Ksentini A Naimi and M Gueroui ldquoToward an im-provement of H264 video transmission over IEEE 80211ethrough a cross-layer architecturerdquo IEEE CommunicationsMagazine vol 44 no 1 pp 107ndash114 2006

[22] W Ye J Heidemann and D Estrin ldquoAn energy-efficientMAC protocol for wireless sensor networksrdquo in Proceedings ofthe 21st International Annual Joint Conference of the IEEEComputer and Communications Societies (INFOCOM 2002)New York NY USA June 2002

[23] J Polastre J Hill and D Culler ldquoVersatile low power mediaaccess for wireless sensor networks categories and subjectdescriptorsrdquo in Proceedings of the Second InternationalConference on Embedded Networked Sensor Systems Sen-Sysrsquo04 pp 95ndash107 Baltimore MA USA November 2004

[24] T Danmanee K N Nakorn and K Rojviboonchai ldquoCU-MAC a duty-cycle MAC protocol for internet of things inwireless sensor networksrdquo ECTI Transactions on ElectricalEngineering Electronics and Communications vol 16 no 2pp 30ndash43 2018

[25] T Van Dam and K Langendoen ldquoAn adaptive energy-efficientMAC protocol for wireless sensor networksrdquo in Proceedings ofthe 1st International Conference on Embedded NetworkedSensor Systems Los Angeles CA USA November 2003

[26] P Xie and J Cui ldquoR-MAC an energy-efficient MAC protocolfor underwater sensor networksrdquo in International Conferenceon Wireless Algorithms Systems and Applications (WASA2007) Chicago IL USA August 2007

[27] A Rahim N Javaid M Aslam U Qasim and Z A KhanldquoAdaptive-reliable medium access control protocol forwireless body area networksrdquo in Proceedings of the AnnualIEEE Communications Society Conference on Sensor Meshand Ad Hoc Communications and Networks (SECON) SeoulKorea June 2012

[28] H S Kwak S Ullah D H Kwak and C H Lee ldquoA powerefficient MAC protocol for wireless body area networksrdquoJournal of the Korean Institute of Intelligent Systems vol 8no 6 pp 131ndash140 2009

[29] C Omeni A J Burdett and M Park ldquoEnergy efficientmedium access protocol for wireless medical body area sensornetworksrdquo in Proceedings of the 4th IEEEEMBS InternationalSummer School and Symposium on Medical Devices andBiosensors pp 29ndash32 Cambridge UK August 2007

[30] L Sitanayah C J Sreenan and K N Brown ldquoER-MAC ahybrid MAC protocol for emergency response wireless sensornetworksrdquo in Proceedings of the Fourth International Con-ference on Sensor Technologies and Applications pp 244ndash249Mestre Italy July 2010

[31] Y Sun and D B Johnson ldquoRI-MAC a receiver-initiatedasynchronous duty cycle MAC protocol for dynamic trafficloads in wireless sensor networksrdquo in Proceedings of the 6thACM conference on Embedded network sensorsystemsmdashSenSysrsquo08 pp 1ndash14 Raleigh NC USA November2008

[32] M Buettner G V Yee E Anderson and R Han ldquoX-MAC ashort preamble MAC protocol for duty-cycled wireless sensornetworksrdquo in Proceedings of the 4th International Conferenceon Embedded Networked Sensor Systems SenSys 2006pp 307ndash320 Boulder Colorado USA October 2006

[33] S Boulfekhar and M Benmohammed ldquoSynchronous receiverinitiated MAC protocol for long-lived sensor networksrdquoComputers amp Electrical Engineering vol 40 no 2 pp 504ndash516 2014

Journal of Computer Networks and Communications 31

[34] S Mehta and K S Kwak ldquoH-MAC a hybrid Mac protocol forwireless sensor networksrdquo International Journal of ComputerNetworks amp Communications vol 2 no 2 pp 108ndash117 2010

[35] I Rhee A Warrier M Aia J Min and M L Sichitiu ldquoZ-MAC a hybrid MAC for wireless sensor networksrdquo IEEEACM Transactions on Networking vol 16 no 3 pp 511ndash5242008

[36] S Arshad A Al-sadi and A Barnawi ldquoZ-MAC performanceevaluation and enhancementsrdquo Procedia Computer Sciencevol 21 pp 485ndash490 2013

[37] Y Liu I Elhanany and H Qi ldquoAn energy-efficient QoS-aware media access control protocol for wireless sensornetworksrdquo in Proceedings of the IEEE International Confer-ence on Mobile ad hoc and Sensor Systems pp 9ndash11 Wash-ington DC USA December 2005

[38] J Decotignie ldquoWiseMAC an ultra low power MAC protocolfor the downlink of infrastructure wireless sensor networksrdquoin Proceedings of the Ninth International Symposium onComputers And Communications pp 244ndash251 AlexandriaEgypt June 2004

[39] R A Rashid ldquoDevelopment of energy aware TDMA-basedMAC protocol for wireless sensor network systemrdquo EuropeanJournal of Scientific Research vol 30 no 4 pp 571ndash578 2009

[40] H P Van Hoesel ldquoA lightweight medium access protocol(LMAC) for wireless sensor networksrdquo in Proceedings of theInternational Workshop on Networked Sensing Systems Bal-timore MD USA November 2004

[41] C Li P Wang H Chen andM Guizani ldquoA cluster based on-demand multi-channel MAC protocol for wireless multi-media sensor networksrdquo in Proceedings of the IEEE In-ternational Conference on Communications pp 2371ndash2376Beijing China May 2008

[42] J L Sobrinho and A S Krishnakumar ldquoQuality-of-service inad hoc carrier sense multiple access wireless networksrdquo IEEEJournal on Selected Areas in Communications vol 17 no 8pp 1353ndash1368 1999

[43] J P John ldquoProviding MAC QoS for multimedia traffic in80211e based multi-hop ad hoc wireless networksrdquo ComputerNetworks vol 51 no 1 pp 153ndash176 2007

[44] S Lofty and N Padmavati ldquoA survey on mobility basedprotocols in WSNsrdquo in Proceedings of the InternationalConference Communication and Manufacturing pp 12ndash15Hangzhou China July 2014

[45] S Deng L Shen and J Li ldquoMobility-based clustering pro-tocol for wireless sensor networks with mobile nodesrdquo IETWireless Sensor Systems vol 1 no 1 pp 39ndash47 2011

[46] M Khalil A Khalid F U Khan and A Shabbir ldquoA review ofrouting protocol selection for wireless sensor networks insmart citiesrdquo in Proceedings of the 24th Asia-Pacific Confer-ence on Communications (APCC) Ningbo China November2014

[47] O Message Q Telemetry and T Mqtt Committee Specifi-cation Draft 02 April 2014

[48] M Collina M Bartolucci A Vanelli-coralli andG E Corazza ldquoInternet of things application layer protocolanalysis over error and delay prone linksrdquo in Proceedings ofthe 7th Advanced Satellite Multimedia Systems Conference andthe 13th Signal Processing for Space CommunicationsWorkshop Livorno Italy September 2014

[49] A Talaminos-barroso M A Estudillo-valderrama L M RoaJ Reina-tosina and F Ortega-ruiz ldquoA machine-to-machineprotocol benchmark for e health applicationsmdashuse case re-spiratory rehabilitationrdquo Computer Methods and Programs inBiomedicine vol 129 pp 1ndash11 2016

[50] V Goyal O Pandey A Sahai and B Waters ldquoAttribute-based encryption for fine-grained access control of encrypteddatardquo in Proceedings of the 13th ACM Conference on Com-puter and Communications Security pp 89ndash98 AlexandriaVA USA October 2006

[51] M Singh M A Rajan V L Shivraj and P BalamuralidharldquoSecure MQTT for internet of things (IoT)rdquo in Proceedings ofthe Fifth International Conference on Communication Systemsand Network Technologies Gwalior India April 2015

[52] A Stanford-Clark and H L Truong MQTT for Sensor Net-works (MQTT-SN) Protocol Specification httpmqttorgnewwp-contentuploads200906MQTT-SN_spec_v12pdf 2013

[53] RFC ldquoExtensible messaging and presence protocol (XMPP)rdquoin Proceedings of the IETF pp 1ndash9 San Diego CA USAAugust 2005

[54] S Vinoski and I Technologies ldquoQueuing protocolrdquo IEEEInternet Computing vol 10 no 6 pp 87ndash89 2006

[55] J L Fernandes I C Lopes J J P C Rodrigues and S UllahldquoPerformance evaluation of restful web services and AMQPprotocolrdquo in Proceedings of the Fifth International Conferenceon Ubiquitous and Future Networks (ICUFN) pp 2ndash7 DaNang Vietnam July 2013

[56] M Asim ldquoA survey on application layer protocols for internetof things (IoT)rdquo International Journal of Advanced Researchin Computer Science vol 8 no 3 pp 996ndash1000 2017

[57] K An A Gokhale D Schmidt S Tambe P Pazandak andG Pardo-castellote ldquoContent-based filtering discovery pro-tocol (CFDP) scalable and efficient OMG DDS discoveryprotocolrdquo in Proceedings of the 8th ACM InternationalConference on Distributed Event-Based Systems New YorkNY USA July 2014

[58] E Web What is a web service pp 52ndash57 2010[59] M Laine RESTful Web Services for the Internet of ings

pp 2ndash4 2012 httpmediaTkk[60] IETF -e WebSocket protocol pp 1ndash71 2011[61] V Pimentel ldquoCommunicating and displaying real-time data

with websocketrdquo IEEE Internet Computing vol 16 no 4pp 45ndash53 2012

[62] Wikipedia 2019 httpsenwikipediaorgwikiStreaming_Text_Oriented_Messaging_Protocol2019

[63] K Raghavendra and S Nireshwalya ldquoApplication layer se-curity issues and its solutionsrdquo IJCSET vol 2 no 6pp 1266ndash1269 2012

[64] Y Xue B Ramamurthy andM C Vuran ldquoSDRCS a service-differentiated real-time communication scheme for eventsensing in wireless sensor networksrdquo Computer Networksvol 55 no 15 pp 3287ndash3302 2011

[65] C Sonmez S Isik M Y Donmez O D Incel and C ErsoyldquoSUIT a cross layer image transport protocol with fuzzy logicbased congestion control for wireless multimedia sensornetworksrdquo in Proceedings of the 5th International Conferenceon New Technologies Mobility and Security (NTMS) pp 1ndash6IEEE Istanbul Turkey May 2012

[66] T Melodia and I Akyildiz ldquoCross-layer QoS-aware com-munication for ultra wide band wireless multimedia sensornetworksrdquo IEEE Journal on Selected Areas in Communica-tions vol 28 no 5 pp 653ndash663 2010

[67] Q Wang and M A Abu-Rgheff ldquoCross-layer signalling fornext-generation wireless systemsrdquo in Proceedings of the 2003IEEE Wireless Communications and Networking 2003WCNC 2003 vol 2 pp 1084ndash1089 IEEE New Orleans LAUSA March 2003

[68] P Liu Z Tao S Narayanan T Korakis and S PanwarldquoCoopMAC a cooperative MAC for wireless LANsrdquo IEEE

32 Journal of Computer Networks and Communications

Journal on Selected Areas in Communications vol 25 no 2pp 340ndash354 2007

[69] H Shan H T Cheng and W Zhuang ldquoCross-layer co-operative MAC protocol in distributed wireless networksrdquoIEEE Transactions onWireless Communications vol 10 no 8pp 2603ndash2615 2011

[70] F Liu T Korakis Z Tao and S Panwar ldquoAMAC-PHY cross-layer protocol for wireless Ad-Hoc networksrdquo in Proceedingsof the 2008 IEEE Wireless Communications and NetworkingConference pp 1792ndash1797 IEEE Las Vegas NV USA March2008

[71] L Jacob and H R Shamna ldquoEfficient cooperative MAC androuting in wireless networksrdquo Transactions on Networks andCommunications vol 3 no 5 2015

[72] S Rao and K Shama ldquoCross layer protocols for multimediatransmission in wireless networksrdquo International Journal ofComputer Science amp Engineering Survey vol 3 no 3pp 15ndash28 2012

[73] L Han S Park S Kang and H Peter ldquoAn adaptive cross-layer FEC mechanism for video transmission over 80211WLANsrdquo in Proceedings of the 2009 International Conferenceon Internet pp 209ndash215 Las Vegas NV USA December2009

[74] N Li J Martınez and V H Dıaz ldquo-e balanced cross-layerdesign routing algorithm in wireless sensor networks usingfuzzy logicrdquo Sensors vol 15 no 8 pp 19541ndash19559 2015

[75] G Sun J Qi Z Zang and Q Xu ldquoA reliable multipathrouting algorithm with related congestion control scheme inwireless multimedia sensor networksrdquo in Proceedings of the2011 3rd International Conference on Computer Research andDevelopment vol 4 pp 229ndash233 IEEE Shanghai ChinaMarch 2011

[76] M Li Y Jing and C Li ldquoA robust and efficient cross-layeroptimal design in wireless sensor networksrdquoWireless PersonalCommunications vol 72 no 4 pp 1889ndash1902 2013

[77] J Yan M Zhou and Z Ding ldquoRecent advances in energy-efficient routing protocols for wireless sensor networks areviewrdquo IEEE Access vol 4 pp 5673ndash5686 2016

[78] A Ahmad S Ahmad M H Rehmani and N U Hassan ldquoAsurvey on radio resource allocation in cognitive radio sensornetworksrdquo IEEE Communications Surveys amp Tutorials vol 17no 2 pp 888ndash917 2015

Journal of Computer Networks and Communications 33

International Journal of

AerospaceEngineeringHindawiwwwhindawicom Volume 2018

RoboticsJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Active and Passive Electronic Components

VLSI Design

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Shock and Vibration

Hindawiwwwhindawicom Volume 2018

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawiwwwhindawicom

Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Control Scienceand Engineering

Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Journal ofEngineeringVolume 2018

SensorsJournal of

Hindawiwwwhindawicom Volume 2018

International Journal of

RotatingMachinery

Hindawiwwwhindawicom Volume 2018

Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Navigation and Observation

International Journal of

Hindawi

wwwhindawicom Volume 2018

Advances in

Multimedia

Submit your manuscripts atwwwhindawicom

Page 7: ReviewArticle - downloads.hindawi.com · ReviewArticle QoSStrategiesforWirelessMultimediaSensorNetworksinthe ContextofIoTattheMACLayer,ApplicationLayer,and Cross-LayerAlgorithms MuwongeSsajjabbiBernard

T-MAC design the major forms of energy consump-tion in T-MAC are through (i) idle listening (ii) col-lisions (iii) protocol overhead and (iv) overhearing Incontrast in the active period the node remains lis-tening and sending data If no activation event occursfor a certain time t then the active period ends -eactivation event might be data reception on a radiocommunication sensing on a radio and end oftransmission for the data packet of the node or ACKknowing that the neighbor has finished transmitting

Clustering and synchronization on waking up a nodebegins by listening to the channel If after a certain timethere is nothing heard it chooses a schedule fortransmitting Synchronization (SYNC) packetsappended with the time of the next frame startup Ifduring this time it hears another nodersquos SYNC packetit follows its schedule subsequently transmitting itsown SYNC Following both schedules enables it to havean activation incident at the beginning of both frames

RTS and t choice if a node does not get a CTS messageafter sending an RTS then either (i) there is a collisionpreventing the destination node from hearing the RTS(ii) RTS or CTS is overheard and destination node barredfrom responding or (iii) the destination node is in thesleep mode So if after time t there is no answer thesource node switches to the sleepmode It should only doso if none of its neighbors is still communicating A newinterval t can be triggered by a neighborrsquos RTS or CTSreception A node may be out of range to hear theinitiating RTS-e length t of the interval should be longenough to get the CTS packetrsquos beginning [25] and t isthus given by the comparison in the following equation

tgtC + R + T (1)

where C contention interval length RRTS packet lengthand T turn-around time

Advantages

(i) It saves power through adaptive duty cycling

Disadvantages

(i) It is unscalable and more complex(ii) It offers less throughput and enhanced delay

214 CU-MAC -is protocol supports IoT standards re-quiring request-response transmissions It employs a multi-channel mechanism for constant bidirectional transmissionof packets at low duty cycle to resolve hidden-terminal issuesAccording to Danmanee et al [24] CU-MAC is an SI-MACprotocol that is aimed at improving bandwidth of the channelby initiating connections and transmitting data throughdifferent channels using the multichannel approach

Protocol design the protocol has the following threeproperties that make it perform more efficiently withWSNs in the IoT

(i) It enables multichannel transmission using 9useable channels in IEEE 802154

(ii) It can enable linked data transmission using oneconnection which reduces additional overheadand delay

(iii) It can support continuous transmission of packets-is is enabled depending on how the bufferoverflow protection policy is set in the buffer stack-is stack has 2 parts (a) bidirectional buffer slotsfor enabling sensor nodes to support linked datatransmission and (b) normal buffer slots fornormal data queuing Buffer stack policies are setas follows

(a) A packet is only dropped in the buffer when itexpires

(b) Unless the packet is from a target node to which areceiver is sending the data the receiver deniespacket reception when buffer slots fill up

(c) -e bidirectional buffer slot is used for bi-directional buffer transfer

CU-MAC can reduce power wastage by choosing toemploy only a single channel for transmitting both controlpackets and data in case there is only one sender in the sameregion of coverage It can further enable multiple trans-missions through another usable channel if there are othersending nodes in this area [24] -e protocol has two phases(i) advertising phase in which a source node announces toneighbors before transmission and (ii) data transfer phasewhich commences when the source node receives a Ready-to-Receive Information packet from the destination node

Advantages

(i) It has a higher packet delivery ratio compared toother MAC protocols

(ii) It performs well in high traffic situations since itsduty-cycle is lower

(iii) It reduces power wastage by employing only asingle channel for transmitting both controlpackets and data

(iv) It can enable multiple transmissions through an-other usable channel if there are other sendingnodes

215 R-MAC -is is a reservation-based MAC protocoldesigned majorly with an aim to be fair and efficient in termsof energy use To eliminate collision of data packets R-MAC[26] schedules control and data packet transmissions at thesource and destination nodes rather than using RTSCTSmessage exchanges Channel utilization in R-MAC is en-hanced by an ARQ method called burst-based acknowl-edgment which together with other scheduling algorithmsaddresses the exposed terminal problem besides enhancingthe network throughput

R-MAC design all nodes in this protocol intermittentlyfunction in listen and sleep states so as to avoid wastingenergy in overhearing and idle modes -e time takento listensleep is the same for all nodes each of which

Journal of Computer Networks and Communications 7

chooses a schedule for itself rendering central sched-uling and synchronization irrelevant So in the absenceof traffic in the locality the node simply does periodiclistening and sleeping and to eliminate collisionstransmissions are distributively synchronized using areservation-based model in case there is a sender totransmit data R-MAC differs from S-MAC in that ittransmits data to a number of nodes using the sleepmode [24] -ere are 3 stages in this protocol (i) la-tency detection (ii) period announcement and (iii)periodic operation of which (i) and (ii) are for nodesynchronization and (iii) is for listensleep procedures

(i) Latency detection phase here all nodes broadcast aNeighbor Discovery control packet (ND) at an ar-bitrarily chosen time since they are powered onWhen the ND is received from the neighbor nodethe receiver notes its arrival time and chooses arandom time to send an acknowledgment (ACK-ND)mdashthe same as the receivedND in sizemdashinwhichthe duration from ND arrival to ACK-ND trans-mission I2 is specified and the time from ND packettransmission to ACK-ND arrival I1 is computed

(a) Propagation latency between the 2 nodesL (I1 minus I2)2

(b) -us propagation latency can be defined as ldquotheinterval from the time the first node sends the firstbit of a packet to the time the second node receivesthe last bit of the packetrdquo [24]

(ii) Period announcement phase here nodes choosetheir own random listensleep time and start timeand broadcast these and when packets are re-ceived the node turns the received schedule intoits own All nodes will record their neighborsrsquoschedules in relation to theirs -ese two phasesmake a number of rounds to ensure all nodes areinformed about their neighboring nodes

(iii) Periodic operation phase here there is datatransmission [26] and nodes intermittently go intothe sleep mode and wake up One period is a listensleep cycle and nodes possess similar periods Twoparts make a period T0 listen window TL and sleepwindow TS as shown in the following equation

T0 TL + TS (2)

In this protocol the nodes make use of control packetsfor communication-ese include REV (reservation packet)ACKREV (acknowledgment packet for REV) and DATAACK-DATA (acknowledgment packet for data) messageexchanges all of which are similar in size

Advantages

(i) It is fair and efficient in terms of energy use(ii) It eliminates collision of data packets

216 AR-MAC (Adaptive-Reliable Media Access Control)-is is a TDMA-based MAC protocol for wireless body areanetworks (WBANs) designed to reduce energy consumption

as proposed by Rahim et al [27] -e protocol allocates aguaranteed time slot (GTS) to all nodes for transmissiondepending on their needs It utilizes periodic sleep andwakeup to minimize overhearing and idle listening based onthe needs of the node It has a central node (CN) having bigbatteries with higher computational power and single ordouble transceivers For double transceivers the sum of thetime frame TFrame is assigned for node transmission But inAR-MAC a single transceiver CN is considered with TFramehaving 3 subdivisions contention-free period (CFP) forsensor connection contention access period (CAP) foremergencies and time TMS to communicate with themonitoring station (MS)

Channel selection the CN first scans for redundantchannels from which it chooses one for transmissionand broadcasts its address and other information to thenodes -e destination nodes search for channels byscanning the radio frequency (RF) channels If free itswitches to the next and if busy it will wait for time TCPlistening to packets and again switch to the subsequentchannel if it does not get the channel packets When thenode finally succeeds in getting a channel packettransmission commences and an ACK packet is sent tothe CNTime slot assignment a Time Slot Request (TSR) packetis sent to the CN by the sensor node after it has selectedthe RF channel -e TSR packet has data rate and timeslot info for the node Static-size time slots with staticguard-band time TGB are proposed in [28] -isprotocol employs the adaptive algorithm of time slot(TS) and GB time and the CN allocates TS whilesending Time Slot Request Reply (TSRR) based on thenode traffic pattern -e size of the time slots variesdepending on the node requirements Based on thetransmissionmodel the data packet transmission ACKpacket reception and some amount of delay can betolerated by allocated TS To prevent interference be-tween adjacent time slots resulting from the node andCN clock drift TGB is inserted between them and itsvalue is given as follows

TGB1

FxTS1100

TGBn

FxTSn

100

TGBnn+1

F

100X12

TSn + TSn+11113858 1113859

(3)

where FGB factor depending on the mean driftvalue Nodes go to sleep after slots have been suc-cessfully assigned and wake up to transmit in assignedslots thus saving energy wasted in idle listening sinceeven allocated slots are collision freeSynchronization for periodic synchronization TDMAneeds additional energy [29] It needs a lot of energy tosynchronize nodes after several cycles To reduce en-ergy wastage and control collisions AR-MAC employs

8 Journal of Computer Networks and Communications

a new synchronization scheme When a data packetarrives the current and expected packet arrival timeswith allowable delay (D) are compared by the CN and adrift value (DV) is computed based on the differenceΔT -e DV is sent to the node in the ACK packet forchanging the slot in subsequent transmissions thoughthe value is based on allowable delay and F In caseΔTgtD the CN will send the DV in the SYNC-ACKpacket for subsequent SYNC to nodes or else a mereACK packet is sent by the CN for the received packetFor any forthcoming data transmissions the node willchange its wake-up time plan based on the DV Withsuch a sync approach a sleeping node will not lose syncfor all cycles N Acceptable delay (D) is given as

D min TS1 TSn( 1113857XF

100 (4)

where F is the guard-band factor

ΔT expected arrival time minus current arrival time

DV 0 if |ΔT|ltD

ΔT if |ΔT|gtD1113896

(5)

Frame format the AR-MAC protocol employs twoforms of packets namely (i) data packets in which anode transmits in its assigned time slot and employsCAP for emergency data and (ii) control packets thatinclude channel packet Time Slot Request (TSR)packet Time Slot Request Reply (TSRR) packet Syn-chronization-Acknowledgment (SYNC-ACK) packetData Request (DR) packet and Acknowledgment(ACK) packet [27]Energy consumption if N is the number of cyclesenergy consumption for these is measured as follows

ETotal 1113944N

k1ESleepk

+ 1113944N

k1EActivek

ESleep TSleep x ISleep x V

TSleep Tframe minus Tactive

(6)

where ISleep current from the voltage source V whenthe node is asleep If Esw is the switching energy Etransthe transmission energy Erec the receiving energy andETout the time-out energy then

EActive 2Esw + Etrans + Erec + ETout (7)

Advantages

(i) It reduces energy consumption(ii) It minimizes overhearing and idle listening based

on the needs of the node via periodic sleep andwakeup

217 ER-MAC It is a combination of TDMA and CSMAmethods making it flexible and adaptable to topological andtraffic changes and hence applicable in rapid response

situations inWSNs Collision-free slots can be planned usingTDMA [30] ER-MAC is designed in such a way that thefollowing are met

(a) It has a higher ratio of packet delivery and low delayunder lower energy consumption levels out-performing Z-MAC

(b) It keeps two priority queues in order to distinguishhigh from least priority packets

(c) It permits contention in TDMA slots so as to handleenormous traffic volumes

(d) It has a harmonized slot shape and the nodes areable to locally change their plans enabling themconnect to or exit the network with easeProtocol design it discovers the topology in such away that it uses the simple flooding mechanism tobuild the tree as initiated by the base station (BS) -etopology discovery aims at neighbor discovery andchange tracking in addition to setting up a routingtree A TOPOLOGY_DISCOVERY message com-prising hop_count new_parent_id and old_-parent_id is generated by the BS A node broadcasts itto discover if it has any potential children and aresponse to the parent In case of the desire to changethe parent the former parent is notified For all nodesthe hop count to the BS parent ID children and one-hop neighbor are registered with the BS at this stageTDMA slot assignment when a BS sends a SYNCmessage it switches to the TDMAmode which whenreceived by the child uses the broadcast slot for itschildrenrsquos synchronization Neighboring nodescannot share a slot since slots are assigned andschedules are exchanged accordingly Slot assign-ment begins with a leaf node in a bottom-up modesuch that the flow of messages to the BS may besmooth Before getting the child-node schedules anonleaf node will not transmit its data via a unicastslot descendantrsquos data via many unicast slots andchild synchronization via a broadcast slot Childnodes send SCHEDULE_NOTIFICATION mes-sages to the BS to end this phase

Local time synchronization is done via parent-childbroadcast sync since they both need to operate the same clockso that when one is transmitting the other is ready to receivePacket prioritization happens in such a way that a highpriority queue is first emptied before a lower priority queue isallowed to transmit In case a queue is filled up the shortestslack packet is dropped and then fairness over source isconsidered as queue modification is done to enable the BSbalance info from different nodes Energy conservation is bysender turning off the radio in absence of any info to transmitand the receiver will switch to the sleep mode in case nopackets are received after a certain timeout [30]

Advantages

(i) It is flexible and adaptable to topological and trafficchanges

Journal of Computer Networks and Communications 9

(ii) It is applicable in rapid response situations inWSNs

(iii) It has a higher ratio of packet delivery and lowdelay under lower energy consumption levels

218 RI-MAC (Receiver-Initiated MAC) -is is an asyn-chronous duty-cycle protocol that tries to reduce timeoccupied by the communicating nodes on the medium toagree on time for data exchange-e sending node waits forthe receiving node to signify the beginning of the com-munication session through a short beacon frame that itsends -e channel is only occupied by the beacon and datatransmission reducing channel occupancy and thus en-abling data exchange for other nodes too -is implies animprovement in channel utilization that subsequently leadsto increment in throughput packet delivery ratio andpower efficiency over a wide-ranging traffic load [31] Beinga receiver-initiated model it ensures substantial reductionin overhearing and lowers the probability of colliding andthe cost of recovery compared to B-MAC [23] or X-MAC[32] RI-MAC is further capable of using a beacon withACK and Ready-to-Receive packets though it has aproblem if a sending node with Ready packets needs to havethe radio on till a ready-to-receive node is awake [13] whichthen sends it a beacon and transmission begins immediatelyand later is acknowledged by another beacon Nodes in-termittently wake up to listen to any possible frames meantfor them depending on their schedules A node announcesits readiness to receive a frame through a broadcast beaconon waking up when it switches on the radio When a dataframe has been received the receiver will spend some moretime known as ldquodwell timerdquo in the active state so as toreceive queued packets which is the time in RI-MAC thatdepends on the number of contending senders If there areseveral contending senders that might not have beenpredicted there is a challenge in trying to reduce the re-ceiverrsquos active time to optimize power efficiency and reducethe cost of collision detection and lost data recovery even inthe case of hidden senders For this protocol this can beaddressed by a beacon frame from the receiving node forcoordinating data frame transmissions of contendingsending nodes-e protocol also beats B-MAC and X-MACin minimizing the cost of collision detection and datarecovery through coordination of the receiversrsquo dataframes -is is because the receiving node is aware of themaximum delay prior to the arrival of the frame-e abilityto broadcast is supported by RI-MAC when it sends data asunicast to each senderrsquos neighboring node or continuouslytransmits the frame in a back-to-back mode for as long asthe sleep interval [31]

Advantages

(i) It improves channel utilization leading to in-crement in throughput packet delivery ratio andpower efficiency over a wide-ranging traffic load

(ii) It ensures substantial reduction in overhearing andlowers the probability of colliding and the cost ofrecovery

219 SRI-MAC (Synchronous Receiver-Initiated MAC)-is is a synchronous duty-cycle protocol that implementsthe method of receiver-initiated data transmission with anaim of enhancing network lifetime through energy wastageavoidance via collision overhearing and idle listening To dothis and also minimize duty cycle it uses a sequence ofadaptive beacons and RTSCTS packets In this protocolsending nodes do not go to sleep till they have got CTSpackets from receiving nodes to begin transmitting exceptfor nonsending nodes that return into the sleep mode rightaway as if the channel was idle [33] Since sending nodeshave separate schedules it reduces the contention proba-bility leading to more energy conservation A time frame isdivided into the information period the allocation periodand the communication period which are discussed asfollows

Information period at frame initiation a beacon is sentby the receiving node declaring its return from sleepand readiness to accept data packets in case the channelis not busy -e beacon is a small packet alerting otherlistening nodes of the transmitterrsquos readiness to get arequest for channel It consists of the receiverrsquos ID andthe duration allocation period (DAP) that relies onnumber of the receiving nodersquos neighbors [13]Allocation period here the receiving node stays awake forsome little time after broadcasting its beacon to confirm ifthere is any node having anRTS packet to transmitWhenthe beacon has been received an RTS packet (with 3fieldsmdashsender ID receiver ID and data size) is sent by thesender prompting the receiver to send a CTS for SYNC asthere is neighbor information exchangeCommunication period SRI-MAC is capable of or-ganizing and scheduling transmissions by means ofsensor node IDs -e receiver assigns order andtransmitting time for every sender such that they switchoff their radios till their time to transmit is dueHowever the receiving node remains in the wake-upmode to get all sending nodersquos data SRI-MAC canenable sensors have lengthy sleeping times duringtransmission such that energy can be saved as long as itis neither sending nor receiving data -e protocolrsquoscommunication period is premised on the TDMAtechnique which conserves energy since the radiorsquosduty cycle is lowered and collision is eliminated [33]

Advantages

(i) Since it employs the TDMA principle it eliminatescollisions while maintaining energy conservationtoo

(ii) Noncommunicating nodes switch off their radioand get into the sleep mode which eliminatesoverhearing

(iii) Since receiving nodes broadcast SYNC signals toall nodes in the form of CTS packets overemittingis avoided

(iv) It decreases duty cycling as sleep time increasesand lowers idle listening

10 Journal of Computer Networks and Communications

2110 H-MAC -is protocol is built on the power savingmechanism (PSM) of IEEE 80211 and slotted Aloha pro-tocols to enhance performance using multiple slots Timehere comprises large frames each having an active and asleeping part [34] If any node has got data to send it will askfor slots from the receiver node during active time and latertransmit the packets in the prenegotiated slots Nodes gointo the sleep mode during sleep-time slots if they do nothave data to transmit

Latency in IEEE 80211 if N is the number of hops nthe value for the current hop tcsn the back-off time andttx the transmission delay the total latency is calculatedusing the following equation

D(N) 1113944N

n1tcsn + ttx1113872 1113873 (8)

and the average latency is given by the followingequation

E[D(N)] N tcsn + ttx1113872 1113873 (9)

-roughput in H-MAC is given by the followingequation

Th npnm

tact + Cts (10)

where np is the number of transmitted packets nm is themaximum number of nodes connected in a time frameC is the number of slots with the same length tact is thelistening time and ts is the sleep delayAdvantages

(i) It enhances performance using multiple slots

2111 Z-MAC (ZebraMAC) -is rides on the strengths andoffsets weaknesses of TDMA and CSMA It yields very goodchannel usage and low delay under low contention and lowerscollision among two-hop neighbors at a reduced cost Itsperformance is strong via error synchronization failure inassigning slots and channel conditions that change with timeIn the set-up phase the protocol operates through certainsteps including neighbor discovery slot assignment localframe exchange and global time synchronization whichunless the topology of the network changes significantly donot run again When the data are being transmitted there isimprovement in network throughput and energy efficiencywhich covers up for the initial upfront operational costs [35]

Disadvantages

(i) Initial slot assignment is quite a cumbersome taskhere since it is done at the beginning yet this maynot be at the same time in reality It further as-sumes static links until setup happens again whichtoo is impractical

(ii) It also suffers from the hidden-terminal problemsince slot owners have a small contention windowand priority for slots allocated to them due to theECN messages

(iii) Synchronizing time also poses a challenge in thisprotocol due to a large clock drift between nodes

(iv) Source and destination nodes too are not wellcoordinated in Z-MAC [36]

2112 Q-MAC (QoS-Aware Media Access Control) -isprotocol offers high QoS with improved energy efficiencyand accesses the wireless channel using MACAW as theunderlying protocol [37] Traffic from different nodes isprioritized for QoS satisfaction depending on how criticalthe data are It mainly has two scheduling algorithms inWSNs (i) intranode and (ii) internode scheduling -eintranode scheduling scheme adopts a first-in first-out-(FIFO-) based queuing algorithm based on application andMAC layer abstraction for data packet classification In-ternode scheduling reduces idle listening and collision on thechannel for improved energy consumption

Advantages

(i) It gives a high QoS with improved energy efficiency

2113 WiseMAC (Wireless Sensor MAC) WiseMAC isbased on the Aloha protocol and employs preamble sam-pling in achieving low-power transmissions in infrastructuresensor networks by occasionally sampling the channel tocheck for activity [3238] -e protocol employs the samemethod as B-MAC except that the sender shortens theextended preamblersquos length by scheduling its transmissionsaccordingly after learning the receiverrsquos awake schedules-is is done by the receiver node putting its subsequentawake time on the ACK packet such that when the senderneeds to send to it again it starts the preamble shortly beforeit wakes up so as to avoid energy wastage in transmitting thepreamble [32] -e data frame is repeated instead of theextended preamble for less traffic transmissions that have alengthier preamble than the data frame in WiseMAC Afterprocessing the frame a node that is not the target recipientwill go back into the sleep mode while the recipient staysawake till the session ends thereby sending an ACK packetFor an idle channel the power consumption is rather lowwith this method Nevertheless there is a problem of longwake-up preamble limiting throughput and causing anenormous power consumption overhead in reception [38]All nodes overhearing a transmission bear the overheadtogether with the recipient node -e protocol can addressissues concerning low-power transmission but has nomechanism of nodes adapting to varying traffic patterns[32] To compensate for drift between clock at the accesspoint and on the sensor node there is a need to compute thewake-up preamble duration -e drift varies directly withtime since previous resynchronization and is given by thefollowing equation

TP min 4θl TW( 1113857 (11)

where θ is the time-base quartz frequency tolerance and l isthe interval between two transmissions Increase in trafficvolume naturally reduces overhearing

Journal of Computer Networks and Communications 11

Radiomodel for protocols with lower power transitiondelays between transceiver states and their powerconsumption are modeled Some of the states includeDOZE where the transceiver neither transmits norreceives but is prepared to power on to the receivetransmit state very fast RXwhere the transceiver listensto the channel receives data or demodulates data fromthe noisyidle channel and TX which is the trans-mission state by the transceiver Let TS be the set-uptime needed to turn the transceiver from the DOZE tothe RX or TX state TT the turn-around time needed toswitch between RX and TX states and B the trans-ceiverrsquos bit rate PZ PR and PT are the values of powerconsumed in the DOZE RX and TX statesrespectively

1113954PR PR minus PZ

1113954PT PT minus PZ(12)

where 1113954PR and 1113954PT are the changes in power as a result ofbeing in RX and TX states respectivelyTraffic model let N be the number of sensor nodesconnected to an access point (AP) and λ be the global rateat which the downlink Poisson traffic arrives at the APfrom the fixed network Traffic toward each sensor node isthe same and is given by λN Average packet interarrivaltime Lwith which data packets arrive at a node is given byL Nλ If TC and TD are the control packets and datapackets respectively assuming a small traffic volume theglobal interarrival 1λ far exceeds the summation ofdurations for the data packet and control packet and theturn-around time Mathematically it is given as

1λ≫TD + TC + TT (13)

-e average power consumed by WiseMAC andtransmission delay incurred are given by equations (14)and (15) [38]

PW PZ +1113954PR TS +(1B)( 1113857

TW+

1113954PR X + TD + TT( 1113857

L

+ 1113954PR(N minus 1)Y

L

(14)

where

X 2θL 1 minus eminus TD4θL( )1113874 1113875

Y T2D + 12TDθL

2TW1 minus e

minus TW4θL( )1113874 1113875

DW TD +TW

21 minus e

minus TW4θL( )1113874 1113875

+ 2θL 2 minus eminus TD4θL( ) minus e

minus TW4θL( )1113874 1113875

(15)

where PW and DW are the power consumed byWiseMAC and the transmission delay for WiseMACrespectivelyAdvantages

(i) It achieves low-power transmissions in in-frastructure sensor networks through preamblesampling while checking for activity

(ii) It reduces overhearing due to increased trafficvolume

Disadvantages

(i) It has a problem of long wake-up preamble limitingthroughput and causing enormous power con-sumption overhead

2114 X-MAC It is a simple minimal transmission powerMAC protocol that uses a short preamble with decouplingfor the sending and receiving node sleep plans inWSNs-isleads to a significant reduction in energy usage at sendingand receiving nodes and decrease in per-hop latency andgives flexibility in adapting to bursty and periodic sensordata sources [32]

Protocol design X-MACrsquos design is aimed at increasingenergy efficiency reducing data latency and increasingthroughput -is enables application in different kindsof packetized and bit stream digitized radios as well assimplified low-overhead distributed application -edesign is as such in order to address the low-powerlistening problems including overhearing excessivepreamble and incompatibility with packetizing radiosas in [32]Asynchronous duty cycling a node with information totransmit will send the packets after sending a preamblewhereas the remaining nodes will preserve their sleepschedules that are not synchronized -e receiver willsample the link on waking up and stays awake if it detectsa preamble and if it is the target it will return to sleep if itis not the target after getting the whole preambleEmbedding the target ID in the preamble to avoidoverhearing the lengthy preamble is divided into manyshortened packets embedded with the receiver nodersquosID so as to eliminate overhearing One lengthy pre-amble is constituted by the shortened preamble packetstream Any node on receiving the short preamble afterwaking up will examine the target ID on the packet Itgoes back into the sleepmode right away resuming dutycycling if it is not the targeted recipient or stay awakefor successive packets if it is the one -is approachchecks energy wastage due to network density It canalso be used on several types of radiosReducing excessive preamble using strobing it ispossible to enable low-power transmission and alsosave energy by reducing time for preamble trans-mission through the use of enhanced preambles inaddition to preamble sampling X-MAC does not sendan endless preamble packet stream but puts slight

12 Journal of Computer Networks and Communications

pauses in between packets to allow the sending node topause and listen to the channel and allow the receivingnode to resend an early ACK packet during that pause-is makes the sending node begin sending datapackets and the receiving node cut excessive preamblesand thus decrease per-hop latency and energy wastageon unnecessary waiting and transmissionPacketizing radios low-power listening is limited incapability for supporting packetizing radios butX-MACrsquos short strobed preambles can support all typesof digital radiosAdaptation to traffic load while a number of WSNapplications are reliable in traffic production they stillmust adapt to changing traffic Nodes are going to haveseparate sleep schedules depending on traffic loadConsidering systems that have time-fluctuating trafficall predetermined static schedules are suboptimalOptimality Buettner et al [32] model the expectationof energy as follows

Es (preamble energy + energy perACK listen)

lowast(expected preamble minus listen iterations)

+(energy to send parket)

Es PTxSp + PRxSal1113872 1113873 Rl + Rs( 1113857

Rl minus Sp+ SdPTx

(16)

where Es expected energy to send a packet PTx andPRx are the transmission and receiver power re-spectively Sp Sal and Sd are the duration of thesenderrsquos preamble ACK listen and data transmissionperiods respectively and Rl and Rs are the receiverrsquoslisten and sleep periods respectively Expected energyto receive a packet is given as

Er (listen cycle energy + sleep cycle energy)

lowast(expected iterations for preamble arrival)

+(energy to sendACK) +(energy to receive packet)

Er PsRs + PRxRl( 1113857

1 minus 1 minus Pd(t)( 1113857 Rl + Rs( 1113857+ RaPTx + RdPRx

(17)

And the one-hop expected latency is given as

Lat (preamble duration + ACK listen)

lowast(expected number of iterations)

+ duration to send packet

Lat Sp + Sal1113872 1113873 Rl + Rs( 1113857

Rl minus Sp

(18)

Advantages

(i) It increases energy efficiency(ii) It reduces data latency(iii) It increases throughput

2115 A-MAC (Advanced MAC) A-MAC is a TDMA-based protocol whose design is aimed at enhancing thelifetime of the network for low rate and reliability of datatransmission [39] To assign time slots it does not depend ona central controller like most protocols but it gathers in-formation from its neighbors so as to allocate the slotsthrough a distributed algorithm Energy is controlled byscheduling power when there is not a single event Its designcomprises many frames with each having many time slotswhich at the beginning have got a beacon transmitted tosynchronize and exchange information with neighborsNodes to participate in the subsequent session are selectedby the controlled node so they do not go into the sleep modewith others [13] -is protocol operates under four time slotdivisions namely initial wait discover and active states Anode in the initial state begins by listening to the medium forthe beacon packet from other nodes for network synchro-nization which when received makes it change the timerthrough subtraction of beacon transmission time from re-ception time To avoid synchronization problems due todrift and allow for continuity the node has to get thestrongest beacon signal It then enters the wait state afterchoosing a number of waiting frames so as to reduce theprobability that nodes get into the discover state simulta-neously In case the synchronization beacon is lost while stillin the wait state it will return to the initial state for anotherone On expiry of the waiting counter it enters the discoverstate and collects information from neighbors by listening totheir beacon messages On successful selection of a time slotit enters the active state where it endlessly transmits beaconsat the initiation of the slot A node goes to sleep when abeacon has been transmitted and there are no more data fortransmission or in case the neighborrsquos beacon received in-dicates no data packets are coming in [39]

Advantages

(i) It can enable multicasting(ii) It enhances the lifetime of the network(iii) It offers reliable data transmission

2116 L-MAC (Lightweight MAC) Based on TDMA thisprotocol is aimed at minimizing the number of transceiverswitches adapting the nodal sleep interval to data trafficamount and making implementation less complex [40] Apacket is transmitted comprising a control and a data partControl packets have a fixed size and the time slot con-trollerrsquos ID -ey show the distance from the transmittingnode for simplified routing to the network gateway as well asaddressing target node and reporting data unit length andinternode synchronization Adjacent nodes are more in-terested in getting control packets from nearby nodes -enodersquos transceivers are switched off until the subsequenttime slot unless the node is also addressed or the message isomnicast If so it listens to the data unit which may not fillthe remaining time slot in totality-e transceivers on eitherside are turned off when the transfer of packets is successfulEnergy wastage through idle listening in uncontrolled timeslots is avoided via a short time-out interval Network nodes

Journal of Computer Networks and Communications 13

are able to communicate without colliding-is saves energyand increases the lifetime of the network A node in L-MACcan only transmit one message per frame [13] -e networkis set up in such a way that all nodes are not synchronized onbeing switched on-e gateway takes control of a timeslot soas to have them synchronized and its immediate neighborgets the control messages thereby synchronizing their timerswith the gatewayrsquos Neighboring nodes learn of multiplegatewaysrsquo time slots in their vicinity just after one frame-ejust synchronized nodes randomly choose time slots forcontrolling as long as they are not occupied -e nodes thuspreserve a table for the neighboring nodes which makes atimeslot reusable after a minimum of 3 hops and preventsmessage collision All nodes keep their slots before thebattery is down or learn of an impending collision All nodescan track their hop distances to a selected gateway node andthe information is communicated in control messages [40]

Advantages

(i) It minimizes the number of transceiver switches(ii) It adapts the nodal sleep interval to data traffic

amount(ii) It makes implementation less complex(iii) It offers improved network lifetime(iv) Energy wastage through idle listening in un-

controlled time slots is avoided via a short time-out interval

Disadvantages

(i) L-MAC is not very good for mobile SNs since slotsare calculated only once

(ii) It suffers idle listening because nodes keep listeningto slotsrsquo control parts so as to get data and enableother nodes to join the network all the time [13]

2117 C-MAC -is was designed for WSNs with 3 seg-ments namely (i) aggressive RTS having double channelcheck to assess the channel (ii) anycast to initialize the flowand (iii) convergent packet forwarding to stabilize the flowIt promises an improvement in energy efficiency lowerlatency and enhanced throughput while avoiding syn-chronization overhead though it operates only in low duty-cycling apps especially where there is less traffic But whenthere is more traffic the protocol employs anycast-basedpacket forwarding for waking the nodes up or finding aforwarder quickly and converges from route-suboptimal toroute-optimal unicast After this energy could be preservedby the nodes using a synchronized wake-up plan [13]

Advantages

(i) It improves energy efficiency(ii) It lowers latency(iii) It enhances throughput avoiding synchronization

overhead

Disadvantages

(i) It operates only in low duty-cycling applicationsespecially where there is less traffic

22 MAC Protocols for WMSNs

221 Diff-MAC -is is a CSMACA-based QoS-awarehybrid-priority-based MAC protocol for WMSNs It targetsenhancing the channel utilization while making use of ef-ficient service differentiation to do coordination for mediumaccess of each traffic class while giving fair quick datatransfer It offers QoS using the following features [14]

(i) It balances energy consumption and delay throughsensor node duty-cycling adaptation depending onthe dominant traffic class

(ii) It has a feature for fragmentation and messagepassing that breaks lengthy video frames intosmaller packets that are transmitted as a burst thusreducing retransmission costs if there are any MACfailures

(iii) -ere is fair data delivery amongst the nodes andtraffic classes respectively resulting from intranodeand intraqueue prioritization reducing unbearableperformance

(iv) -e number of collisions can be reduced and packetlatencies are kept within limits by adjusting the sizeof the congestion window in this protocol as per thetraffic needsAdvantages

(i) Diff-MAC is very suitable for WMSNs since itadapts quickly to changes in networkconditions

(ii) It is a fair protocol(iii) It is scalable

Disadvantages

(i) It is quite difficult to monitor the statistics ofthe network

(ii) Adapting dynamically to the network condi-tions is not an easy task

(iii) It exhibits some degree of latency in packetdelivery

-e implementation of Diff-MAC on Imote2 indicatesthat resource requirements can be met on the sensorhardware Results of extensive simulation runs showed thatDiff-MAC outperforms its competitors

222 Cluster-Based On-Demand Multichannel MAC Pro-tocol for WMSN (COM-MAC) -is is an energy-efficientprotocol in WMSNs with high throughput and reliable datatransmission It achieves energy efficiency by schedulingmultichannel media access to reduce channel contentionand remove collision idle listening and overhearing [41]

Design it has a three-phased operation with the (i)request phase (ii) scheduling phase and (iii) datatransmission phaseRequest phase here all nodes have two protocols eachfor sending requests to the cluster head (CH) ie thecontention-based and contention-free TDMAFDMA

14 Journal of Computer Networks and Communications

protocols -e contention-based protocol allocates achannel per node for sending the request message andfree channels are useable as control channels Forcontention-free protocols a control slot is allocatedwhen a network is deployed Here a slot per channel isallocated to a different node enabling requests to be sentwithout interfering with transmissions from othernodes -e contention-based protocol operates undertwo steps (i) Control channel assignment phase achannel is assigned per node for transmitting a request(REQ) message A channel is assigned to multiplenodes and nodes are evenly distributed to availablechannels for congestion control (ii) Request trans-mission phase nodes with data send a REQmessage viathe assigned control channel to notify the CH whichresponds with an ACK message -e contention-freeTDMAFDMA protocol too has two phases includingthe control slot assignment and request transmissionphases For unreliable channel conditions and lowtraffic load the contention-based protocol is fit for usebecause of less collisions and congestion whereas thecontention-free protocol is preferable in heavy trafficload and reliable channel applicationsScheduling phase from the above phase a schedule isgenerated by the CH for coordination and this enablesnodes to transmit and broadcast their data via controlchannels-e schedule for broadcasting only comprisesdata for nodes assigned to a channel -e schedule isrebroadcast for transmission reliability enhancementbut for energy efficiency the nodersquos transceiver isswitched off when all the schedule is completely de-livered till it is the time for transmitting again-e timeslot and data communication radio channel are allincluded in the schedule generated for each nodeData transmission phase on getting the schedule thesensor node sends its data as per the time slot and thechannel assigned -e time slot is divided into datatransmission andACK sections ACK supports link layererror control Using implicit selective repeat ARQ theCH sends an ACK message for all received packets Ifthere is no ACK the packet is assumed lost and hence aretransmission during the subsequent interval is done-is enhances reliability of transmissions though excessdelay is possible here which is also intolerable for real-time critical apps like video or audio -is gives rise to ahybrid MAC protocol to exploit the unused spectrumduring data transmission sessions [41]Advantages

(i) It achieves energy efficiency by scheduling mul-tichannel media access to reduce channel con-tention and remove collision idle listening andoverhearing

(ii) It maximizes network throughput by dynamicallyallocating time slots and channels for nodes byexecution of a traffic-adaptive QoS-aware sched-uling algorithm depending on needs for QoS andtraffic data information

(iii) It increases transmission reliability by in-corporating spectrum-aware ARQ to avoid spec-trum wastage

223 EQoSA -is protocol is developed for video andimage transfer across WMSNs in a hybrid mode withprovision of QoS Concerning the active sensor nodes andtheir traffic loads it employs dynamic session sizes andchanges bitmap-assisted (BMA) MACrsquos fixed session sizeNodes declare any data available for transmission duringcontention time -is forces the cluster head to make aschedule of how slots have been assigned It communicatesthis schedule to the nodes giving each node a given numberof slots per session and so enabling the busty traffic to beaccommodated [14] -e protocol does not perform wellwith DiffServe mechanisms but does well with schemessupporting busty traffic loads and strong cluster heads arenecessary in WMSNs that can assign and broadcast slots

224 QoS-Supported Energy-Efficient MAC (QE-MAC)Based on the IEEE 80211e standard this protocol ensuresfairness and has low latency and jitter and better efficiencyon energy usage than other QoS-aware MAC protocols forWMSNs It operates under a two-phased routine

-e first phase introduces the innovating prioritymechanism in IEEE 80211e QoS and several data types areassigned priorities Its coordination function is a combi-nation of distributed contention-based channel allocationand centralized polling-based channel access mechanisms

-e second phase preserves the nodesrsquo energy usingdynamic duty cycling -ere is an exchange of RTSCTSpackets to supplement CSMACA by muting nodes near thesenderreceiver during packet duration [42]

Advantages

(i) It ensures fairness(ii) It has low latency and jitter and better efficiency on

energy usage than other QoS-awareMAC protocolsfor WMSNs

225 Black-Burst (BB) Contention A contention-basedMAC scheme for enhancing instantaneous QoS access tocarrier-sense WSNs is proposed in [42] In this schemenodes contend for the channel until it is free Real-timepackets are given higher priority over non-real-time packetsand if a node is transmitting real-time packets it is prior-itized over other nodes [15] Voice and video are some of thereal-time apps in consideration here which periodicallyaccess the common radio channel when transmitting -eseneed constrained end-to-end delay of packets at the MAClayer -is protocol majorly does the following [42]

(i) Prioritization of real-time traffic(ii) Implementation of the round-robin mechanism in

real-time nodes(iii) Bounded access delays among real-time packets

Journal of Computer Networks and Communications 15

(iv) Implementation of real-time communication withvarying needs of bandwidth good for multimediatraffic

Amechanism for dynamically assigning packet prioritiesbased on their deadlines and the traversed hops is proposedin [43] -e ReAllocative Priority (ReAP) mechanism ismeant to provide QoS at the MAC layer to video traffic -isscheme introduces adaptive TXOP by modifying the TXOPdynamically based on the queue length [15] Other protocolsin this category include dynamic duty cycle and adaptivecontention window-based MAC protocol frame sharing(FRASH) MAC protocol and real-time independentchannels (RICH) MAC protocol

Advantages

(i) It can implement real-time communication withvarying needs of bandwidth

23 Challenges and Opportunities(a) One of the challenges in satellite transmission is

cross-talk leading to interference in different fre-quency bands It is necessary to increase the rise andfall times as well as protect sensitive nodes as wecircumvent floating ones if we are to regulate cross-talk as in satellite communication therefore Wefurther need to investigate the possibility of usingdifferential signaling -is constitutes an opportu-nity for further research

(b) Another challenge is that the hidden and exposedterminal problems are still noticeable despite the useof RTSCTS solutions in MAC since collisions arestill possible from transmissions of separate nodesAt the MAC layer ACKs might block efficientchannel usage by prohibiting exposed terminalsfrom reusing the channel and may be eliminatedleaving no proof of sender data packet receptionUsing a directional antenna is one of the proposedtechniques of high-speed WSNs like IEEE 80211-is constitutes an opportunity for further study intothese issues for better solutions though

(c) Another challenge is that since nodes in contention-based MAC protocols are mobile there are manycollisions by the packets and schedule in-consistencies in schedule-based protocols in the two-hop neighborhood information when nodes areentering or leaving [44] Deng et al [45] propose amobility-based clustering (MBC) protocol for mo-bile nodes to enhance the packet delivery rate -isprotocol in comparison with CBR has 25 chancesof decreasing collision-induced packet losses and50 chances in comparison with low-energyadaptive clustering hierarchy (LEACH)

(d) -e process of analyzing and evaluating routingperformance for WMSNs is quite complex becauseof the bursty multimedia traffic Because of the wayencoding scheme frames are structured as well as thechanges in between the scenes that occur naturally

the compressed video frames display some substantialburstiness on several time scales -ere will also befrequent route recalculation that is as well energy in-tensive for nodes frequently capturing and transmittingenergy -e solution here could as well possibly be theuse of the MBC protocol Route recalculation tooshould be dynamic to cater for the mobile nodes whichmight be in and out of the range at different instants oftime Recently the use of smart dumpsters for effectivewaste management in smart cities has also been pro-posed as a possible solution [46]

(e) For WSNs their routing protocols do not look atmultimedia applications which require enormousbandwidth and processing energy and should betransmitted in real time with utmost fidelity Extrachallenges in routing are imposed in WMSNs indeployment of heterogeneous sensors since therecould be audio video and still pictures in addition toscalar data during transmission Heterogeneous datahandled also have different QoS requirements andwith many different business needs different ser-vices with varying requirements give rise to a bigchallenge for the routing design -ere is a need todesign protocols that can support WMSNs such asEQoSA [14] for this purpose

Table 2 summarizes the reviewed MAC protocols forWSNs and the ones designed for suitability in WMSNs

3 QoS Strategies at the Application Layer

We discuss many protocols used to solve different needs ofcommunication between machines

31 ConstrainedApplication Protocol (CoAP) It is a requestresponse protocol designed by the Internet Engineering TaskForce (IETF) for constrained devices such as those havinginsufficient RAM or CPU and WPANs with low-powerconstraints resulting in poor transfer of packets andenormous overhead -e protocol was designed for ma-chine-to-machine (M2M) apps and system automation toreduce bandwidth requirements increase packet transferdecrease the overhead and make work simpler leading tolightweight implementation HTTP commands like GETPOST PUT and DELETE are used in a client-server ar-chitecture Using the publishsubscribe architecture CoAPcan support multiple users and hence yield better perfor-mance via multicasting enabling asynchronous messageexchange -is protocol can support both multicasting andunicasting via UDP and hence is a reliable and simpleprotocol -is is done via the messaging sublayer (for re-liability based on UDP) and the requestresponse sublayer(for communication and interaction) [16]

It uses the following messages

(i) Confirmable message this guarantees reliabilityusing ACK messages sent by the receiver node-ese could be synchronous or asynchronous ifthere is a need for more computational time

16 Journal of Computer Networks and Communications

(ii) Nonconfirmable message here no ACK messagesare sent back

(iii) Acknowledgment (ACK) message this confirmsreceipt of a confirmable message

(iv) Reset message this confirms receipt of an un-processed message in case any critical parts nec-essary for interpreting the message are missing

(v) Piggybacked response the receiver directly re-sponds while getting the ACK message

(vi) Separate response the reply by the receiver is dis-crete from the ACK message

CoAP has no in-built security features It is secured bythe datagram transport layer security (DTLS) that runs ontop of UDP though it was not designed for the IoT -ismakes its suitability debatable since DTLS offers no supportfor multicasting It also needs more packets for handshakingthereby increasing the traffic volume and computationalresources and so shortening the lifespan of mobile devicessince they use batteries [17]

Advantages

(i) It can support multiple users and hence yieldbetter performance via multicasting

Table 2 Comparison of MAC layer protocols used in QoS provisioning

Protocol Features QoS parameter Comments Access Priorityassignment Energy eff Designed for

S-MAC[22] Collision avoidance Energy latency Requires periodic

synchronization CSMA Hybrid Medium WSNs

B-MAC[23] Collision avoidance Energy delay

throughput Flexible interface CSMA Dynamic High WSNs

T-MAC[25] Collision avoidance Energy Complex nonscalable CSMA Dynamic High WSNs

C-MAC[13]

Dynamic channelallocation Energy Multichannel MAC CSMA Dynamic Medium WSNs

CU-MAC[24]

Multichannelmechanism

-roughputlatency High packet delivery ratio TDMA Dynamic mdash WSNs

A-MAC[39]

Per flow serviceguarantees Delay m-ary tree model

dynamic prioritiesTDMACSMA Dynamic Medium WSNs

ER-MAC[30]

Emergencyresponse

Energy delaydelivery ratio

Adapts to traffic andtopology changes

TDMACSMA Hybrid Higher than Z-

MAC WSNs

RI-MAC[31] Collision detection Delay packet

delivery ratioHigh packet delivery ratio

and throughput CSMA Dynamic Medium WSNs

SRI-MAC[33] Collision avoidance Energy delay Good data transfer and

energy consumption TDMA Dynamic High WSNs

H-MAC[34] Slotted sleep time -roughput

latencyData transmission only in

sleep timeCSMAAloha Dynamic Medium WSNs

Z-MAC[36] Collision avoidance Energy delay

throughputVery good channel usage

and error syncCSMATDMA Dynamic Medium WSNs

Q-MAC[37]

Intrainternodescheduling Energy High QoS improved

energy efficiency CSMA Hybrid High WSNs

X-MAC[32]

Overhearingavoidance

Energy delaythroughput

Reduces excessivepreamble by strobing CSMA Adaptive

static Medium WSNs

L-MAC[40] Collision avoidance Network

lifetime energyReduces the number oftransceiver switches TDMA Static Medium WSNs

WiseMAC[38]

No downlinkchannel collision Power delay Significantly low-power

consumptionAlohaCSMA Static Higher than

ZigbeeInfrastructure

WSNsAR-MAC[27] Collision avoidance Energy delay Uses data packets and

control packets TDMA Static Higher thanIEEE 802154 WBANs

BB-MAC[42]

Real-time trafficprioritization Delay Employs round robin in

real-time nodesCSMATDMA Dynamic High WMSNs

COM-MAC [41] Collision avoidance Packet delay

throughputBalances reliability and

retransmissionTDMAFDMA Dynamic High WMSNs

EQoSA[14]

Busty traffic loadaccommodation Energy Provides QoS support for

video and imagesTDMACDMA mdash High WMSNs

QE-MAC[42] Collision avoidance Energy jitter

latencyRuns central polling-like

channel access CSMA Dynamic High WMSNs

R-MAC[26] Collision avoidance Energy delay Reservation based mdash Static High Underwater

SNs

Journal of Computer Networks and Communications 17

(ii) It enables asynchronous message exchange(iii) It is a reliable and simple protocol

32 Message Queue Telemetry Transport (MQTT)Developed by IBM and later improved by OASIS with anintention of decreasing bandwidth MQTT is a simple open-source publishsubscribe protocol analogous to the client-server version It is appropriate for lightweight IoT appli-cations and M2M transmissions in constrained situationslike limited power computation capability and memory orbandwidth -e network bandwidth and the computationalresource usage are decreasing since clients are not updatedmaking publishsubscribe protocols outperform requestresponse protocols via IoT requirements [16] -is protocolcan run over TCPIP and so has the following advantages

(i) Delivering of packets with reliable guarantee(ii) Enabling multicasting(iii) Remote device session establishment(iv) Reduction of transport overhead and protocol ex-

change to minimize network traffic(v) Provision of the notification mechanism when

anomalies occur [47]

MQTTrsquos design targets economical use of bandwidthand battery so it has a major application in FacebookMessenger MQTT ensures QoS and reliability of messagedelivery by providing the following three QoS levels

(1) Fire and forget a message is sent once depending onthe networkrsquos best effort and there is no need forACK (lowest QoS level)

(2) Delivered at least once a message is sent at least onceand an ACK message is needed to avoid duplicates

(3) Delivered exactly once by a four-way handshake itensures a message is sent exactly one time (highestQoS level)-e architecture it has three key components (i)publishersmdashlightweight nodes sending data to thebroker before returning to sleep (ii) broker whichorganizes sensory data in topics and sends them tointerested subscribers and (iii) subscribersmdashIoTapps interested in data sent by sensors through thebroker as shown in Figure 3

MQTT is a simple lightweight protocol that is betterthan CoAP via low and high packet loss probability interms of throughput and exhibits lower latency in lowsampling rates [48] It supports a number of imple-mentations for embedded devices and mobile applicationsand offers web-socket support too Nevertheless it hassome shortcomings

(i) It is only employed in simple data types(ii) It exhibits high latency in high sampling rates(iii) It is unsuitable for real-time apps in such cases(iv) It does not provide automatic discovery andor

sufficient security at the protocol level [49]

33 Secure Message Queue Telemetry Transport (SMQTT)-is is an extension of MQTT that is based on lightweightattribute-based encryption using elliptic curve cryptogra-phy -e protocol has three entities as shown in Figure 4(i) publisher that publishes information under a certaintopic (ii) subscriber that obtains the same topic in-formation via a broker and (iii) public key generator(PKG) or broker that acts as the trusted third party -eprotocol has four phases namely (a) set-up phase (b)encrypt phase (c) publish phase and (d) decrypt phase

(a) Set-up phase publisher and subscriber provide aunique ID with attributes for registration with thePKG -is generates a secret master key set andpublic parameters that it publishes with set U A1A2 An For ciphertext policy-attribute-basedencryption (CP-ABE) the PKG gives a key set todevices with attributes [50]

(b) Encrypt phase based on the access tree the publishercreates an access policy An access tree is sent by thepublisher to the PKG for generating the key-gen-eration policy in the key policy- (KP-) based ABEalgorithm -e publishers know beforehand thetopics and subscribers to access as well as the rulesfor access Subscribers are issued with private keysfor all communication sessions Publishers in theCP-ABE algorithm come up with access trees andrules where the payload is encrypted by the pub-lisher who also issues more info needed for de-cryption and the rules With the help of public keycryptography data are encrypted giving rise to ci-phertext on the basis of KPCP-ABE and the receivergets SUBACK from the broker [51]

(c) Publish phase ciphertext is embedded in theSPublish command as payload and topic name isdefined in the header by the publisher -e brokerreceives the SPublish packet and replies withPUBACK prompting the publisher to respond withthe PUBREL packet -e broker then broadcasts totopic subscribers and deletes the data forwarding thePUBCOMP packet to the sender node

(d) Decrypt phase the encrypted information isdecrypted by the subscriber with the help of a secretkey Considering KP-ABE on verification that itsatisfies the access policy the subscriber invokes thePKG for a corresponding key which it sends afterverifying the request With CP-ABE subscribers usethe private key for decryption after authenticationthereby enabling offline interaction [51]Advantages

(i) -e information being encrypted makes it moresecure

34 MQTT for Sensor Networks (MQTT-SN) -is is a ver-sion of MQTT designed to operate under environments oflow bandwidth high link failures and short message lengthIt is a publishsubscribe (pubsub) protocol designed for

18 Journal of Computer Networks and Communications

UDP-based WSNs to visualize communication for low-power devices [5152] It comprises a publisher device actingas the MQTT-SN client to forward communications to agateway that also changes the MQTT-SN message to theMQTTmessage before delivering it to the broker that sendsit to MQTT-SN subscribers [51]

It has a gateway for changing protocols from MQTT toMQTT-SN (Figure 5) [52] -is protocol supports nodes in

the sleeping mode using an offline keep-alive procedure thatcan be used by battery-operated devices to get into thesleeping mode To support sleeping clients MQTT-SN has anew offline keep-alive procedure with which battery-oper-ated devices can go to the sleeping state in which theirmessages are buffered at the server until they are awake InMQTT-SN clients with no preconfigured gateway addresscan figure out the real address of their operational gateway

Publisher

Subscriber

Subscriber

Broker

Publisher

4G

Figure 3 Architecture of the MQTT protocol

Register by mentioning URI Register by mentioning URI

Access tree

Set-up

Publisher ReceiverPKG(broker)

Key management Key management

Encrypt data using public parametersand random key This is done using

CP-ABE Publish it as

Subscribe (topics)SUBACK

S-Publish (TopicEncrypt (data))S-Publish (TopicEncrypt (data))

Encryptand publish

PUBACK

PUBACK

PUBRELPUBRELPUBCOMP

PUBCOMP

Figure 4 SMQTT protocol

Journal of Computer Networks and Communications 19

within the network through a discovery procedure Simul-taneously we may have several gateways in a single wirelessnetwork that can be cooperative via load sharing

MQTT-SN architecture it is composed of MQTT-SN clients gateways and forwarders (Figure 5)rough the MQTT-SN gateway (GW) the MQTT-SN protocol is used by the MQTT-SN clients toconnect to the MQTT server Integration betweenthe MQTT-SN GW and the MQTT-SN server mightbe possible or not Translating between the MQTT-SN and the MQTT for a stand-alone GW is doneusing the MQTT-SN protocol If a GW is detachedfrom the network it can be accessed by the MQTT-SN clients through a forwarder is just encapsu-lates the received MQTT-SN frames that are on thewireless side before forwarding them to the GW un-altered It then decapsulates the received frames tooresending them to the client as sent by the GW [52]ere are 2 types of gateways (a) transparent gatewayand (b) aggregating gateway as shown in Figure 6

(i) Transparent gateway this launches and sustains adirect MQTT connection between the MQTT-SNclient and the MQTT server which is entirely re-served for end-to-end client-server messaging enumber of servers and that of MQTT-SN clientsconnected to the GW are the same is transparent

GW translates syntax between these protocols and iseasier to implement compared to the aggregatingGW But the MQTT server must support dierentsessions for all connected clients Some MQTTservers are congured to enable few parallel sessions

(ii) Aggregating gateway this maintains a single MQTTconnection to the server on behalf of all clients Allexchanged messages from the MQTT-SN client to theaggregating GW are directed to the GW which sortsthe information to forward to the server e aggre-gating GW decreases the number of MQTT sessionssimultaneously supported by the server in WSNs withmany battery-operated sensors and actuators (SAs)having limited storage and processing capabilities It ishowever more complex to implement compared to atransparent GWe general format of the MQTT-SNprotocol message has two or four message headeroctets and n octets of variable header that is optionaldepending on the message in question MQTT-SN has256 message types (MsgType) with those from0X1Eminus 0XFD 0X19 and 0XFF reserved for futureusage [51]

Advantages

(i) e protocol can operate under environmentsof low bandwidth high link failures and shortmessage length

MQTT-SN gateway

MQTT-SN

MQTT

MQTT-SN

MQTT-SN

MQTT-SN

MQTT-SN client

MQTT-SN clients

MQTT broker

MQTT-SN client

MQTT-SN client

MQTT-SN gateway

MQTT-SN forwarder

Figure 5 MQTT-SN architecture

20 Journal of Computer Networks and Communications

(ii) It is a pubsub protocol designed for UDP-basedWSNs to visualize communication for low-power devices

(iii) It can be optimal for low-cost battery-operatednodes having low processing power and lessstorage

(iv) It supports nodes in the sleeping mode using theobrvbarine keep-alive procedure used by battery-operated devices

35e Extensible Messaging and Presence Protocol (XMPP)e standardization of this protocol by the IETF made itpopular for usage over the Internet especially in instantmessaging and chatting platforms It is however susceptibleto spamming attacks and its failure to have worldwidesupport has made Google too withdraw its support of thestandard [17] e protocol is mainly useable in the client-server architecture but can also be applicable in client-clientXMPP communication [53] Among the application layerprotocols XMPP is the only one that supports both asyn-chronous (publishsubscribe) and synchronous (requestresponse) services while running over TCP but can still runover other protocols

Given the simplicity of its implementation and ease ofusability in XMPP instant messaging (IM) is suitable forXMPP and is implemented using the message stanzaltmessagegt ese stanzas are usable in one-to-one ormultiuser sessions or chatting or in error reporting andsending headlines Attribute type of the stanza denes themessage type e attribute type could be a chat group chatheadline normal or error An instant messaging network inXMPP can support many XMPP clients and servers with agateway congured to enable other IM network users toconnect to other XMPP users e XMPP server mainlycontrols sessions and XML streams to clients and servers inaddition to routing XML stanzas in the streams [53] eTCP ports 5269 and 5222 have been respectively registeredfor the XMPP server-to-server connections and XMPP clientconnections by the Internet Assigned Numbers Authority(IANA) XMPP enables specication of XMPP ExtensionProtocols (XEP) to enhance its usefulness [17]

XMPP security clients connect to the server usingtransport layer security (TLS) and simple authentication andsecurity layer (SASL) protocols which are at the same timeused by the servers for interdomain communications Fordata condentiality and integrity TLS will encrypt thestreams of XML for all sessions To conrm the identity of anyclient that needs to access the server the SASL protocol doesthe authentication of the XML stream For any communi-cating session with a server to start the client needs to resolvethe serverrsquos DNS hostname For protection of the server of theclient from third-party attacks the server maintains the au-thentication details secret ie IP address and access methodand only original connections of the server are required

In server-to-server communications the SASL protocolis employed for authentication and condentiality Server-to-server communications may as well be deactivated from aserver by an administrator depending on the organizationalpolicy To guard against domain-spoong attacks serverdialback can be a good solution since servers support itsecuring more the XML stanzas Nevertheless it cannot beused to authenticate secure or encrypt in-server streamsand the server identication resulting therein is not strongenough to be relied on It further cannot secure against DNSpoisoning attacks and IP session hijacking for remote do-mains [53] High-security domains can use TLS and SASL

Advantages

(i) It is popular for usage over the Internet especially ininstant messaging and chatting platforms

Disadvantages

(i) It is susceptible to spamming attacks(ii) It has failed to have worldwide support making

Google withdraw support of the standard(iii) It is not secure against DNS poisoning attacks and

IP session hijacking for remote domains

36 Advanced Message Queuing Protocol (AMQP) isprotocol can be utilized by a number of platforms formessage exchange since it is open source and written inmany languages It is a publishsubscribe model depending

MQTT-SNclients Broker

MQTT

(a)

MQTT-SNclients Broker

(b)

Figure 6 (a) Transparent gateway and (b) aggregating gateway

Journal of Computer Networks and Communications 21

on a reliable and efficient messaging queue with capacity forinstant messaging or e-mail for different applications es-pecially in IoT environments It differs from others in thesense that message types can be specified and have theirsource traceable and how to do trade-off between perfor-mance security and reliability For reliability of messagetransmission AMQPmakes use of some delivery guaranteeslike at most once at least once and exactly once It can alsohowever utilize the TCP transport layer for guaranteeingreliability -e AMQPrsquos publishsubscribe method has twocomponents namely (i) exchange queue meant for routingof messages to an appropriate order in the queue and (ii)message queue that acts as storage pending their delivery toreceivers [16] In this protocol model the broker can makerouting decisions contrary to othermessaging systems whereapps using the queue have in them embedded the decision-making logic for recipients or senders -e developer mayneed to check each of the affected apps so as to change thelogic for routing and delivery of the messages [54]

Advantages

(i) It is convenient(ii) It has capacity for interoperability between dif-

ferent vendors(iii) It can use open standards to enable connection

between business partners(iv) It can enable innovations based on AMQP foun-

dations [55](v) It can enable supporting of mission-critical ap-

plications in e-commerce(vi) It can guarantee reliable delivery of messages

Disadvantages

(i) It is not very appropriate for real-timeapplications

(ii) It may not offer automatic discovery(iii) Interoperability is not guaranteed(iv) It lacks open-source libraries for constrained

gadgets [49]

37 Data Distribution Service (DDS) A product of theObject Management Group (OMG) DDS is a pubsubprotocol meant for M2M communications with two sub-layers -ese are the data-centric pubsub sublayer re-sponsible for connecting anonymous data publishers tosubscribers and the data-local reconstruction sublayer that isoptional and integrates DDS into the application layer [56]-e DDS architecture connects system components asshown in Figure 7 Publishers and subscribers are discon-nected with respect to (i) time (ii) space (iii) flow andbehavior (iv) platforms and (v) programming languages[57] A DDS data publisher generates topics and publisher-subscriber coupling is represented as topic name data typeschema and publishers and subscribersrsquo QoS attributesDelivering of data is done by the publisher layer -e datawriter and publishers together make a decision about anynecessary alterations for subsequent forwarding to thesubscribers who need to send data to IoT applications Data

readers are responsible for delivering topics to the sub-scribers after reading through them [56]

Some of the major components of the DDS architectureinclude the following [57]

(i) Topic this is a logical path specifying publicationand subscriptionrsquos data type between DataWritersand DataReaders For a successful session to begintopic names types and DataWriters and Data-Readersrsquo QoS should correspond

(ii) Domain this is a logical setting for transmissionthat can be utilized for separating and optimizingnetwork transmission in a setup of similar appli-cations If DDS apps share a domain ID then theycan exchange data among themselves

(iii) DataWriter and DataReader these are terminalobjects for writing and reading messages re-spectively to and from a universal data space

(iv) Participant this is an object that stands for apublisher or subscriber of the DDS app in a do-main acting like a container for other objects

Advantages

(i) It is highly reliable and offers QoS support(ii) It is highly scalable(iii) It is appropriate for real-time apps(iv) It is fault tolerant(v) It has high level of interoperability(vi) It does not need brokers that may act as

bottlenecks(vii) It offers automatic discovery(viii) It has a user-defined data structure for topic(ix) It supports publisher and subscriber decoupling

Disadvantages

(i) It is originally developed for only stand-aloneLANs

(ii) A lot of memory is needed(iii) It has no open-source libraries to support con-

strained devices(iv) It is hard to design and configure compared to

other protocols [49]

38 RESTful Services REpresentational State Transfer (REST)-is is an architectural style based on a set of principles thatdefine and address networked resources It offers web servicesthat enable transmission from the HTTP device to the devicewithin the IoT architecture and enables clients to reach theserver [16]-is protocol utilizes HTTP commands to backupthe requestresponse messaging model HTTP is a popularand secure protocol in the WWW using TLSSSL [16] ARESTful application has the following characteristics

(i) State and functionality are separated into distrib-uted resources

(ii) All resources are distinctively named by the use ofunique HTTP commands (GET POST PUT orDELETE over the Internet)

22 Journal of Computer Networks and Communications

(iii) It is a clientserver stateless layered protocol thatsupports caching REST uses the same methods asHTTP for all requests or responses with POST andGET for creating and retrieving resources PUT forupdating and changing the state of resources andDELETE for removing resourcesAdvantages of REST over arbitrary web services(SOAP) in the IoT context

(i) It reduces overhead(ii) It reduces parsing complexity(iii) It is stateless(iii) It integrates more tightly with HTTP [58](iv) It is easy to implement and learnis together

with its natural t over HTTP makes it apreferred method for Web 20 applications toshowcase their data and secure M2M sessionsin the IoT

(v) Applications that support RESTful web ser-vices exhibit better performance than the re-source-constrained WSN nodes [59]

(vi) REST is important for the IoT since it supportscommercial M2M cloud platforms

(vii) Its implementation in smart phones and tabletapps is also easy since it needs only an HTTPlibrary that is accessible to all operating systemplatforms [56]

Disadvantages

(i) Although it enjoys wide usage in commercialM2M platforms it is not likely to dominatebecause it is hard to implement

(ii) Its HTTP usage makes it incompatible withconstrained M2M devices leaving its usage toend applications

39 WebSockets It is a web-based protocol operating on asingle TCP channel and providing full-duplex transmissions

It permits secure duplex communication between a clientexecuting unreliable code in a restricted environment and adistant node communicating from that code while using theorigin-based security model employed by web browsers overTCP [60] e protocol is not a requestresponse model andis not a publishsubscribe model So unlike other protocolsa web-socket session is established by the client initiating ahandshake to the server to launch the session is onceestablished a full-duplex client-to-server connection beginsasynchronously and runs until terminated by either party[56] It gives a model for browser-based apps utilizing bi-directional communication with no need of many HTTPconnections to servers

WebSocket comprises two parts ie (i) the handshakehaving a clientrsquos message and the serverrsquos handshake responseand (ii) the data transfer HTTP headers should be repeated byapplications in all client requests and server responses arisingfrom the endless polling which might increase the commu-nication overhead subject to the application [61] WebSocketcan be used to come up with scalable real-time web appli-cations since it can enable full-duplex transmission via onesocket across the Web Its security can be ensured with thehelp of TLSSSL is protocol is suitable for resource-con-strained devices and IoTapplications will not have a good userexperience with it since it operates a clientserver architectureNevertheless it can outcompete all TCP-based protocolsbecause of the following advantages

(i) It is t for real-time transmission(ii) It has improved security(iii) It minimizes transmission overhead(iv) It can provide ecient messaging systems using

WebSocket ApplicationMessaging Protocol (WAMP)[17]

310 Streaming (Simple) Text-Oriented Messaging Protocol(STOMP) is is a TCP-based protocol designed for textmessage-oriented middleware (MOM) with the use of

Topic

Datareader

Datareader

Datareader

Datawriter

Datawriter

Datawriter

Subscriber PublisherPublisher

Data bus (DDS domain)

Subscriber

Topic Topic

Figure 7 DDS architecture

Journal of Computer Networks and Communications 23

HTTP-like commands for interoperability among platformslanguages and brokers Data communication takes placebetween the client and the broker in line with commands(like CONNECT DISCONNECT ACK NACK SUB-SCRIBE UNSUBSCRIBE SEND BEGIN COMMIT orABORT) then headers in the form ltkeygt ltvaluegt (one perline) then a blank line and then body content terminatingin a null character Server-client communication takes placeover a MESSAGE RECEIPT or ERROR frame with headerand body content format being analogous MOM productssupporting STOMP include Apache ActiveMQ or FuseMessage Broker HornetQ Open Message Queue(OpenMQ) RabbitMQ and syslog-ng through its STOMPdestination plugin [62] In addition to these other appli-cation layer protocols include simple media control protocol(SMCP) lightweight local automation protocol (LLAP)simple sensor interface (SSI) lightweight M2M (LWM2M)XMPP-IOT and simple object access protocol (SOAP)

Advantages

(i) It is lightweight for constrained networks(ii) It has flexibility to choose quality of services with

the given functionality(iii) It is standardized by OASIS Technical Committee(iv) It is easy and quick to implement

Challenges and opportunities

Here security threats are imminent because of the open-endedness of the application layer [63] -ese could includethe following

(i) -ere is poor or lack of a security design for anapplicationrsquos function Coming up with a gooddesign to address this could be a good ideaopportunity

(ii) Some of the programs might have backdoorsevading otherwise secure controls to allow un-authorized access to network resources We requiremechanisms to patch up such loopholes

(iii) Applications that lack or have weak authenticationmechanisms can be easily targeted by unlicensedusers to abuse the network Such mechanisms needto be strengthened further

(iv) Excessively complicated access control rules maybecause of their misunderstanding or poor writinggive unauthorized access allowing dubious activityrather than protecting the network they purport toprotect -ese should be simple and clear to un-derstand by all users

(v) In booting workstations without disks and managingnetwork devices we usually use the TFTP protocolalthough access does not require authenticating theuser by usernamepassword -is makes an intruderable to gain easy access to configaccess info as longas he can guess the filenames -is needs to be fixed

A few steps can be used to secure the application layerTo protect network data transmitted across the network

applications need to strengthen their encryption and au-thentication mechanisms Apps too need to enable strictcontrols of privileges for data access using good mechanismsto balance between usability and effectiveness [63] Appli-cations handling sensitive data need detailed logging andaudit capability as well as testing and reviewing

Table 3 summarizes the application layer protocols dis-cussed with their weaknesses and differences among them

4 Cross-Layer QoS Strategies

41 Service-Differentiated Real-Time Communication Scheme(SDRCS) -is is an event-driven routing protocol that routesreal-time traffic by featuring a cross-layer packet-forwardingdesign in which it integrates the real-time routing functionalitywith a new priority-based MAC scheme -e protocol ap-proximates distributed packet traversal speed for traffic clas-sification and admission control on the basis of this design Itfurther localizes decision-making by prioritizing packet for-warding to maximize the speed of packets [64]

Advantages

(i) It improves bandwidth utilization bandwidthdegradation due to unschedulable packets can beavoided

(ii) It offers a lightweight packet schedulability estima-tion mechanism Here it uses received signalstrength and admission control and then earlymissed-deadline packet-dropping policies are made

(iii) It delivers high end-to-end throughput via highersource data rates alongside strict end-to-end latencyneeds

(iv) Nodes which meet real-time requirements of lesstraffic load and better channel quality are givenhighest priority when it comes to packet forwarding

-e SDRCS operates via 5 mechanisms namely (1) RSS-(Rich Site SummaryReally Simple Syndication-) basedgrouping (2) admission control (3) prioritized queuing(4) real-time MAC and (5) dynamic forwarding -isprotocol has a queuing policy that employs per-hopdeadline-based queues and the nodes schedule theirpackets using FIFO priority-based queues [1] -e prin-cipal features of this protocol are mentioned as follows

(i) It is event driven and is easily adaptable to anynetwork changes

(ii) It does not need additional hardware for locali-zation or multichannel transmission

(iii) -e cross-layer packet-forwarding strategy can beused when transmitting multimedia traffic

(iv) It is able to efficiently circumvent any voids therein

Early deadline miss (EDM) policy drops all unscheduledpackets -e SDRCS circumvents wrong packet dropsdue to EDM since it adjusts to network changes easily

42 Sensor Fuzzy-Based Image Transport (SUIT) -is pro-tocol regulates congestion using a scheme based on fuzzy

24 Journal of Computer Networks and Communications

logic It transmits packets with reduced quality to allowablelevels if there is congestion enabling it to transmit lower-quality packetsframes without being dropped-e numberof packets delivered per second thus rises and while thisenhances video streaming apps it is more appropriatelymade for those that quickly transmit non-real-time videosuch as in video surveillance SUIT can stream JPEG pic-tures other than the traditional predictive video codingmethods that are not suitable for sensor devices because oftheir constraints of energy memory and computationalspeed Progressive JPEG-PJPEG performs better via packettransmission rate and delay though it degrades the imagequality because of congestion [65] -is protocol employscross-layer methods for interlayer data exchange though itstill employs some layering making it still possible topreserve all advantages of the layered design since it hasapplication and transport layers but has noMAC or routinglayer

For congestion detection and control SUIT bases on thepercentage number of in-bound and out-bound frames perwindow number of contenders and buffer occupancy of thenext-hop node Since buffer management has an effect oneffectiveness and QoS inWMSNs it is useful in the transportlayer so the protocol prioritizes packets for better QoS SUITreorders packets by tracking the sequence numbers of thedelivered packets since they all have a source ID framenumber and sequence number that it uses to detect anymissing packets -e protocol however does not providepacket reliability [65]

Advantages

(i) -e protocol prioritizes packets for better QoS(ii) It can transmit lower-quality packetsframes

without being dropped

Disadvantages

(i) It does not provide packet reliability

43 Network Layer QoS Support Enforced by a Cross-LayerController (NLQS) -is scheme permits packet-level servicedifferentiation as a function of throughput end-to-endpacket error rate and delay [66] (see Figure 8) -is en-hances network layer QoS It comprises a cross-layer controlunit (XLCU) for configuring and controlling networkingfunctionalities at physical MAC and network layers [18]-is depends upon the unified logic which affects choices forapplication layer needs and status of functional blocks whichdo implementation of networking functions [66]

Advantages

(i) Cross-layer interactions can be controlled withoutweakening the upgradability modularity andsimplicity of designing the system

44 Cross-Layer Signaling Shortcuts (CLASS) CLASS has ahigh level of effectiveness flexibility and comprehensive-ness According to [67] it has the following features

(i) Direct signaling among remote layers leading tofaster transmission

(ii) Lightweighted internal message format standard-ized protocols are not suitable for use in internalsignaling since they are not usually lightweight forinstance transmitting against faults in the networkCLASS requires only three fields namely destina-tion address event type and event contents

(iii) -e setup of its exterior messages is standardconsidering external signaling ICMP and TCPIPare suitable for general messages and short notifi-cations respectively

(iv) Message control protocol should ensure optimizedand organized exchange of dense simultaneousmessages across layers for high efficiency andavoidance of possible conflicts -e protocol appliesto many cross-layer signaling situations

Table 3 Differences and problems with application layer protocols

Protocol Architecture Transport QoS options Security Weaknesses

CoAP Requestresponse UDP DTLS No in-built security features DTLS does not supportmulticast

MQTT Publishsubscribe TCP TLSSSL Insufficient security at the protocol level sincepayload values are not encrypted

SMQTT Publishsubscribe TCP CPKP-ABE Key revocation and group pubsub for distributedSMQTT are still a challenge

MQTT-SN Publishsubscribe UDP TLSSSL SSLTLS suffers from attacks like BEAST CRIMERC4 and Heartbleed

XMPP RequestresponsePublishsubscribe TCP X TLSSSL Susceptible to spamming attacks and lacks worldwide

supportAMQP Publishsubscribe TCP TLSSSL Inappropriate for real-time applications

DDS Publishsubscribe TCPUDP TLSSSL Memory intensive and no open-source libraries forconstrained devices

REST Requestresponse HTTP X HTTPS Hard to implement Uses HTTP so incompatiblewith constrained apps

WebSockets PublishsubscribeClientserver TCP X TLSSSL Bad user experience for IoT apps since it runs the

clientserver architectureSTOMP Clientserver TCP HTTP -e broker can act as a bottleneck

Journal of Computer Networks and Communications 25

Advantages

(i) e protocol is scalable and has a very lowpropagation delay

(ii) It is very ecient and exible since it uses directsignaling between any two layers

(iii) It is very comprehensive and eective

45 Cross-Layer and Multipath-Based Video Transmission(CMVT) CMVT is a hybrid of application and networklayers At the application layer it carries out the encoding ofvideo streams to video data frames (I- P- and B-frames) bymeans of the MPEG-4 encoding format At the networklayer route discovery and data transfer take place Twoalgorithms greedy forwarding and rollback are used todiscover a number of paths from the source to the sink nodevia the route discovery method [20] A specied node icalculates the evaluation of its adjacent node j using thefollowing equation

fij (1 minus α)d2(j D) minus d2min(i)d2max(i) minus d

2min(i)

+ αeinit(j) minus eres(j)

einit(j)

(19)

where fij evaluation value for the node i to j d2(j D) isthe distance from the node j to the destination node D

d2min(i) and d2max(i) are the minimum and maximum dis-

tances of neighbors of the node i to D respectively einit(j) isthe initial energy of the node j eres(j) is the current residualenergy of the node j and α is the energy coecient given asfollows

α emax(i) minus emin(i)

emax(i) (20)

where emax(i) and emin(i) are respectively the maximumand minimum energy left of all node irsquos neighbors

e network layer further transmits video streams withCMVT doing status evaluation choosing an appropriatecommunication path for any packets and the QoS guaranteelevel for the path i is calculated as

fi (1 minus ω)hisum hi

+ ωnisum ni

(21)

where fi is the evaluation value for the path i hi are the hopsfor the path i ni is the summation of packets sent via isum ni isthe summation of packets sent by sources and ω is theenergy consumption factor [18]

Advantages

(i) e CMVT protocol is very superior in mediatransmission particularly large-scale WMSNs

To network peers

Flow requirements

-

Atimes

Admissioncontrol

Capacityassignment

Next-hopselection

Coding rate

Data rate

QoScontracts

Cross-layer controlunit (XLCU)

Localization subsystem

Synchronizationsubsystem

Networkcoordinationunit (NCU)

UWB modulator

Channel coder

QoS scheduler

Georouting

Source coder

Rate control

Figure 8 Cross-layer controller architecture

26 Journal of Computer Networks and Communications

46 Cross-Layer Cooperative MAC (CoopMAC)CoopMAC [68] is based on the IEEE 80211 DCF approach-ere is a reduction in interference of neighboring cellsleading to uniform coverage in densely deployed networks[68] A connection is established by two-way authenticationbetween the sender and the receiver When a Clear-to-Send(CTS) packet and short interframe space (SIFS) are obtainedby the sender packets are transmitted right to the receiverwith no valuable cooperation If available the two nodes findout whether they can exchange data -is is done by the useof a helper identification signal to establish how feasiblecooperative communication is It is established in thepresence of a signal and if it is not available direct com-munication is activated For distributed systems helper-initiated cooperation is preferable because of the presence ofthe RTSCTS packets [69]

Advantages

(i) CoopMAC has an advantage over others due to thespatial diversity amongst the three nodes [70]

(ii) It has the potential to achieve significantthroughput and delay with no complexity in thedesign of the system

47 Cooperative MAC Protocol for Multihop Networks (M-CMAC) Just as CoopMAC M-CMAC is made in such away that high data rate stations help low data rate stations toforward traffic for broadcasting [71] -ere are also helperschosen in such a way that 2 fast-hop transmissions replace aslow hop transmission -e helper via whom there existsleast delay from the sender to the receiver is considered bestbecause of theminimum two-hop transmission rate and so isthe chosen neighbor to the duo Here the working as-sumption is that all nodes have their position coordinatesknown and thus the Euclidean distances between all pairscan be calculated and converted to the data rate for that linkAll nodes have cooperative tables (CT) of potential helperswith destination and helper MAC address Euclidean dis-tance and total distance through the helper Any sender withdata to transmit checks out for the existence of any helper inthe CT for the destination and if available forwards RTS forchannel reservation for single-hop duration M-CMAC hasgot an RTS format having a shape with five fields similar tothe one given as follows

Framecontrol Duration Source

addressDestination

addressHelperaddress

-e destination nodersquos address is kept by the source in thepart for the helper address that is in turn saved as the desti-nation address Nodes inspect helper and destination addressfields on receipt of RTS whereby the node acts as the othernodersquos helper in case the helper address field is not the sameand the helper transmits CTS back to the source if at all it wantsto forward data On delivery of the CTS packet the sourcetransmits to the helper which in turn sends the packets to thedestination that also sends ACK to the helper on receipt of thepacket In M-CMAC there exists a higher level of channelreuse (parallel transmissions) due to the increased number of

nodes therein that subsequently raises availability of helpers fordata forwarding -is results in increased throughput incomparison with CoopMAC and IEEE 80211 DCF [71]

Advantages

(i) It promises a higher throughput compared toCoopMAC and IEEE 80211 DCF

48 Cluster-Based Cooperative Routing (CBCR) Protocol-e CBCR protocol has a multihop data-forwarding func-tion realized at the link layer with cooperative links that useM-CMAC -is protocol encompasses two stages namelythe routing relay selection phase and data forwarding phase

(i) Routing relay selection phase all nodes announcetheir presence to their neighbors by broadcastingperiodical beacon messages that carry the MAC ad-dress of the node Each of these constructs a relaytable that contains all neighboring nodes with whichit is able to communicate -e node further broad-casts its neighbor list in case of any change to itsentries since the previous broadcast -e MAC ad-dresses of nodes next to the node X are contained inits relay tablersquos column one and the neighbor nodersquosrow has MAC addresses for neighbors of the adjacentnode Based on its relay table each node in-dependently chooses routing relays Choosing a nodeas a relay node depends on the number of nodes itconnectsmdashit should be the highest number [71]

(ii) Data forwarding phase a node with packets totransmit needs to first verify if the receiver is in thesame cluster and if that receiver has a helper to whichthe packet is thence forwarded but in whose absencethe packet is delivered right to the receiver In casethe intended receiver is in a different cluster therelay table will be checked to see whether that re-ceiver is reachable via other routing relays If it isreachable packets will be sent to the routing relaydirectly or via the helper if at all the relay has one Incase of destination unreachable via relays packetsare multicast to all relays by the node [71]Advantages

(i) It enables multicasting

49 MAC-PHY Cross-Layer Protocol A cooperative cross-layer standard for cooperation at the physical layer in next-generation WMSNs is developed in [70] It provides acomplete MAC layer algorithm that gives a supportive shellfor the PHY-MAC layer Like CoopMAC its scheme de-pends on the middle node for communication betweennodes-eMAC layer protocol has been changed to regulateinformation exchanges at the physical layer -e receivernode receives duplicate packets from source and helpernodes to decode the data [70]

410 MAC-Centric Approach -is cross-layer protocoltargets multimedia applications by using the MPEG-4

Journal of Computer Networks and Communications 27

scheme [21] It is characterized by 4 access categories(AC3ndashAC0) according to their priority in transmission It ismeant for supporting different QoS needs in upcomingvideo applications and enabling MAC layer differentiationfor H264 partitioning [18] Because of low bandwidth delayand other QoS challenges that cause inefficiency in trans-mittingmultimedia data inWMSNs a number of algorithmsdepending on IEEE 80211e have been proposed to supporttransmission of quality videos [72] Selecting an AC dependson the measures for QoS such as the loss rate and delay-us AC3 having the highest priority is mapped to theparameter set concept since the stream is sensitive to bit lossespecially in video transmission [21]

411 Adaptive Cross-Layer Forward Error Correction(ACFEC) In the ACFEC [73] model data packets are ex-changed among nodes through the access point (AP) whichoperates the infrastructure mode in which it adds FEC to videodata [72] -ese data are dealt by encapsulation using astreaming server to the receiver through thewireless AP as RTPpackets An adaptive FEC controller senses the category ofpackets out of the RTP header and recovers the header ofpackets from UDP -e encoder generates some error-cor-recting packets whose number is determined by the sourcepacket number of the block Multimedia transmissions aremonitored by the controller using MAC failure data and itscounter is increased by one if there is a failed transmission-econtroller uses the failure counter to change the packet numberproduced after transmitting a block [18] In case there are lostpackets it changes the redundancy rates and produces extrapackets to replace the lost ones and satisfy the requirements ofthe receiving node [72] -e FEC packet number is enhancedor reduced to satisfy the receiverrsquos requirements and stoppacket losses-is is done through accurate detection of packetlosses and adjustment of redundancy rates If all video datapackets are well received there will not be generation of FECpackets [73]

Advantages

(i) It promises a good QoS through packet loss re-duction and redundancy rate adjustment

412 Balanced Cross-Layer Fuzzy Logic (BCFL) DesignRouting Algorithm in WSNs A new fuzzy logic-basedrouting algorithm (BCFL) [74] was designed using disper-sion of the cross-layer parameter as the fuzzy logic inferencesystem input Each cross-layer parameter has a dynamicweight depending on the value of dispersion -e designcomes with some innovations as follows (i) for fuzzy logicinference system input the absolute parameter value issubstituted by parameter dispersion thereby significantlyreducing algorithmic complexity (ii) dispersion does notchange with the order of magnitude according to the dis-persion formula and (iii) the weight of the parameter de-pends on the size of its dispersion and the two are inverselyproportional to each other -is enables BCFL to have somenobility unlike other algorithms as per the following dis-tinguishing properties making it advantageous

(i) It has got simple if-then rules which remain con-stant even when the constraints increase

(ii) It is capable of dealing with many constraints withno increment in complexity

(iii) It has capacity to yield a more balanced solutionthan other algorithms

(iv) It is easily adaptable to changes in network con-ditions and topology even when the changes arefrequent like in underwater WSNs -e algorithm isuseable in choosing the CH in cluster-based routingprotocols [74]

413 Minimum Hop Disjoint Multipath Routing Algorithmwith Time Slice Load-Balancing Congestion Control SchemeMHDMwTS is a two-phased routing protocol comprisingpath build-up and path acknowledgment phases It consists ofmultiple sources each of which has three build-up disjointpaths namely primary alternate and back-up paths Onactivation the source node requests to build up a route to thenearest hop neighbor which is the path build-up phase [19] Inthis phase step one has source activation in which therequested node for path building adds its number and time-stamp and sends to the least hop-count neighboring node-isgoes on until the least time latency sink having the informationneeded to construct the primary route is reached [75] In steptwo there is path extraction when the new package fromanother path arrives whereby the extracted path is comparedwith the primary path If the node is shared the package will berejected or an alternative path will be searched for to get abackup through comparison of the preceding two paths Phasetwo is path acknowledgment where we have the third step inwhich the sink returns the ACK packet to the sender with pathinformation having nodes and their time information after ithas computed it using the timestamp [1819]

Advantages

(i) -e protocol minimizes end-to-end latency(ii) It enhances congestion control

414 Cross-Layer Optimal Design (CLOD) Authors in [76]come up with a CLOD for scheduling at the data link layerrouting at the network layer and controlling congestion atthe transport layer with an assumption of fixed link capacity-rough congestion control energy efficiency is improvedTransport layer congestion at the nodes is minimized bycompressed sensing (CS) in which transmitted bits are re-duced whereas optimally allocating resources decreasescongestion on the links at the data link layer CLODpromises minimized computational complexity and betterperformance in light traffic scenarios [77] Generating andstoring CLOD is more efficient compared to Gaussianrandom matrices It prolongs network lifetime and savesenergy

Advantages

(i) It does congestion control subsequently increasingenergy efficiency

28 Journal of Computer Networks and Communications

(ii) It reduces computational complexity(iii) It improves performance in light traffic scenarios

415 Challenges and Opportunities of the Cross-Layer Design(CLD)

(i) Different layers in a real network assume differentfunctions andor services A layer only communi-cates with its neighboring layers -e layered modeldegrades system performance because of many

features in wireless transmission QoS parameters inthe layers of WMSNs are shared by routing pro-tocols to optimize the performance Neverthelessthere is a need to design cross-layer models forincreased efficiency in routing [1]

(ii) -e physical layer plays a highly vital role in CLDRate adaptation and channel allocation take place atthe physical layer via signal processing to enhanceQoS End-to-end performance in wireless media isaffected by the changes therein given the effect on

Table 4 A summary of some of the reviewed cross-layer models

Protocol Layers Aims Comments QoS parameters

SDRCS MACPHY Routing real-time traffic Transmits multimedia traffic viacross-layer packet forwarding

-roughputlatency

SUIT Applicationtransport Transmission of non-real-time video Uses fuzzy logic to regulate

congestionTransmission rate

delay

CLASS Any twoDesign serves as a framework fordifferent implementations ofdifferent application scenarios

Scalable very efficient flexible andhas very low propagation delay

Propagation delayjitter

NLQS PHYMACnetwork

Permits packet-level servicedifferentiation as a function of

throughput packet error rate anddelay

Network layer QoS is enhanced -roughput delaypacket error rate

CMVT Applicationnetwork

Encodes video streams to video dataframes via MPEG-4 encoding anddoes route discovery and data

transfer

Greedy forwarding and rollback areused to find source-to-sink paths Energy

CoopMAC PHYMAC Offers spatial diversity among thethree nodes

Helper ID signal finds out howfeasible cooperative communication

is

-roughput ratedelay

M-CMAC PHYMAC Increases end-to-end throughputand packet delivery ratio

Euclidean distances between nodesare calculated and converted to the

data rate for a link

-roughputpacket delivery

ratio

CBCR PHYMACMinimizes control overhead andtime consumed in establishing thecooperative paths than M-CMAC

Energy consumption is moreuniformly distributed in a network

enhancing network lifetime

-roughputpacket deliveryratio energy

MAC-PHY MACPHYMeant to enable PHY layer

cooperation and maximize gains ofcooperation at the MAC layer

Leverages both spatial diversity andcoding gain -roughput delay

MAC-centric MACAPPL

Meant to support QoS needs in newvideo apps and enable MAC layer

differentiation for H264partitioning

Targets multimedia applications byusing the MPEG-4 scheme

Delay packet lossrate

ACFEC MACnetwork(UDP)

Meant to enhance the quality ofvideo streaming over 80211WLANs

and overcome packet losses

Adjusts redundancy rates toovercome channel fluctuations and

detect and reduce packet lossPacket loss rate

MHDMwTS mdash

Meant to provide reliable datatransfer with the multipath routingand load-balancing congestioncontrol method in WMSNs

More reliable than basic routingschemes for transport multimedia

data

Latency packagetransmit rate

BCFL mdash

Introduces dispersion into fuzzylogic-based routing and sets every

cross-layer parameter with adynamic weight

Can be used to select a CH in cluster-based routing protocols andproposes a dispersion formula

Node utilitydispersion

CLODDatalinknetworktransport

Prolongs network lifetime andachieves congestion control anddesigned for lightly loaded WSNs

Assumes fixed link capacity andintegrates compressed sensing

technology

-roughputaverage energy CS

error ratio

Journal of Computer Networks and Communications 29

the purposefulness of the protocols at the networklayer CLD offers solutions to conserving powerminimizing energy and controlling both the flowand congestion in the network as well as fault tol-erance so it is an opportunity for designers con-sidering other layers We further desire to developCLDs that shall make an impact on network oper-ation and get close attention [18]

(iii) CLD interface standardization the architecturemust deliver module functionality although thereexist some queries amongst modules about possibleinterfaces predicted by the necessity to share in-formation among remote protocol layers Technicalchallenges include developing designing andstandardizing of cross-layer interfaces and algo-rithms that satisfy the cross-layer optimization re-quirements among protocol layers

Table 4 summarizes these cross-layer models with a fewcomments

5 Future Research Direction

51 At theMACLayer In WMSNs achieving duty cycling istricky since video traffic is volatile Listening to adjacentnodes wastes energy We need to switch states from awake tosleep -is requires further research

Self-organizing networks and green communicationalthough most CRWSN energy-saving algorithms aredesigned to reduce node transmission power energyconsumption is inevitable during the operational mode ofBSs because of internal processing Scholars are trying withpolicy-makers to encourage a shift to greenWSNs to reduceoperational costs and carbon footprints With self-orga-nizing CRWSNs for automatic switch-off when idle energyis saved which calls for cognitive switching algorithms Sointegrating cognitive switching with radio resource allo-cation in CRWSNs is a promising research area [78]

52 At the Application Layer Considering a WMSN ar-chitecture end users access the network via the BS to andfrom which they send their sensed data We desire to getguaranteed QoS and energy efficiency by designing a newarchitecture in the form of hardware and software whichneeds to be investigated further for a fix of the same [1]

QoS and energy efficiency are very crucial in WMSNsand mostly real-time apps that require guaranteed band-width and throughput in their network lifetime Mostprotocols ignore base station (BS) mobility and WMSNnodes Traffic management telemedicine and battlefieldsurveillance apps require mobile nodes or BSs so designingdynamic routing protocols to be adaptable in these cir-cumstances is necessary

53 Cross-Layer Security of data transmitted on WMSNseg in military surveillance and e-commerce is key withQoS and energy efficiency WMSNs are vulnerable to attacks

like worms and sinkhole Moreover their computationpower is not that high complicating implementation ofstrong secure protocols -is needs further attention

For fairness and priority issues in CRWSNs nodes areassigned diverse priorities depending on importance andurgency of their data For better network performance thereis a need to develop fair resource allocation mechanismswith priority-based fairness in sensors-ere is still potentialin this research area [78]

We must develop multiobjective adaptive protocols tooptimize QoS metrics involved in routing for best trans-mission results to get a trade-off between different opti-mization metrics We need to design standardized cross-layer algorithms which are able to meet cross-layer opti-mization standards

6 Conclusion

We have reviewed the different QoS strategies for WSNs inthe context of IoT from the MAC layer and application layeras well as the cross-layer paradigm For the MAC layer wehave reviewed protocols for WSNs such as the hybrid of thecontention-free and contention-based MAC protocols andthose for WMSNs We further reviewed a number of ap-plication layer protocols including many machine-to-ma-chine requestresponse and publishsubscribe protocols Forsystem optimization cross-layer QoS strategies are veryimportant in wireless communication We have reviewed anumber of cross-layer strategies For all categories reviewedchallenges and opportunities are discussed Finally somepossible future directions are discussed for research andapplication showing a promising potential for future re-search in this relatively new area especially as more WMSNapplications emerge recently

Conflicts of Interest

-e authors hereby declare no conflicts of interest regardingthe publication of this work

Acknowledgments

-is research was funded by the National Natural ScienceFoundation of China (Nos 61602398 61672447 and61711540306)

References

[1] V Bhandary A Malik and S Kumar ldquoRouting in wirelessmultimedia sensor networks a survey of existing protocolsand open research issuesrdquo Journal of Engineering vol 2016Article ID 9608757 27 pages 2016

[2] I F Akyildiz T Melodia and K R Chowdhury ldquoWirelessmultimedia sensor networks applications and testbedsrdquoProceedings of the IEEE vol 96 no 10 pp 1588ndash1605 2008

[3] M Iqbal M Naeem A Ahmed M Awais A Anpalagan andA Ahmad ldquoSwarm intelligence based resource managementfor cooperative cognitive radio network in smart hospitalsrdquoWireless Personal Communications vol 98 no 1 pp 571ndash5922018

30 Journal of Computer Networks and Communications

[4] S A Alvi B Afzal G A Shah L Atzori and W MahmoodldquoInternet of multimedia things vision and challengesrdquo AdHoc Networks vol 33 pp 87ndash111 2015

[5] Q F Hassan A U R Khan and S A Madani Internet ofings Challenges Advances and Applications CRC PressTaylor amp Francis Group A Chapman amp Hall Group BocaRaton FL USA 2018

[6] Q Wang Y Zhao W Wang et al ldquoMultimedia IoT systemsand applicationsrdquo in Proceedings of the 2017 Global Internet ofings Summit (GIoTS) vol 2 Geneva Switzerland June 2017

[7] A Aslam and E Curry ldquoTowards a generalized approach fordeep neural network based event processing for the internet ofmultimedia thingsrdquo IEEE Access vol 6 pp 25573ndash25587 2018

[8] M Jridi T Chapel V Dorez L Bougeant and A Le BotlanldquoSoC-based edge computing gateway in the context of theinternet of multimedia things experimental platformrdquoJournal of Low Power Electronics and Applications vol 8no 1 p 1 2018

[9] H Noura A Chehab L Sleem andM N Rapha ldquoOne roundcipher algorithm for multimedia IoT devicesrdquo MultimediaTools and Applications vol 77 no 14 pp 18383ndash18413 2018

[10] T Balan D Robu and F Sandu ldquoMultihoming for mobileinternet of multimedia thingsrdquo Mobile Information Systemsvol 2017 Article ID 6965028 16 pages 2017

[11] H Shen and G Bai ldquoRouting in wireless multimedia sensornetworks a survey and challenges aheadrdquo Journal of Networkand Computer Applications vol 71 pp 30ndash49 2016

[12] A Ahmad M H Rehmani H Tembine O A Mohammedand A Jamalipour ldquoIEEE access special section Editorial optimization for emerging wireless Networks IoT 5G andsmart grid communication networksrdquo IEEE Access vol 5pp 2096ndash2100 2017

[13] B A Muzakkari M A Mohamed M F A Kadir andZ Mohamad ldquoRecent advances in energy efficient-QoS awareMAC protocols for wireless sensor networksrdquo InternationalJournal of Advanced Computer Research vol 8 no 38pp 212ndash228 2018

[14] M A Yigitel O D Incel and C Ersoy ldquoQoS-aware MACprotocols for wireless sensor networks a surveyrdquo ComputerNetworks vol 55 no 8 pp 1982ndash2004 2011

[15] A M Abbas and O Kure ldquoQuality of Service in mobile ad hocnetworks a surveyrdquo International Journal of Ad Hoc andUbiquitous Computing vol 6 no 2 p 75 2010

[16] M Q Shatnawi ldquoApplication layer protocols for the internetof things a surveyrdquo in Proceedings of the InternationalConference on Engineering amp MIS (ICEMIS) Agadir Mo-rocco September 2016

[17] V Karagiannis P Chatzimisios F Vazquez-gallego andJ Alonso-zarate ldquoA survey on application layer protocols forthe internet of things research motivationrdquo InternationalJournal of Advanced Research in Computer Science vol 8no 3 pp 1ndash10 2015

[18] M S Bernard T Pei Z Li and K Li ldquoQoS strategies forwireless multimedia sensor networks in the context of IoTrdquoLecture Notes of the Institute for Computer Sciences SocialInformatics and Telecommunications Engineering vol 4pp 228ndash253 2019

[19] A AlAmri and M Abdullah ldquoCross-layer quality of serviceprotocols for wireless multimedia sensor networksrdquo in Pro-ceedings of the International Conference on CommunicationManagement and Information Technology pp 649ndash658Cosenza Italy April 2017

[20] J Guo L Sun and R Wang ldquoA cross-layer and multipathbased video transmission scheme for wireless multimedia

sensor networksrdquo Journal of Networks vol 7 no 9pp 1334ndash1340 2012

[21] A Ksentini A Naimi and M Gueroui ldquoToward an im-provement of H264 video transmission over IEEE 80211ethrough a cross-layer architecturerdquo IEEE CommunicationsMagazine vol 44 no 1 pp 107ndash114 2006

[22] W Ye J Heidemann and D Estrin ldquoAn energy-efficientMAC protocol for wireless sensor networksrdquo in Proceedings ofthe 21st International Annual Joint Conference of the IEEEComputer and Communications Societies (INFOCOM 2002)New York NY USA June 2002

[23] J Polastre J Hill and D Culler ldquoVersatile low power mediaaccess for wireless sensor networks categories and subjectdescriptorsrdquo in Proceedings of the Second InternationalConference on Embedded Networked Sensor Systems Sen-Sysrsquo04 pp 95ndash107 Baltimore MA USA November 2004

[24] T Danmanee K N Nakorn and K Rojviboonchai ldquoCU-MAC a duty-cycle MAC protocol for internet of things inwireless sensor networksrdquo ECTI Transactions on ElectricalEngineering Electronics and Communications vol 16 no 2pp 30ndash43 2018

[25] T Van Dam and K Langendoen ldquoAn adaptive energy-efficientMAC protocol for wireless sensor networksrdquo in Proceedings ofthe 1st International Conference on Embedded NetworkedSensor Systems Los Angeles CA USA November 2003

[26] P Xie and J Cui ldquoR-MAC an energy-efficient MAC protocolfor underwater sensor networksrdquo in International Conferenceon Wireless Algorithms Systems and Applications (WASA2007) Chicago IL USA August 2007

[27] A Rahim N Javaid M Aslam U Qasim and Z A KhanldquoAdaptive-reliable medium access control protocol forwireless body area networksrdquo in Proceedings of the AnnualIEEE Communications Society Conference on Sensor Meshand Ad Hoc Communications and Networks (SECON) SeoulKorea June 2012

[28] H S Kwak S Ullah D H Kwak and C H Lee ldquoA powerefficient MAC protocol for wireless body area networksrdquoJournal of the Korean Institute of Intelligent Systems vol 8no 6 pp 131ndash140 2009

[29] C Omeni A J Burdett and M Park ldquoEnergy efficientmedium access protocol for wireless medical body area sensornetworksrdquo in Proceedings of the 4th IEEEEMBS InternationalSummer School and Symposium on Medical Devices andBiosensors pp 29ndash32 Cambridge UK August 2007

[30] L Sitanayah C J Sreenan and K N Brown ldquoER-MAC ahybrid MAC protocol for emergency response wireless sensornetworksrdquo in Proceedings of the Fourth International Con-ference on Sensor Technologies and Applications pp 244ndash249Mestre Italy July 2010

[31] Y Sun and D B Johnson ldquoRI-MAC a receiver-initiatedasynchronous duty cycle MAC protocol for dynamic trafficloads in wireless sensor networksrdquo in Proceedings of the 6thACM conference on Embedded network sensorsystemsmdashSenSysrsquo08 pp 1ndash14 Raleigh NC USA November2008

[32] M Buettner G V Yee E Anderson and R Han ldquoX-MAC ashort preamble MAC protocol for duty-cycled wireless sensornetworksrdquo in Proceedings of the 4th International Conferenceon Embedded Networked Sensor Systems SenSys 2006pp 307ndash320 Boulder Colorado USA October 2006

[33] S Boulfekhar and M Benmohammed ldquoSynchronous receiverinitiated MAC protocol for long-lived sensor networksrdquoComputers amp Electrical Engineering vol 40 no 2 pp 504ndash516 2014

Journal of Computer Networks and Communications 31

[34] S Mehta and K S Kwak ldquoH-MAC a hybrid Mac protocol forwireless sensor networksrdquo International Journal of ComputerNetworks amp Communications vol 2 no 2 pp 108ndash117 2010

[35] I Rhee A Warrier M Aia J Min and M L Sichitiu ldquoZ-MAC a hybrid MAC for wireless sensor networksrdquo IEEEACM Transactions on Networking vol 16 no 3 pp 511ndash5242008

[36] S Arshad A Al-sadi and A Barnawi ldquoZ-MAC performanceevaluation and enhancementsrdquo Procedia Computer Sciencevol 21 pp 485ndash490 2013

[37] Y Liu I Elhanany and H Qi ldquoAn energy-efficient QoS-aware media access control protocol for wireless sensornetworksrdquo in Proceedings of the IEEE International Confer-ence on Mobile ad hoc and Sensor Systems pp 9ndash11 Wash-ington DC USA December 2005

[38] J Decotignie ldquoWiseMAC an ultra low power MAC protocolfor the downlink of infrastructure wireless sensor networksrdquoin Proceedings of the Ninth International Symposium onComputers And Communications pp 244ndash251 AlexandriaEgypt June 2004

[39] R A Rashid ldquoDevelopment of energy aware TDMA-basedMAC protocol for wireless sensor network systemrdquo EuropeanJournal of Scientific Research vol 30 no 4 pp 571ndash578 2009

[40] H P Van Hoesel ldquoA lightweight medium access protocol(LMAC) for wireless sensor networksrdquo in Proceedings of theInternational Workshop on Networked Sensing Systems Bal-timore MD USA November 2004

[41] C Li P Wang H Chen andM Guizani ldquoA cluster based on-demand multi-channel MAC protocol for wireless multi-media sensor networksrdquo in Proceedings of the IEEE In-ternational Conference on Communications pp 2371ndash2376Beijing China May 2008

[42] J L Sobrinho and A S Krishnakumar ldquoQuality-of-service inad hoc carrier sense multiple access wireless networksrdquo IEEEJournal on Selected Areas in Communications vol 17 no 8pp 1353ndash1368 1999

[43] J P John ldquoProviding MAC QoS for multimedia traffic in80211e based multi-hop ad hoc wireless networksrdquo ComputerNetworks vol 51 no 1 pp 153ndash176 2007

[44] S Lofty and N Padmavati ldquoA survey on mobility basedprotocols in WSNsrdquo in Proceedings of the InternationalConference Communication and Manufacturing pp 12ndash15Hangzhou China July 2014

[45] S Deng L Shen and J Li ldquoMobility-based clustering pro-tocol for wireless sensor networks with mobile nodesrdquo IETWireless Sensor Systems vol 1 no 1 pp 39ndash47 2011

[46] M Khalil A Khalid F U Khan and A Shabbir ldquoA review ofrouting protocol selection for wireless sensor networks insmart citiesrdquo in Proceedings of the 24th Asia-Pacific Confer-ence on Communications (APCC) Ningbo China November2014

[47] O Message Q Telemetry and T Mqtt Committee Specifi-cation Draft 02 April 2014

[48] M Collina M Bartolucci A Vanelli-coralli andG E Corazza ldquoInternet of things application layer protocolanalysis over error and delay prone linksrdquo in Proceedings ofthe 7th Advanced Satellite Multimedia Systems Conference andthe 13th Signal Processing for Space CommunicationsWorkshop Livorno Italy September 2014

[49] A Talaminos-barroso M A Estudillo-valderrama L M RoaJ Reina-tosina and F Ortega-ruiz ldquoA machine-to-machineprotocol benchmark for e health applicationsmdashuse case re-spiratory rehabilitationrdquo Computer Methods and Programs inBiomedicine vol 129 pp 1ndash11 2016

[50] V Goyal O Pandey A Sahai and B Waters ldquoAttribute-based encryption for fine-grained access control of encrypteddatardquo in Proceedings of the 13th ACM Conference on Com-puter and Communications Security pp 89ndash98 AlexandriaVA USA October 2006

[51] M Singh M A Rajan V L Shivraj and P BalamuralidharldquoSecure MQTT for internet of things (IoT)rdquo in Proceedings ofthe Fifth International Conference on Communication Systemsand Network Technologies Gwalior India April 2015

[52] A Stanford-Clark and H L Truong MQTT for Sensor Net-works (MQTT-SN) Protocol Specification httpmqttorgnewwp-contentuploads200906MQTT-SN_spec_v12pdf 2013

[53] RFC ldquoExtensible messaging and presence protocol (XMPP)rdquoin Proceedings of the IETF pp 1ndash9 San Diego CA USAAugust 2005

[54] S Vinoski and I Technologies ldquoQueuing protocolrdquo IEEEInternet Computing vol 10 no 6 pp 87ndash89 2006

[55] J L Fernandes I C Lopes J J P C Rodrigues and S UllahldquoPerformance evaluation of restful web services and AMQPprotocolrdquo in Proceedings of the Fifth International Conferenceon Ubiquitous and Future Networks (ICUFN) pp 2ndash7 DaNang Vietnam July 2013

[56] M Asim ldquoA survey on application layer protocols for internetof things (IoT)rdquo International Journal of Advanced Researchin Computer Science vol 8 no 3 pp 996ndash1000 2017

[57] K An A Gokhale D Schmidt S Tambe P Pazandak andG Pardo-castellote ldquoContent-based filtering discovery pro-tocol (CFDP) scalable and efficient OMG DDS discoveryprotocolrdquo in Proceedings of the 8th ACM InternationalConference on Distributed Event-Based Systems New YorkNY USA July 2014

[58] E Web What is a web service pp 52ndash57 2010[59] M Laine RESTful Web Services for the Internet of ings

pp 2ndash4 2012 httpmediaTkk[60] IETF -e WebSocket protocol pp 1ndash71 2011[61] V Pimentel ldquoCommunicating and displaying real-time data

with websocketrdquo IEEE Internet Computing vol 16 no 4pp 45ndash53 2012

[62] Wikipedia 2019 httpsenwikipediaorgwikiStreaming_Text_Oriented_Messaging_Protocol2019

[63] K Raghavendra and S Nireshwalya ldquoApplication layer se-curity issues and its solutionsrdquo IJCSET vol 2 no 6pp 1266ndash1269 2012

[64] Y Xue B Ramamurthy andM C Vuran ldquoSDRCS a service-differentiated real-time communication scheme for eventsensing in wireless sensor networksrdquo Computer Networksvol 55 no 15 pp 3287ndash3302 2011

[65] C Sonmez S Isik M Y Donmez O D Incel and C ErsoyldquoSUIT a cross layer image transport protocol with fuzzy logicbased congestion control for wireless multimedia sensornetworksrdquo in Proceedings of the 5th International Conferenceon New Technologies Mobility and Security (NTMS) pp 1ndash6IEEE Istanbul Turkey May 2012

[66] T Melodia and I Akyildiz ldquoCross-layer QoS-aware com-munication for ultra wide band wireless multimedia sensornetworksrdquo IEEE Journal on Selected Areas in Communica-tions vol 28 no 5 pp 653ndash663 2010

[67] Q Wang and M A Abu-Rgheff ldquoCross-layer signalling fornext-generation wireless systemsrdquo in Proceedings of the 2003IEEE Wireless Communications and Networking 2003WCNC 2003 vol 2 pp 1084ndash1089 IEEE New Orleans LAUSA March 2003

[68] P Liu Z Tao S Narayanan T Korakis and S PanwarldquoCoopMAC a cooperative MAC for wireless LANsrdquo IEEE

32 Journal of Computer Networks and Communications

Journal on Selected Areas in Communications vol 25 no 2pp 340ndash354 2007

[69] H Shan H T Cheng and W Zhuang ldquoCross-layer co-operative MAC protocol in distributed wireless networksrdquoIEEE Transactions onWireless Communications vol 10 no 8pp 2603ndash2615 2011

[70] F Liu T Korakis Z Tao and S Panwar ldquoAMAC-PHY cross-layer protocol for wireless Ad-Hoc networksrdquo in Proceedingsof the 2008 IEEE Wireless Communications and NetworkingConference pp 1792ndash1797 IEEE Las Vegas NV USA March2008

[71] L Jacob and H R Shamna ldquoEfficient cooperative MAC androuting in wireless networksrdquo Transactions on Networks andCommunications vol 3 no 5 2015

[72] S Rao and K Shama ldquoCross layer protocols for multimediatransmission in wireless networksrdquo International Journal ofComputer Science amp Engineering Survey vol 3 no 3pp 15ndash28 2012

[73] L Han S Park S Kang and H Peter ldquoAn adaptive cross-layer FEC mechanism for video transmission over 80211WLANsrdquo in Proceedings of the 2009 International Conferenceon Internet pp 209ndash215 Las Vegas NV USA December2009

[74] N Li J Martınez and V H Dıaz ldquo-e balanced cross-layerdesign routing algorithm in wireless sensor networks usingfuzzy logicrdquo Sensors vol 15 no 8 pp 19541ndash19559 2015

[75] G Sun J Qi Z Zang and Q Xu ldquoA reliable multipathrouting algorithm with related congestion control scheme inwireless multimedia sensor networksrdquo in Proceedings of the2011 3rd International Conference on Computer Research andDevelopment vol 4 pp 229ndash233 IEEE Shanghai ChinaMarch 2011

[76] M Li Y Jing and C Li ldquoA robust and efficient cross-layeroptimal design in wireless sensor networksrdquoWireless PersonalCommunications vol 72 no 4 pp 1889ndash1902 2013

[77] J Yan M Zhou and Z Ding ldquoRecent advances in energy-efficient routing protocols for wireless sensor networks areviewrdquo IEEE Access vol 4 pp 5673ndash5686 2016

[78] A Ahmad S Ahmad M H Rehmani and N U Hassan ldquoAsurvey on radio resource allocation in cognitive radio sensornetworksrdquo IEEE Communications Surveys amp Tutorials vol 17no 2 pp 888ndash917 2015

Journal of Computer Networks and Communications 33

International Journal of

AerospaceEngineeringHindawiwwwhindawicom Volume 2018

RoboticsJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Active and Passive Electronic Components

VLSI Design

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Shock and Vibration

Hindawiwwwhindawicom Volume 2018

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawiwwwhindawicom

Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Control Scienceand Engineering

Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Journal ofEngineeringVolume 2018

SensorsJournal of

Hindawiwwwhindawicom Volume 2018

International Journal of

RotatingMachinery

Hindawiwwwhindawicom Volume 2018

Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Navigation and Observation

International Journal of

Hindawi

wwwhindawicom Volume 2018

Advances in

Multimedia

Submit your manuscripts atwwwhindawicom

Page 8: ReviewArticle - downloads.hindawi.com · ReviewArticle QoSStrategiesforWirelessMultimediaSensorNetworksinthe ContextofIoTattheMACLayer,ApplicationLayer,and Cross-LayerAlgorithms MuwongeSsajjabbiBernard

chooses a schedule for itself rendering central sched-uling and synchronization irrelevant So in the absenceof traffic in the locality the node simply does periodiclistening and sleeping and to eliminate collisionstransmissions are distributively synchronized using areservation-based model in case there is a sender totransmit data R-MAC differs from S-MAC in that ittransmits data to a number of nodes using the sleepmode [24] -ere are 3 stages in this protocol (i) la-tency detection (ii) period announcement and (iii)periodic operation of which (i) and (ii) are for nodesynchronization and (iii) is for listensleep procedures

(i) Latency detection phase here all nodes broadcast aNeighbor Discovery control packet (ND) at an ar-bitrarily chosen time since they are powered onWhen the ND is received from the neighbor nodethe receiver notes its arrival time and chooses arandom time to send an acknowledgment (ACK-ND)mdashthe same as the receivedND in sizemdashinwhichthe duration from ND arrival to ACK-ND trans-mission I2 is specified and the time from ND packettransmission to ACK-ND arrival I1 is computed

(a) Propagation latency between the 2 nodesL (I1 minus I2)2

(b) -us propagation latency can be defined as ldquotheinterval from the time the first node sends the firstbit of a packet to the time the second node receivesthe last bit of the packetrdquo [24]

(ii) Period announcement phase here nodes choosetheir own random listensleep time and start timeand broadcast these and when packets are re-ceived the node turns the received schedule intoits own All nodes will record their neighborsrsquoschedules in relation to theirs -ese two phasesmake a number of rounds to ensure all nodes areinformed about their neighboring nodes

(iii) Periodic operation phase here there is datatransmission [26] and nodes intermittently go intothe sleep mode and wake up One period is a listensleep cycle and nodes possess similar periods Twoparts make a period T0 listen window TL and sleepwindow TS as shown in the following equation

T0 TL + TS (2)

In this protocol the nodes make use of control packetsfor communication-ese include REV (reservation packet)ACKREV (acknowledgment packet for REV) and DATAACK-DATA (acknowledgment packet for data) messageexchanges all of which are similar in size

Advantages

(i) It is fair and efficient in terms of energy use(ii) It eliminates collision of data packets

216 AR-MAC (Adaptive-Reliable Media Access Control)-is is a TDMA-based MAC protocol for wireless body areanetworks (WBANs) designed to reduce energy consumption

as proposed by Rahim et al [27] -e protocol allocates aguaranteed time slot (GTS) to all nodes for transmissiondepending on their needs It utilizes periodic sleep andwakeup to minimize overhearing and idle listening based onthe needs of the node It has a central node (CN) having bigbatteries with higher computational power and single ordouble transceivers For double transceivers the sum of thetime frame TFrame is assigned for node transmission But inAR-MAC a single transceiver CN is considered with TFramehaving 3 subdivisions contention-free period (CFP) forsensor connection contention access period (CAP) foremergencies and time TMS to communicate with themonitoring station (MS)

Channel selection the CN first scans for redundantchannels from which it chooses one for transmissionand broadcasts its address and other information to thenodes -e destination nodes search for channels byscanning the radio frequency (RF) channels If free itswitches to the next and if busy it will wait for time TCPlistening to packets and again switch to the subsequentchannel if it does not get the channel packets When thenode finally succeeds in getting a channel packettransmission commences and an ACK packet is sent tothe CNTime slot assignment a Time Slot Request (TSR) packetis sent to the CN by the sensor node after it has selectedthe RF channel -e TSR packet has data rate and timeslot info for the node Static-size time slots with staticguard-band time TGB are proposed in [28] -isprotocol employs the adaptive algorithm of time slot(TS) and GB time and the CN allocates TS whilesending Time Slot Request Reply (TSRR) based on thenode traffic pattern -e size of the time slots variesdepending on the node requirements Based on thetransmissionmodel the data packet transmission ACKpacket reception and some amount of delay can betolerated by allocated TS To prevent interference be-tween adjacent time slots resulting from the node andCN clock drift TGB is inserted between them and itsvalue is given as follows

TGB1

FxTS1100

TGBn

FxTSn

100

TGBnn+1

F

100X12

TSn + TSn+11113858 1113859

(3)

where FGB factor depending on the mean driftvalue Nodes go to sleep after slots have been suc-cessfully assigned and wake up to transmit in assignedslots thus saving energy wasted in idle listening sinceeven allocated slots are collision freeSynchronization for periodic synchronization TDMAneeds additional energy [29] It needs a lot of energy tosynchronize nodes after several cycles To reduce en-ergy wastage and control collisions AR-MAC employs

8 Journal of Computer Networks and Communications

a new synchronization scheme When a data packetarrives the current and expected packet arrival timeswith allowable delay (D) are compared by the CN and adrift value (DV) is computed based on the differenceΔT -e DV is sent to the node in the ACK packet forchanging the slot in subsequent transmissions thoughthe value is based on allowable delay and F In caseΔTgtD the CN will send the DV in the SYNC-ACKpacket for subsequent SYNC to nodes or else a mereACK packet is sent by the CN for the received packetFor any forthcoming data transmissions the node willchange its wake-up time plan based on the DV Withsuch a sync approach a sleeping node will not lose syncfor all cycles N Acceptable delay (D) is given as

D min TS1 TSn( 1113857XF

100 (4)

where F is the guard-band factor

ΔT expected arrival time minus current arrival time

DV 0 if |ΔT|ltD

ΔT if |ΔT|gtD1113896

(5)

Frame format the AR-MAC protocol employs twoforms of packets namely (i) data packets in which anode transmits in its assigned time slot and employsCAP for emergency data and (ii) control packets thatinclude channel packet Time Slot Request (TSR)packet Time Slot Request Reply (TSRR) packet Syn-chronization-Acknowledgment (SYNC-ACK) packetData Request (DR) packet and Acknowledgment(ACK) packet [27]Energy consumption if N is the number of cyclesenergy consumption for these is measured as follows

ETotal 1113944N

k1ESleepk

+ 1113944N

k1EActivek

ESleep TSleep x ISleep x V

TSleep Tframe minus Tactive

(6)

where ISleep current from the voltage source V whenthe node is asleep If Esw is the switching energy Etransthe transmission energy Erec the receiving energy andETout the time-out energy then

EActive 2Esw + Etrans + Erec + ETout (7)

Advantages

(i) It reduces energy consumption(ii) It minimizes overhearing and idle listening based

on the needs of the node via periodic sleep andwakeup

217 ER-MAC It is a combination of TDMA and CSMAmethods making it flexible and adaptable to topological andtraffic changes and hence applicable in rapid response

situations inWSNs Collision-free slots can be planned usingTDMA [30] ER-MAC is designed in such a way that thefollowing are met

(a) It has a higher ratio of packet delivery and low delayunder lower energy consumption levels out-performing Z-MAC

(b) It keeps two priority queues in order to distinguishhigh from least priority packets

(c) It permits contention in TDMA slots so as to handleenormous traffic volumes

(d) It has a harmonized slot shape and the nodes areable to locally change their plans enabling themconnect to or exit the network with easeProtocol design it discovers the topology in such away that it uses the simple flooding mechanism tobuild the tree as initiated by the base station (BS) -etopology discovery aims at neighbor discovery andchange tracking in addition to setting up a routingtree A TOPOLOGY_DISCOVERY message com-prising hop_count new_parent_id and old_-parent_id is generated by the BS A node broadcasts itto discover if it has any potential children and aresponse to the parent In case of the desire to changethe parent the former parent is notified For all nodesthe hop count to the BS parent ID children and one-hop neighbor are registered with the BS at this stageTDMA slot assignment when a BS sends a SYNCmessage it switches to the TDMAmode which whenreceived by the child uses the broadcast slot for itschildrenrsquos synchronization Neighboring nodescannot share a slot since slots are assigned andschedules are exchanged accordingly Slot assign-ment begins with a leaf node in a bottom-up modesuch that the flow of messages to the BS may besmooth Before getting the child-node schedules anonleaf node will not transmit its data via a unicastslot descendantrsquos data via many unicast slots andchild synchronization via a broadcast slot Childnodes send SCHEDULE_NOTIFICATION mes-sages to the BS to end this phase

Local time synchronization is done via parent-childbroadcast sync since they both need to operate the same clockso that when one is transmitting the other is ready to receivePacket prioritization happens in such a way that a highpriority queue is first emptied before a lower priority queue isallowed to transmit In case a queue is filled up the shortestslack packet is dropped and then fairness over source isconsidered as queue modification is done to enable the BSbalance info from different nodes Energy conservation is bysender turning off the radio in absence of any info to transmitand the receiver will switch to the sleep mode in case nopackets are received after a certain timeout [30]

Advantages

(i) It is flexible and adaptable to topological and trafficchanges

Journal of Computer Networks and Communications 9

(ii) It is applicable in rapid response situations inWSNs

(iii) It has a higher ratio of packet delivery and lowdelay under lower energy consumption levels

218 RI-MAC (Receiver-Initiated MAC) -is is an asyn-chronous duty-cycle protocol that tries to reduce timeoccupied by the communicating nodes on the medium toagree on time for data exchange-e sending node waits forthe receiving node to signify the beginning of the com-munication session through a short beacon frame that itsends -e channel is only occupied by the beacon and datatransmission reducing channel occupancy and thus en-abling data exchange for other nodes too -is implies animprovement in channel utilization that subsequently leadsto increment in throughput packet delivery ratio andpower efficiency over a wide-ranging traffic load [31] Beinga receiver-initiated model it ensures substantial reductionin overhearing and lowers the probability of colliding andthe cost of recovery compared to B-MAC [23] or X-MAC[32] RI-MAC is further capable of using a beacon withACK and Ready-to-Receive packets though it has aproblem if a sending node with Ready packets needs to havethe radio on till a ready-to-receive node is awake [13] whichthen sends it a beacon and transmission begins immediatelyand later is acknowledged by another beacon Nodes in-termittently wake up to listen to any possible frames meantfor them depending on their schedules A node announcesits readiness to receive a frame through a broadcast beaconon waking up when it switches on the radio When a dataframe has been received the receiver will spend some moretime known as ldquodwell timerdquo in the active state so as toreceive queued packets which is the time in RI-MAC thatdepends on the number of contending senders If there areseveral contending senders that might not have beenpredicted there is a challenge in trying to reduce the re-ceiverrsquos active time to optimize power efficiency and reducethe cost of collision detection and lost data recovery even inthe case of hidden senders For this protocol this can beaddressed by a beacon frame from the receiving node forcoordinating data frame transmissions of contendingsending nodes-e protocol also beats B-MAC and X-MACin minimizing the cost of collision detection and datarecovery through coordination of the receiversrsquo dataframes -is is because the receiving node is aware of themaximum delay prior to the arrival of the frame-e abilityto broadcast is supported by RI-MAC when it sends data asunicast to each senderrsquos neighboring node or continuouslytransmits the frame in a back-to-back mode for as long asthe sleep interval [31]

Advantages

(i) It improves channel utilization leading to in-crement in throughput packet delivery ratio andpower efficiency over a wide-ranging traffic load

(ii) It ensures substantial reduction in overhearing andlowers the probability of colliding and the cost ofrecovery

219 SRI-MAC (Synchronous Receiver-Initiated MAC)-is is a synchronous duty-cycle protocol that implementsthe method of receiver-initiated data transmission with anaim of enhancing network lifetime through energy wastageavoidance via collision overhearing and idle listening To dothis and also minimize duty cycle it uses a sequence ofadaptive beacons and RTSCTS packets In this protocolsending nodes do not go to sleep till they have got CTSpackets from receiving nodes to begin transmitting exceptfor nonsending nodes that return into the sleep mode rightaway as if the channel was idle [33] Since sending nodeshave separate schedules it reduces the contention proba-bility leading to more energy conservation A time frame isdivided into the information period the allocation periodand the communication period which are discussed asfollows

Information period at frame initiation a beacon is sentby the receiving node declaring its return from sleepand readiness to accept data packets in case the channelis not busy -e beacon is a small packet alerting otherlistening nodes of the transmitterrsquos readiness to get arequest for channel It consists of the receiverrsquos ID andthe duration allocation period (DAP) that relies onnumber of the receiving nodersquos neighbors [13]Allocation period here the receiving node stays awake forsome little time after broadcasting its beacon to confirm ifthere is any node having anRTS packet to transmitWhenthe beacon has been received an RTS packet (with 3fieldsmdashsender ID receiver ID and data size) is sent by thesender prompting the receiver to send a CTS for SYNC asthere is neighbor information exchangeCommunication period SRI-MAC is capable of or-ganizing and scheduling transmissions by means ofsensor node IDs -e receiver assigns order andtransmitting time for every sender such that they switchoff their radios till their time to transmit is dueHowever the receiving node remains in the wake-upmode to get all sending nodersquos data SRI-MAC canenable sensors have lengthy sleeping times duringtransmission such that energy can be saved as long as itis neither sending nor receiving data -e protocolrsquoscommunication period is premised on the TDMAtechnique which conserves energy since the radiorsquosduty cycle is lowered and collision is eliminated [33]

Advantages

(i) Since it employs the TDMA principle it eliminatescollisions while maintaining energy conservationtoo

(ii) Noncommunicating nodes switch off their radioand get into the sleep mode which eliminatesoverhearing

(iii) Since receiving nodes broadcast SYNC signals toall nodes in the form of CTS packets overemittingis avoided

(iv) It decreases duty cycling as sleep time increasesand lowers idle listening

10 Journal of Computer Networks and Communications

2110 H-MAC -is protocol is built on the power savingmechanism (PSM) of IEEE 80211 and slotted Aloha pro-tocols to enhance performance using multiple slots Timehere comprises large frames each having an active and asleeping part [34] If any node has got data to send it will askfor slots from the receiver node during active time and latertransmit the packets in the prenegotiated slots Nodes gointo the sleep mode during sleep-time slots if they do nothave data to transmit

Latency in IEEE 80211 if N is the number of hops nthe value for the current hop tcsn the back-off time andttx the transmission delay the total latency is calculatedusing the following equation

D(N) 1113944N

n1tcsn + ttx1113872 1113873 (8)

and the average latency is given by the followingequation

E[D(N)] N tcsn + ttx1113872 1113873 (9)

-roughput in H-MAC is given by the followingequation

Th npnm

tact + Cts (10)

where np is the number of transmitted packets nm is themaximum number of nodes connected in a time frameC is the number of slots with the same length tact is thelistening time and ts is the sleep delayAdvantages

(i) It enhances performance using multiple slots

2111 Z-MAC (ZebraMAC) -is rides on the strengths andoffsets weaknesses of TDMA and CSMA It yields very goodchannel usage and low delay under low contention and lowerscollision among two-hop neighbors at a reduced cost Itsperformance is strong via error synchronization failure inassigning slots and channel conditions that change with timeIn the set-up phase the protocol operates through certainsteps including neighbor discovery slot assignment localframe exchange and global time synchronization whichunless the topology of the network changes significantly donot run again When the data are being transmitted there isimprovement in network throughput and energy efficiencywhich covers up for the initial upfront operational costs [35]

Disadvantages

(i) Initial slot assignment is quite a cumbersome taskhere since it is done at the beginning yet this maynot be at the same time in reality It further as-sumes static links until setup happens again whichtoo is impractical

(ii) It also suffers from the hidden-terminal problemsince slot owners have a small contention windowand priority for slots allocated to them due to theECN messages

(iii) Synchronizing time also poses a challenge in thisprotocol due to a large clock drift between nodes

(iv) Source and destination nodes too are not wellcoordinated in Z-MAC [36]

2112 Q-MAC (QoS-Aware Media Access Control) -isprotocol offers high QoS with improved energy efficiencyand accesses the wireless channel using MACAW as theunderlying protocol [37] Traffic from different nodes isprioritized for QoS satisfaction depending on how criticalthe data are It mainly has two scheduling algorithms inWSNs (i) intranode and (ii) internode scheduling -eintranode scheduling scheme adopts a first-in first-out-(FIFO-) based queuing algorithm based on application andMAC layer abstraction for data packet classification In-ternode scheduling reduces idle listening and collision on thechannel for improved energy consumption

Advantages

(i) It gives a high QoS with improved energy efficiency

2113 WiseMAC (Wireless Sensor MAC) WiseMAC isbased on the Aloha protocol and employs preamble sam-pling in achieving low-power transmissions in infrastructuresensor networks by occasionally sampling the channel tocheck for activity [3238] -e protocol employs the samemethod as B-MAC except that the sender shortens theextended preamblersquos length by scheduling its transmissionsaccordingly after learning the receiverrsquos awake schedules-is is done by the receiver node putting its subsequentawake time on the ACK packet such that when the senderneeds to send to it again it starts the preamble shortly beforeit wakes up so as to avoid energy wastage in transmitting thepreamble [32] -e data frame is repeated instead of theextended preamble for less traffic transmissions that have alengthier preamble than the data frame in WiseMAC Afterprocessing the frame a node that is not the target recipientwill go back into the sleep mode while the recipient staysawake till the session ends thereby sending an ACK packetFor an idle channel the power consumption is rather lowwith this method Nevertheless there is a problem of longwake-up preamble limiting throughput and causing anenormous power consumption overhead in reception [38]All nodes overhearing a transmission bear the overheadtogether with the recipient node -e protocol can addressissues concerning low-power transmission but has nomechanism of nodes adapting to varying traffic patterns[32] To compensate for drift between clock at the accesspoint and on the sensor node there is a need to compute thewake-up preamble duration -e drift varies directly withtime since previous resynchronization and is given by thefollowing equation

TP min 4θl TW( 1113857 (11)

where θ is the time-base quartz frequency tolerance and l isthe interval between two transmissions Increase in trafficvolume naturally reduces overhearing

Journal of Computer Networks and Communications 11

Radiomodel for protocols with lower power transitiondelays between transceiver states and their powerconsumption are modeled Some of the states includeDOZE where the transceiver neither transmits norreceives but is prepared to power on to the receivetransmit state very fast RXwhere the transceiver listensto the channel receives data or demodulates data fromthe noisyidle channel and TX which is the trans-mission state by the transceiver Let TS be the set-uptime needed to turn the transceiver from the DOZE tothe RX or TX state TT the turn-around time needed toswitch between RX and TX states and B the trans-ceiverrsquos bit rate PZ PR and PT are the values of powerconsumed in the DOZE RX and TX statesrespectively

1113954PR PR minus PZ

1113954PT PT minus PZ(12)

where 1113954PR and 1113954PT are the changes in power as a result ofbeing in RX and TX states respectivelyTraffic model let N be the number of sensor nodesconnected to an access point (AP) and λ be the global rateat which the downlink Poisson traffic arrives at the APfrom the fixed network Traffic toward each sensor node isthe same and is given by λN Average packet interarrivaltime Lwith which data packets arrive at a node is given byL Nλ If TC and TD are the control packets and datapackets respectively assuming a small traffic volume theglobal interarrival 1λ far exceeds the summation ofdurations for the data packet and control packet and theturn-around time Mathematically it is given as

1λ≫TD + TC + TT (13)

-e average power consumed by WiseMAC andtransmission delay incurred are given by equations (14)and (15) [38]

PW PZ +1113954PR TS +(1B)( 1113857

TW+

1113954PR X + TD + TT( 1113857

L

+ 1113954PR(N minus 1)Y

L

(14)

where

X 2θL 1 minus eminus TD4θL( )1113874 1113875

Y T2D + 12TDθL

2TW1 minus e

minus TW4θL( )1113874 1113875

DW TD +TW

21 minus e

minus TW4θL( )1113874 1113875

+ 2θL 2 minus eminus TD4θL( ) minus e

minus TW4θL( )1113874 1113875

(15)

where PW and DW are the power consumed byWiseMAC and the transmission delay for WiseMACrespectivelyAdvantages

(i) It achieves low-power transmissions in in-frastructure sensor networks through preamblesampling while checking for activity

(ii) It reduces overhearing due to increased trafficvolume

Disadvantages

(i) It has a problem of long wake-up preamble limitingthroughput and causing enormous power con-sumption overhead

2114 X-MAC It is a simple minimal transmission powerMAC protocol that uses a short preamble with decouplingfor the sending and receiving node sleep plans inWSNs-isleads to a significant reduction in energy usage at sendingand receiving nodes and decrease in per-hop latency andgives flexibility in adapting to bursty and periodic sensordata sources [32]

Protocol design X-MACrsquos design is aimed at increasingenergy efficiency reducing data latency and increasingthroughput -is enables application in different kindsof packetized and bit stream digitized radios as well assimplified low-overhead distributed application -edesign is as such in order to address the low-powerlistening problems including overhearing excessivepreamble and incompatibility with packetizing radiosas in [32]Asynchronous duty cycling a node with information totransmit will send the packets after sending a preamblewhereas the remaining nodes will preserve their sleepschedules that are not synchronized -e receiver willsample the link on waking up and stays awake if it detectsa preamble and if it is the target it will return to sleep if itis not the target after getting the whole preambleEmbedding the target ID in the preamble to avoidoverhearing the lengthy preamble is divided into manyshortened packets embedded with the receiver nodersquosID so as to eliminate overhearing One lengthy pre-amble is constituted by the shortened preamble packetstream Any node on receiving the short preamble afterwaking up will examine the target ID on the packet Itgoes back into the sleepmode right away resuming dutycycling if it is not the targeted recipient or stay awakefor successive packets if it is the one -is approachchecks energy wastage due to network density It canalso be used on several types of radiosReducing excessive preamble using strobing it ispossible to enable low-power transmission and alsosave energy by reducing time for preamble trans-mission through the use of enhanced preambles inaddition to preamble sampling X-MAC does not sendan endless preamble packet stream but puts slight

12 Journal of Computer Networks and Communications

pauses in between packets to allow the sending node topause and listen to the channel and allow the receivingnode to resend an early ACK packet during that pause-is makes the sending node begin sending datapackets and the receiving node cut excessive preamblesand thus decrease per-hop latency and energy wastageon unnecessary waiting and transmissionPacketizing radios low-power listening is limited incapability for supporting packetizing radios butX-MACrsquos short strobed preambles can support all typesof digital radiosAdaptation to traffic load while a number of WSNapplications are reliable in traffic production they stillmust adapt to changing traffic Nodes are going to haveseparate sleep schedules depending on traffic loadConsidering systems that have time-fluctuating trafficall predetermined static schedules are suboptimalOptimality Buettner et al [32] model the expectationof energy as follows

Es (preamble energy + energy perACK listen)

lowast(expected preamble minus listen iterations)

+(energy to send parket)

Es PTxSp + PRxSal1113872 1113873 Rl + Rs( 1113857

Rl minus Sp+ SdPTx

(16)

where Es expected energy to send a packet PTx andPRx are the transmission and receiver power re-spectively Sp Sal and Sd are the duration of thesenderrsquos preamble ACK listen and data transmissionperiods respectively and Rl and Rs are the receiverrsquoslisten and sleep periods respectively Expected energyto receive a packet is given as

Er (listen cycle energy + sleep cycle energy)

lowast(expected iterations for preamble arrival)

+(energy to sendACK) +(energy to receive packet)

Er PsRs + PRxRl( 1113857

1 minus 1 minus Pd(t)( 1113857 Rl + Rs( 1113857+ RaPTx + RdPRx

(17)

And the one-hop expected latency is given as

Lat (preamble duration + ACK listen)

lowast(expected number of iterations)

+ duration to send packet

Lat Sp + Sal1113872 1113873 Rl + Rs( 1113857

Rl minus Sp

(18)

Advantages

(i) It increases energy efficiency(ii) It reduces data latency(iii) It increases throughput

2115 A-MAC (Advanced MAC) A-MAC is a TDMA-based protocol whose design is aimed at enhancing thelifetime of the network for low rate and reliability of datatransmission [39] To assign time slots it does not depend ona central controller like most protocols but it gathers in-formation from its neighbors so as to allocate the slotsthrough a distributed algorithm Energy is controlled byscheduling power when there is not a single event Its designcomprises many frames with each having many time slotswhich at the beginning have got a beacon transmitted tosynchronize and exchange information with neighborsNodes to participate in the subsequent session are selectedby the controlled node so they do not go into the sleep modewith others [13] -is protocol operates under four time slotdivisions namely initial wait discover and active states Anode in the initial state begins by listening to the medium forthe beacon packet from other nodes for network synchro-nization which when received makes it change the timerthrough subtraction of beacon transmission time from re-ception time To avoid synchronization problems due todrift and allow for continuity the node has to get thestrongest beacon signal It then enters the wait state afterchoosing a number of waiting frames so as to reduce theprobability that nodes get into the discover state simulta-neously In case the synchronization beacon is lost while stillin the wait state it will return to the initial state for anotherone On expiry of the waiting counter it enters the discoverstate and collects information from neighbors by listening totheir beacon messages On successful selection of a time slotit enters the active state where it endlessly transmits beaconsat the initiation of the slot A node goes to sleep when abeacon has been transmitted and there are no more data fortransmission or in case the neighborrsquos beacon received in-dicates no data packets are coming in [39]

Advantages

(i) It can enable multicasting(ii) It enhances the lifetime of the network(iii) It offers reliable data transmission

2116 L-MAC (Lightweight MAC) Based on TDMA thisprotocol is aimed at minimizing the number of transceiverswitches adapting the nodal sleep interval to data trafficamount and making implementation less complex [40] Apacket is transmitted comprising a control and a data partControl packets have a fixed size and the time slot con-trollerrsquos ID -ey show the distance from the transmittingnode for simplified routing to the network gateway as well asaddressing target node and reporting data unit length andinternode synchronization Adjacent nodes are more in-terested in getting control packets from nearby nodes -enodersquos transceivers are switched off until the subsequenttime slot unless the node is also addressed or the message isomnicast If so it listens to the data unit which may not fillthe remaining time slot in totality-e transceivers on eitherside are turned off when the transfer of packets is successfulEnergy wastage through idle listening in uncontrolled timeslots is avoided via a short time-out interval Network nodes

Journal of Computer Networks and Communications 13

are able to communicate without colliding-is saves energyand increases the lifetime of the network A node in L-MACcan only transmit one message per frame [13] -e networkis set up in such a way that all nodes are not synchronized onbeing switched on-e gateway takes control of a timeslot soas to have them synchronized and its immediate neighborgets the control messages thereby synchronizing their timerswith the gatewayrsquos Neighboring nodes learn of multiplegatewaysrsquo time slots in their vicinity just after one frame-ejust synchronized nodes randomly choose time slots forcontrolling as long as they are not occupied -e nodes thuspreserve a table for the neighboring nodes which makes atimeslot reusable after a minimum of 3 hops and preventsmessage collision All nodes keep their slots before thebattery is down or learn of an impending collision All nodescan track their hop distances to a selected gateway node andthe information is communicated in control messages [40]

Advantages

(i) It minimizes the number of transceiver switches(ii) It adapts the nodal sleep interval to data traffic

amount(ii) It makes implementation less complex(iii) It offers improved network lifetime(iv) Energy wastage through idle listening in un-

controlled time slots is avoided via a short time-out interval

Disadvantages

(i) L-MAC is not very good for mobile SNs since slotsare calculated only once

(ii) It suffers idle listening because nodes keep listeningto slotsrsquo control parts so as to get data and enableother nodes to join the network all the time [13]

2117 C-MAC -is was designed for WSNs with 3 seg-ments namely (i) aggressive RTS having double channelcheck to assess the channel (ii) anycast to initialize the flowand (iii) convergent packet forwarding to stabilize the flowIt promises an improvement in energy efficiency lowerlatency and enhanced throughput while avoiding syn-chronization overhead though it operates only in low duty-cycling apps especially where there is less traffic But whenthere is more traffic the protocol employs anycast-basedpacket forwarding for waking the nodes up or finding aforwarder quickly and converges from route-suboptimal toroute-optimal unicast After this energy could be preservedby the nodes using a synchronized wake-up plan [13]

Advantages

(i) It improves energy efficiency(ii) It lowers latency(iii) It enhances throughput avoiding synchronization

overhead

Disadvantages

(i) It operates only in low duty-cycling applicationsespecially where there is less traffic

22 MAC Protocols for WMSNs

221 Diff-MAC -is is a CSMACA-based QoS-awarehybrid-priority-based MAC protocol for WMSNs It targetsenhancing the channel utilization while making use of ef-ficient service differentiation to do coordination for mediumaccess of each traffic class while giving fair quick datatransfer It offers QoS using the following features [14]

(i) It balances energy consumption and delay throughsensor node duty-cycling adaptation depending onthe dominant traffic class

(ii) It has a feature for fragmentation and messagepassing that breaks lengthy video frames intosmaller packets that are transmitted as a burst thusreducing retransmission costs if there are any MACfailures

(iii) -ere is fair data delivery amongst the nodes andtraffic classes respectively resulting from intranodeand intraqueue prioritization reducing unbearableperformance

(iv) -e number of collisions can be reduced and packetlatencies are kept within limits by adjusting the sizeof the congestion window in this protocol as per thetraffic needsAdvantages

(i) Diff-MAC is very suitable for WMSNs since itadapts quickly to changes in networkconditions

(ii) It is a fair protocol(iii) It is scalable

Disadvantages

(i) It is quite difficult to monitor the statistics ofthe network

(ii) Adapting dynamically to the network condi-tions is not an easy task

(iii) It exhibits some degree of latency in packetdelivery

-e implementation of Diff-MAC on Imote2 indicatesthat resource requirements can be met on the sensorhardware Results of extensive simulation runs showed thatDiff-MAC outperforms its competitors

222 Cluster-Based On-Demand Multichannel MAC Pro-tocol for WMSN (COM-MAC) -is is an energy-efficientprotocol in WMSNs with high throughput and reliable datatransmission It achieves energy efficiency by schedulingmultichannel media access to reduce channel contentionand remove collision idle listening and overhearing [41]

Design it has a three-phased operation with the (i)request phase (ii) scheduling phase and (iii) datatransmission phaseRequest phase here all nodes have two protocols eachfor sending requests to the cluster head (CH) ie thecontention-based and contention-free TDMAFDMA

14 Journal of Computer Networks and Communications

protocols -e contention-based protocol allocates achannel per node for sending the request message andfree channels are useable as control channels Forcontention-free protocols a control slot is allocatedwhen a network is deployed Here a slot per channel isallocated to a different node enabling requests to be sentwithout interfering with transmissions from othernodes -e contention-based protocol operates undertwo steps (i) Control channel assignment phase achannel is assigned per node for transmitting a request(REQ) message A channel is assigned to multiplenodes and nodes are evenly distributed to availablechannels for congestion control (ii) Request trans-mission phase nodes with data send a REQmessage viathe assigned control channel to notify the CH whichresponds with an ACK message -e contention-freeTDMAFDMA protocol too has two phases includingthe control slot assignment and request transmissionphases For unreliable channel conditions and lowtraffic load the contention-based protocol is fit for usebecause of less collisions and congestion whereas thecontention-free protocol is preferable in heavy trafficload and reliable channel applicationsScheduling phase from the above phase a schedule isgenerated by the CH for coordination and this enablesnodes to transmit and broadcast their data via controlchannels-e schedule for broadcasting only comprisesdata for nodes assigned to a channel -e schedule isrebroadcast for transmission reliability enhancementbut for energy efficiency the nodersquos transceiver isswitched off when all the schedule is completely de-livered till it is the time for transmitting again-e timeslot and data communication radio channel are allincluded in the schedule generated for each nodeData transmission phase on getting the schedule thesensor node sends its data as per the time slot and thechannel assigned -e time slot is divided into datatransmission andACK sections ACK supports link layererror control Using implicit selective repeat ARQ theCH sends an ACK message for all received packets Ifthere is no ACK the packet is assumed lost and hence aretransmission during the subsequent interval is done-is enhances reliability of transmissions though excessdelay is possible here which is also intolerable for real-time critical apps like video or audio -is gives rise to ahybrid MAC protocol to exploit the unused spectrumduring data transmission sessions [41]Advantages

(i) It achieves energy efficiency by scheduling mul-tichannel media access to reduce channel con-tention and remove collision idle listening andoverhearing

(ii) It maximizes network throughput by dynamicallyallocating time slots and channels for nodes byexecution of a traffic-adaptive QoS-aware sched-uling algorithm depending on needs for QoS andtraffic data information

(iii) It increases transmission reliability by in-corporating spectrum-aware ARQ to avoid spec-trum wastage

223 EQoSA -is protocol is developed for video andimage transfer across WMSNs in a hybrid mode withprovision of QoS Concerning the active sensor nodes andtheir traffic loads it employs dynamic session sizes andchanges bitmap-assisted (BMA) MACrsquos fixed session sizeNodes declare any data available for transmission duringcontention time -is forces the cluster head to make aschedule of how slots have been assigned It communicatesthis schedule to the nodes giving each node a given numberof slots per session and so enabling the busty traffic to beaccommodated [14] -e protocol does not perform wellwith DiffServe mechanisms but does well with schemessupporting busty traffic loads and strong cluster heads arenecessary in WMSNs that can assign and broadcast slots

224 QoS-Supported Energy-Efficient MAC (QE-MAC)Based on the IEEE 80211e standard this protocol ensuresfairness and has low latency and jitter and better efficiencyon energy usage than other QoS-aware MAC protocols forWMSNs It operates under a two-phased routine

-e first phase introduces the innovating prioritymechanism in IEEE 80211e QoS and several data types areassigned priorities Its coordination function is a combi-nation of distributed contention-based channel allocationand centralized polling-based channel access mechanisms

-e second phase preserves the nodesrsquo energy usingdynamic duty cycling -ere is an exchange of RTSCTSpackets to supplement CSMACA by muting nodes near thesenderreceiver during packet duration [42]

Advantages

(i) It ensures fairness(ii) It has low latency and jitter and better efficiency on

energy usage than other QoS-awareMAC protocolsfor WMSNs

225 Black-Burst (BB) Contention A contention-basedMAC scheme for enhancing instantaneous QoS access tocarrier-sense WSNs is proposed in [42] In this schemenodes contend for the channel until it is free Real-timepackets are given higher priority over non-real-time packetsand if a node is transmitting real-time packets it is prior-itized over other nodes [15] Voice and video are some of thereal-time apps in consideration here which periodicallyaccess the common radio channel when transmitting -eseneed constrained end-to-end delay of packets at the MAClayer -is protocol majorly does the following [42]

(i) Prioritization of real-time traffic(ii) Implementation of the round-robin mechanism in

real-time nodes(iii) Bounded access delays among real-time packets

Journal of Computer Networks and Communications 15

(iv) Implementation of real-time communication withvarying needs of bandwidth good for multimediatraffic

Amechanism for dynamically assigning packet prioritiesbased on their deadlines and the traversed hops is proposedin [43] -e ReAllocative Priority (ReAP) mechanism ismeant to provide QoS at the MAC layer to video traffic -isscheme introduces adaptive TXOP by modifying the TXOPdynamically based on the queue length [15] Other protocolsin this category include dynamic duty cycle and adaptivecontention window-based MAC protocol frame sharing(FRASH) MAC protocol and real-time independentchannels (RICH) MAC protocol

Advantages

(i) It can implement real-time communication withvarying needs of bandwidth

23 Challenges and Opportunities(a) One of the challenges in satellite transmission is

cross-talk leading to interference in different fre-quency bands It is necessary to increase the rise andfall times as well as protect sensitive nodes as wecircumvent floating ones if we are to regulate cross-talk as in satellite communication therefore Wefurther need to investigate the possibility of usingdifferential signaling -is constitutes an opportu-nity for further research

(b) Another challenge is that the hidden and exposedterminal problems are still noticeable despite the useof RTSCTS solutions in MAC since collisions arestill possible from transmissions of separate nodesAt the MAC layer ACKs might block efficientchannel usage by prohibiting exposed terminalsfrom reusing the channel and may be eliminatedleaving no proof of sender data packet receptionUsing a directional antenna is one of the proposedtechniques of high-speed WSNs like IEEE 80211-is constitutes an opportunity for further study intothese issues for better solutions though

(c) Another challenge is that since nodes in contention-based MAC protocols are mobile there are manycollisions by the packets and schedule in-consistencies in schedule-based protocols in the two-hop neighborhood information when nodes areentering or leaving [44] Deng et al [45] propose amobility-based clustering (MBC) protocol for mo-bile nodes to enhance the packet delivery rate -isprotocol in comparison with CBR has 25 chancesof decreasing collision-induced packet losses and50 chances in comparison with low-energyadaptive clustering hierarchy (LEACH)

(d) -e process of analyzing and evaluating routingperformance for WMSNs is quite complex becauseof the bursty multimedia traffic Because of the wayencoding scheme frames are structured as well as thechanges in between the scenes that occur naturally

the compressed video frames display some substantialburstiness on several time scales -ere will also befrequent route recalculation that is as well energy in-tensive for nodes frequently capturing and transmittingenergy -e solution here could as well possibly be theuse of the MBC protocol Route recalculation tooshould be dynamic to cater for the mobile nodes whichmight be in and out of the range at different instants oftime Recently the use of smart dumpsters for effectivewaste management in smart cities has also been pro-posed as a possible solution [46]

(e) For WSNs their routing protocols do not look atmultimedia applications which require enormousbandwidth and processing energy and should betransmitted in real time with utmost fidelity Extrachallenges in routing are imposed in WMSNs indeployment of heterogeneous sensors since therecould be audio video and still pictures in addition toscalar data during transmission Heterogeneous datahandled also have different QoS requirements andwith many different business needs different ser-vices with varying requirements give rise to a bigchallenge for the routing design -ere is a need todesign protocols that can support WMSNs such asEQoSA [14] for this purpose

Table 2 summarizes the reviewed MAC protocols forWSNs and the ones designed for suitability in WMSNs

3 QoS Strategies at the Application Layer

We discuss many protocols used to solve different needs ofcommunication between machines

31 ConstrainedApplication Protocol (CoAP) It is a requestresponse protocol designed by the Internet Engineering TaskForce (IETF) for constrained devices such as those havinginsufficient RAM or CPU and WPANs with low-powerconstraints resulting in poor transfer of packets andenormous overhead -e protocol was designed for ma-chine-to-machine (M2M) apps and system automation toreduce bandwidth requirements increase packet transferdecrease the overhead and make work simpler leading tolightweight implementation HTTP commands like GETPOST PUT and DELETE are used in a client-server ar-chitecture Using the publishsubscribe architecture CoAPcan support multiple users and hence yield better perfor-mance via multicasting enabling asynchronous messageexchange -is protocol can support both multicasting andunicasting via UDP and hence is a reliable and simpleprotocol -is is done via the messaging sublayer (for re-liability based on UDP) and the requestresponse sublayer(for communication and interaction) [16]

It uses the following messages

(i) Confirmable message this guarantees reliabilityusing ACK messages sent by the receiver node-ese could be synchronous or asynchronous ifthere is a need for more computational time

16 Journal of Computer Networks and Communications

(ii) Nonconfirmable message here no ACK messagesare sent back

(iii) Acknowledgment (ACK) message this confirmsreceipt of a confirmable message

(iv) Reset message this confirms receipt of an un-processed message in case any critical parts nec-essary for interpreting the message are missing

(v) Piggybacked response the receiver directly re-sponds while getting the ACK message

(vi) Separate response the reply by the receiver is dis-crete from the ACK message

CoAP has no in-built security features It is secured bythe datagram transport layer security (DTLS) that runs ontop of UDP though it was not designed for the IoT -ismakes its suitability debatable since DTLS offers no supportfor multicasting It also needs more packets for handshakingthereby increasing the traffic volume and computationalresources and so shortening the lifespan of mobile devicessince they use batteries [17]

Advantages

(i) It can support multiple users and hence yieldbetter performance via multicasting

Table 2 Comparison of MAC layer protocols used in QoS provisioning

Protocol Features QoS parameter Comments Access Priorityassignment Energy eff Designed for

S-MAC[22] Collision avoidance Energy latency Requires periodic

synchronization CSMA Hybrid Medium WSNs

B-MAC[23] Collision avoidance Energy delay

throughput Flexible interface CSMA Dynamic High WSNs

T-MAC[25] Collision avoidance Energy Complex nonscalable CSMA Dynamic High WSNs

C-MAC[13]

Dynamic channelallocation Energy Multichannel MAC CSMA Dynamic Medium WSNs

CU-MAC[24]

Multichannelmechanism

-roughputlatency High packet delivery ratio TDMA Dynamic mdash WSNs

A-MAC[39]

Per flow serviceguarantees Delay m-ary tree model

dynamic prioritiesTDMACSMA Dynamic Medium WSNs

ER-MAC[30]

Emergencyresponse

Energy delaydelivery ratio

Adapts to traffic andtopology changes

TDMACSMA Hybrid Higher than Z-

MAC WSNs

RI-MAC[31] Collision detection Delay packet

delivery ratioHigh packet delivery ratio

and throughput CSMA Dynamic Medium WSNs

SRI-MAC[33] Collision avoidance Energy delay Good data transfer and

energy consumption TDMA Dynamic High WSNs

H-MAC[34] Slotted sleep time -roughput

latencyData transmission only in

sleep timeCSMAAloha Dynamic Medium WSNs

Z-MAC[36] Collision avoidance Energy delay

throughputVery good channel usage

and error syncCSMATDMA Dynamic Medium WSNs

Q-MAC[37]

Intrainternodescheduling Energy High QoS improved

energy efficiency CSMA Hybrid High WSNs

X-MAC[32]

Overhearingavoidance

Energy delaythroughput

Reduces excessivepreamble by strobing CSMA Adaptive

static Medium WSNs

L-MAC[40] Collision avoidance Network

lifetime energyReduces the number oftransceiver switches TDMA Static Medium WSNs

WiseMAC[38]

No downlinkchannel collision Power delay Significantly low-power

consumptionAlohaCSMA Static Higher than

ZigbeeInfrastructure

WSNsAR-MAC[27] Collision avoidance Energy delay Uses data packets and

control packets TDMA Static Higher thanIEEE 802154 WBANs

BB-MAC[42]

Real-time trafficprioritization Delay Employs round robin in

real-time nodesCSMATDMA Dynamic High WMSNs

COM-MAC [41] Collision avoidance Packet delay

throughputBalances reliability and

retransmissionTDMAFDMA Dynamic High WMSNs

EQoSA[14]

Busty traffic loadaccommodation Energy Provides QoS support for

video and imagesTDMACDMA mdash High WMSNs

QE-MAC[42] Collision avoidance Energy jitter

latencyRuns central polling-like

channel access CSMA Dynamic High WMSNs

R-MAC[26] Collision avoidance Energy delay Reservation based mdash Static High Underwater

SNs

Journal of Computer Networks and Communications 17

(ii) It enables asynchronous message exchange(iii) It is a reliable and simple protocol

32 Message Queue Telemetry Transport (MQTT)Developed by IBM and later improved by OASIS with anintention of decreasing bandwidth MQTT is a simple open-source publishsubscribe protocol analogous to the client-server version It is appropriate for lightweight IoT appli-cations and M2M transmissions in constrained situationslike limited power computation capability and memory orbandwidth -e network bandwidth and the computationalresource usage are decreasing since clients are not updatedmaking publishsubscribe protocols outperform requestresponse protocols via IoT requirements [16] -is protocolcan run over TCPIP and so has the following advantages

(i) Delivering of packets with reliable guarantee(ii) Enabling multicasting(iii) Remote device session establishment(iv) Reduction of transport overhead and protocol ex-

change to minimize network traffic(v) Provision of the notification mechanism when

anomalies occur [47]

MQTTrsquos design targets economical use of bandwidthand battery so it has a major application in FacebookMessenger MQTT ensures QoS and reliability of messagedelivery by providing the following three QoS levels

(1) Fire and forget a message is sent once depending onthe networkrsquos best effort and there is no need forACK (lowest QoS level)

(2) Delivered at least once a message is sent at least onceand an ACK message is needed to avoid duplicates

(3) Delivered exactly once by a four-way handshake itensures a message is sent exactly one time (highestQoS level)-e architecture it has three key components (i)publishersmdashlightweight nodes sending data to thebroker before returning to sleep (ii) broker whichorganizes sensory data in topics and sends them tointerested subscribers and (iii) subscribersmdashIoTapps interested in data sent by sensors through thebroker as shown in Figure 3

MQTT is a simple lightweight protocol that is betterthan CoAP via low and high packet loss probability interms of throughput and exhibits lower latency in lowsampling rates [48] It supports a number of imple-mentations for embedded devices and mobile applicationsand offers web-socket support too Nevertheless it hassome shortcomings

(i) It is only employed in simple data types(ii) It exhibits high latency in high sampling rates(iii) It is unsuitable for real-time apps in such cases(iv) It does not provide automatic discovery andor

sufficient security at the protocol level [49]

33 Secure Message Queue Telemetry Transport (SMQTT)-is is an extension of MQTT that is based on lightweightattribute-based encryption using elliptic curve cryptogra-phy -e protocol has three entities as shown in Figure 4(i) publisher that publishes information under a certaintopic (ii) subscriber that obtains the same topic in-formation via a broker and (iii) public key generator(PKG) or broker that acts as the trusted third party -eprotocol has four phases namely (a) set-up phase (b)encrypt phase (c) publish phase and (d) decrypt phase

(a) Set-up phase publisher and subscriber provide aunique ID with attributes for registration with thePKG -is generates a secret master key set andpublic parameters that it publishes with set U A1A2 An For ciphertext policy-attribute-basedencryption (CP-ABE) the PKG gives a key set todevices with attributes [50]

(b) Encrypt phase based on the access tree the publishercreates an access policy An access tree is sent by thepublisher to the PKG for generating the key-gen-eration policy in the key policy- (KP-) based ABEalgorithm -e publishers know beforehand thetopics and subscribers to access as well as the rulesfor access Subscribers are issued with private keysfor all communication sessions Publishers in theCP-ABE algorithm come up with access trees andrules where the payload is encrypted by the pub-lisher who also issues more info needed for de-cryption and the rules With the help of public keycryptography data are encrypted giving rise to ci-phertext on the basis of KPCP-ABE and the receivergets SUBACK from the broker [51]

(c) Publish phase ciphertext is embedded in theSPublish command as payload and topic name isdefined in the header by the publisher -e brokerreceives the SPublish packet and replies withPUBACK prompting the publisher to respond withthe PUBREL packet -e broker then broadcasts totopic subscribers and deletes the data forwarding thePUBCOMP packet to the sender node

(d) Decrypt phase the encrypted information isdecrypted by the subscriber with the help of a secretkey Considering KP-ABE on verification that itsatisfies the access policy the subscriber invokes thePKG for a corresponding key which it sends afterverifying the request With CP-ABE subscribers usethe private key for decryption after authenticationthereby enabling offline interaction [51]Advantages

(i) -e information being encrypted makes it moresecure

34 MQTT for Sensor Networks (MQTT-SN) -is is a ver-sion of MQTT designed to operate under environments oflow bandwidth high link failures and short message lengthIt is a publishsubscribe (pubsub) protocol designed for

18 Journal of Computer Networks and Communications

UDP-based WSNs to visualize communication for low-power devices [5152] It comprises a publisher device actingas the MQTT-SN client to forward communications to agateway that also changes the MQTT-SN message to theMQTTmessage before delivering it to the broker that sendsit to MQTT-SN subscribers [51]

It has a gateway for changing protocols from MQTT toMQTT-SN (Figure 5) [52] -is protocol supports nodes in

the sleeping mode using an offline keep-alive procedure thatcan be used by battery-operated devices to get into thesleeping mode To support sleeping clients MQTT-SN has anew offline keep-alive procedure with which battery-oper-ated devices can go to the sleeping state in which theirmessages are buffered at the server until they are awake InMQTT-SN clients with no preconfigured gateway addresscan figure out the real address of their operational gateway

Publisher

Subscriber

Subscriber

Broker

Publisher

4G

Figure 3 Architecture of the MQTT protocol

Register by mentioning URI Register by mentioning URI

Access tree

Set-up

Publisher ReceiverPKG(broker)

Key management Key management

Encrypt data using public parametersand random key This is done using

CP-ABE Publish it as

Subscribe (topics)SUBACK

S-Publish (TopicEncrypt (data))S-Publish (TopicEncrypt (data))

Encryptand publish

PUBACK

PUBACK

PUBRELPUBRELPUBCOMP

PUBCOMP

Figure 4 SMQTT protocol

Journal of Computer Networks and Communications 19

within the network through a discovery procedure Simul-taneously we may have several gateways in a single wirelessnetwork that can be cooperative via load sharing

MQTT-SN architecture it is composed of MQTT-SN clients gateways and forwarders (Figure 5)rough the MQTT-SN gateway (GW) the MQTT-SN protocol is used by the MQTT-SN clients toconnect to the MQTT server Integration betweenthe MQTT-SN GW and the MQTT-SN server mightbe possible or not Translating between the MQTT-SN and the MQTT for a stand-alone GW is doneusing the MQTT-SN protocol If a GW is detachedfrom the network it can be accessed by the MQTT-SN clients through a forwarder is just encapsu-lates the received MQTT-SN frames that are on thewireless side before forwarding them to the GW un-altered It then decapsulates the received frames tooresending them to the client as sent by the GW [52]ere are 2 types of gateways (a) transparent gatewayand (b) aggregating gateway as shown in Figure 6

(i) Transparent gateway this launches and sustains adirect MQTT connection between the MQTT-SNclient and the MQTT server which is entirely re-served for end-to-end client-server messaging enumber of servers and that of MQTT-SN clientsconnected to the GW are the same is transparent

GW translates syntax between these protocols and iseasier to implement compared to the aggregatingGW But the MQTT server must support dierentsessions for all connected clients Some MQTTservers are congured to enable few parallel sessions

(ii) Aggregating gateway this maintains a single MQTTconnection to the server on behalf of all clients Allexchanged messages from the MQTT-SN client to theaggregating GW are directed to the GW which sortsthe information to forward to the server e aggre-gating GW decreases the number of MQTT sessionssimultaneously supported by the server in WSNs withmany battery-operated sensors and actuators (SAs)having limited storage and processing capabilities It ishowever more complex to implement compared to atransparent GWe general format of the MQTT-SNprotocol message has two or four message headeroctets and n octets of variable header that is optionaldepending on the message in question MQTT-SN has256 message types (MsgType) with those from0X1Eminus 0XFD 0X19 and 0XFF reserved for futureusage [51]

Advantages

(i) e protocol can operate under environmentsof low bandwidth high link failures and shortmessage length

MQTT-SN gateway

MQTT-SN

MQTT

MQTT-SN

MQTT-SN

MQTT-SN

MQTT-SN client

MQTT-SN clients

MQTT broker

MQTT-SN client

MQTT-SN client

MQTT-SN gateway

MQTT-SN forwarder

Figure 5 MQTT-SN architecture

20 Journal of Computer Networks and Communications

(ii) It is a pubsub protocol designed for UDP-basedWSNs to visualize communication for low-power devices

(iii) It can be optimal for low-cost battery-operatednodes having low processing power and lessstorage

(iv) It supports nodes in the sleeping mode using theobrvbarine keep-alive procedure used by battery-operated devices

35e Extensible Messaging and Presence Protocol (XMPP)e standardization of this protocol by the IETF made itpopular for usage over the Internet especially in instantmessaging and chatting platforms It is however susceptibleto spamming attacks and its failure to have worldwidesupport has made Google too withdraw its support of thestandard [17] e protocol is mainly useable in the client-server architecture but can also be applicable in client-clientXMPP communication [53] Among the application layerprotocols XMPP is the only one that supports both asyn-chronous (publishsubscribe) and synchronous (requestresponse) services while running over TCP but can still runover other protocols

Given the simplicity of its implementation and ease ofusability in XMPP instant messaging (IM) is suitable forXMPP and is implemented using the message stanzaltmessagegt ese stanzas are usable in one-to-one ormultiuser sessions or chatting or in error reporting andsending headlines Attribute type of the stanza denes themessage type e attribute type could be a chat group chatheadline normal or error An instant messaging network inXMPP can support many XMPP clients and servers with agateway congured to enable other IM network users toconnect to other XMPP users e XMPP server mainlycontrols sessions and XML streams to clients and servers inaddition to routing XML stanzas in the streams [53] eTCP ports 5269 and 5222 have been respectively registeredfor the XMPP server-to-server connections and XMPP clientconnections by the Internet Assigned Numbers Authority(IANA) XMPP enables specication of XMPP ExtensionProtocols (XEP) to enhance its usefulness [17]

XMPP security clients connect to the server usingtransport layer security (TLS) and simple authentication andsecurity layer (SASL) protocols which are at the same timeused by the servers for interdomain communications Fordata condentiality and integrity TLS will encrypt thestreams of XML for all sessions To conrm the identity of anyclient that needs to access the server the SASL protocol doesthe authentication of the XML stream For any communi-cating session with a server to start the client needs to resolvethe serverrsquos DNS hostname For protection of the server of theclient from third-party attacks the server maintains the au-thentication details secret ie IP address and access methodand only original connections of the server are required

In server-to-server communications the SASL protocolis employed for authentication and condentiality Server-to-server communications may as well be deactivated from aserver by an administrator depending on the organizationalpolicy To guard against domain-spoong attacks serverdialback can be a good solution since servers support itsecuring more the XML stanzas Nevertheless it cannot beused to authenticate secure or encrypt in-server streamsand the server identication resulting therein is not strongenough to be relied on It further cannot secure against DNSpoisoning attacks and IP session hijacking for remote do-mains [53] High-security domains can use TLS and SASL

Advantages

(i) It is popular for usage over the Internet especially ininstant messaging and chatting platforms

Disadvantages

(i) It is susceptible to spamming attacks(ii) It has failed to have worldwide support making

Google withdraw support of the standard(iii) It is not secure against DNS poisoning attacks and

IP session hijacking for remote domains

36 Advanced Message Queuing Protocol (AMQP) isprotocol can be utilized by a number of platforms formessage exchange since it is open source and written inmany languages It is a publishsubscribe model depending

MQTT-SNclients Broker

MQTT

(a)

MQTT-SNclients Broker

(b)

Figure 6 (a) Transparent gateway and (b) aggregating gateway

Journal of Computer Networks and Communications 21

on a reliable and efficient messaging queue with capacity forinstant messaging or e-mail for different applications es-pecially in IoT environments It differs from others in thesense that message types can be specified and have theirsource traceable and how to do trade-off between perfor-mance security and reliability For reliability of messagetransmission AMQPmakes use of some delivery guaranteeslike at most once at least once and exactly once It can alsohowever utilize the TCP transport layer for guaranteeingreliability -e AMQPrsquos publishsubscribe method has twocomponents namely (i) exchange queue meant for routingof messages to an appropriate order in the queue and (ii)message queue that acts as storage pending their delivery toreceivers [16] In this protocol model the broker can makerouting decisions contrary to othermessaging systems whereapps using the queue have in them embedded the decision-making logic for recipients or senders -e developer mayneed to check each of the affected apps so as to change thelogic for routing and delivery of the messages [54]

Advantages

(i) It is convenient(ii) It has capacity for interoperability between dif-

ferent vendors(iii) It can use open standards to enable connection

between business partners(iv) It can enable innovations based on AMQP foun-

dations [55](v) It can enable supporting of mission-critical ap-

plications in e-commerce(vi) It can guarantee reliable delivery of messages

Disadvantages

(i) It is not very appropriate for real-timeapplications

(ii) It may not offer automatic discovery(iii) Interoperability is not guaranteed(iv) It lacks open-source libraries for constrained

gadgets [49]

37 Data Distribution Service (DDS) A product of theObject Management Group (OMG) DDS is a pubsubprotocol meant for M2M communications with two sub-layers -ese are the data-centric pubsub sublayer re-sponsible for connecting anonymous data publishers tosubscribers and the data-local reconstruction sublayer that isoptional and integrates DDS into the application layer [56]-e DDS architecture connects system components asshown in Figure 7 Publishers and subscribers are discon-nected with respect to (i) time (ii) space (iii) flow andbehavior (iv) platforms and (v) programming languages[57] A DDS data publisher generates topics and publisher-subscriber coupling is represented as topic name data typeschema and publishers and subscribersrsquo QoS attributesDelivering of data is done by the publisher layer -e datawriter and publishers together make a decision about anynecessary alterations for subsequent forwarding to thesubscribers who need to send data to IoT applications Data

readers are responsible for delivering topics to the sub-scribers after reading through them [56]

Some of the major components of the DDS architectureinclude the following [57]

(i) Topic this is a logical path specifying publicationand subscriptionrsquos data type between DataWritersand DataReaders For a successful session to begintopic names types and DataWriters and Data-Readersrsquo QoS should correspond

(ii) Domain this is a logical setting for transmissionthat can be utilized for separating and optimizingnetwork transmission in a setup of similar appli-cations If DDS apps share a domain ID then theycan exchange data among themselves

(iii) DataWriter and DataReader these are terminalobjects for writing and reading messages re-spectively to and from a universal data space

(iv) Participant this is an object that stands for apublisher or subscriber of the DDS app in a do-main acting like a container for other objects

Advantages

(i) It is highly reliable and offers QoS support(ii) It is highly scalable(iii) It is appropriate for real-time apps(iv) It is fault tolerant(v) It has high level of interoperability(vi) It does not need brokers that may act as

bottlenecks(vii) It offers automatic discovery(viii) It has a user-defined data structure for topic(ix) It supports publisher and subscriber decoupling

Disadvantages

(i) It is originally developed for only stand-aloneLANs

(ii) A lot of memory is needed(iii) It has no open-source libraries to support con-

strained devices(iv) It is hard to design and configure compared to

other protocols [49]

38 RESTful Services REpresentational State Transfer (REST)-is is an architectural style based on a set of principles thatdefine and address networked resources It offers web servicesthat enable transmission from the HTTP device to the devicewithin the IoT architecture and enables clients to reach theserver [16]-is protocol utilizes HTTP commands to backupthe requestresponse messaging model HTTP is a popularand secure protocol in the WWW using TLSSSL [16] ARESTful application has the following characteristics

(i) State and functionality are separated into distrib-uted resources

(ii) All resources are distinctively named by the use ofunique HTTP commands (GET POST PUT orDELETE over the Internet)

22 Journal of Computer Networks and Communications

(iii) It is a clientserver stateless layered protocol thatsupports caching REST uses the same methods asHTTP for all requests or responses with POST andGET for creating and retrieving resources PUT forupdating and changing the state of resources andDELETE for removing resourcesAdvantages of REST over arbitrary web services(SOAP) in the IoT context

(i) It reduces overhead(ii) It reduces parsing complexity(iii) It is stateless(iii) It integrates more tightly with HTTP [58](iv) It is easy to implement and learnis together

with its natural t over HTTP makes it apreferred method for Web 20 applications toshowcase their data and secure M2M sessionsin the IoT

(v) Applications that support RESTful web ser-vices exhibit better performance than the re-source-constrained WSN nodes [59]

(vi) REST is important for the IoT since it supportscommercial M2M cloud platforms

(vii) Its implementation in smart phones and tabletapps is also easy since it needs only an HTTPlibrary that is accessible to all operating systemplatforms [56]

Disadvantages

(i) Although it enjoys wide usage in commercialM2M platforms it is not likely to dominatebecause it is hard to implement

(ii) Its HTTP usage makes it incompatible withconstrained M2M devices leaving its usage toend applications

39 WebSockets It is a web-based protocol operating on asingle TCP channel and providing full-duplex transmissions

It permits secure duplex communication between a clientexecuting unreliable code in a restricted environment and adistant node communicating from that code while using theorigin-based security model employed by web browsers overTCP [60] e protocol is not a requestresponse model andis not a publishsubscribe model So unlike other protocolsa web-socket session is established by the client initiating ahandshake to the server to launch the session is onceestablished a full-duplex client-to-server connection beginsasynchronously and runs until terminated by either party[56] It gives a model for browser-based apps utilizing bi-directional communication with no need of many HTTPconnections to servers

WebSocket comprises two parts ie (i) the handshakehaving a clientrsquos message and the serverrsquos handshake responseand (ii) the data transfer HTTP headers should be repeated byapplications in all client requests and server responses arisingfrom the endless polling which might increase the commu-nication overhead subject to the application [61] WebSocketcan be used to come up with scalable real-time web appli-cations since it can enable full-duplex transmission via onesocket across the Web Its security can be ensured with thehelp of TLSSSL is protocol is suitable for resource-con-strained devices and IoTapplications will not have a good userexperience with it since it operates a clientserver architectureNevertheless it can outcompete all TCP-based protocolsbecause of the following advantages

(i) It is t for real-time transmission(ii) It has improved security(iii) It minimizes transmission overhead(iv) It can provide ecient messaging systems using

WebSocket ApplicationMessaging Protocol (WAMP)[17]

310 Streaming (Simple) Text-Oriented Messaging Protocol(STOMP) is is a TCP-based protocol designed for textmessage-oriented middleware (MOM) with the use of

Topic

Datareader

Datareader

Datareader

Datawriter

Datawriter

Datawriter

Subscriber PublisherPublisher

Data bus (DDS domain)

Subscriber

Topic Topic

Figure 7 DDS architecture

Journal of Computer Networks and Communications 23

HTTP-like commands for interoperability among platformslanguages and brokers Data communication takes placebetween the client and the broker in line with commands(like CONNECT DISCONNECT ACK NACK SUB-SCRIBE UNSUBSCRIBE SEND BEGIN COMMIT orABORT) then headers in the form ltkeygt ltvaluegt (one perline) then a blank line and then body content terminatingin a null character Server-client communication takes placeover a MESSAGE RECEIPT or ERROR frame with headerand body content format being analogous MOM productssupporting STOMP include Apache ActiveMQ or FuseMessage Broker HornetQ Open Message Queue(OpenMQ) RabbitMQ and syslog-ng through its STOMPdestination plugin [62] In addition to these other appli-cation layer protocols include simple media control protocol(SMCP) lightweight local automation protocol (LLAP)simple sensor interface (SSI) lightweight M2M (LWM2M)XMPP-IOT and simple object access protocol (SOAP)

Advantages

(i) It is lightweight for constrained networks(ii) It has flexibility to choose quality of services with

the given functionality(iii) It is standardized by OASIS Technical Committee(iv) It is easy and quick to implement

Challenges and opportunities

Here security threats are imminent because of the open-endedness of the application layer [63] -ese could includethe following

(i) -ere is poor or lack of a security design for anapplicationrsquos function Coming up with a gooddesign to address this could be a good ideaopportunity

(ii) Some of the programs might have backdoorsevading otherwise secure controls to allow un-authorized access to network resources We requiremechanisms to patch up such loopholes

(iii) Applications that lack or have weak authenticationmechanisms can be easily targeted by unlicensedusers to abuse the network Such mechanisms needto be strengthened further

(iv) Excessively complicated access control rules maybecause of their misunderstanding or poor writinggive unauthorized access allowing dubious activityrather than protecting the network they purport toprotect -ese should be simple and clear to un-derstand by all users

(v) In booting workstations without disks and managingnetwork devices we usually use the TFTP protocolalthough access does not require authenticating theuser by usernamepassword -is makes an intruderable to gain easy access to configaccess info as longas he can guess the filenames -is needs to be fixed

A few steps can be used to secure the application layerTo protect network data transmitted across the network

applications need to strengthen their encryption and au-thentication mechanisms Apps too need to enable strictcontrols of privileges for data access using good mechanismsto balance between usability and effectiveness [63] Appli-cations handling sensitive data need detailed logging andaudit capability as well as testing and reviewing

Table 3 summarizes the application layer protocols dis-cussed with their weaknesses and differences among them

4 Cross-Layer QoS Strategies

41 Service-Differentiated Real-Time Communication Scheme(SDRCS) -is is an event-driven routing protocol that routesreal-time traffic by featuring a cross-layer packet-forwardingdesign in which it integrates the real-time routing functionalitywith a new priority-based MAC scheme -e protocol ap-proximates distributed packet traversal speed for traffic clas-sification and admission control on the basis of this design Itfurther localizes decision-making by prioritizing packet for-warding to maximize the speed of packets [64]

Advantages

(i) It improves bandwidth utilization bandwidthdegradation due to unschedulable packets can beavoided

(ii) It offers a lightweight packet schedulability estima-tion mechanism Here it uses received signalstrength and admission control and then earlymissed-deadline packet-dropping policies are made

(iii) It delivers high end-to-end throughput via highersource data rates alongside strict end-to-end latencyneeds

(iv) Nodes which meet real-time requirements of lesstraffic load and better channel quality are givenhighest priority when it comes to packet forwarding

-e SDRCS operates via 5 mechanisms namely (1) RSS-(Rich Site SummaryReally Simple Syndication-) basedgrouping (2) admission control (3) prioritized queuing(4) real-time MAC and (5) dynamic forwarding -isprotocol has a queuing policy that employs per-hopdeadline-based queues and the nodes schedule theirpackets using FIFO priority-based queues [1] -e prin-cipal features of this protocol are mentioned as follows

(i) It is event driven and is easily adaptable to anynetwork changes

(ii) It does not need additional hardware for locali-zation or multichannel transmission

(iii) -e cross-layer packet-forwarding strategy can beused when transmitting multimedia traffic

(iv) It is able to efficiently circumvent any voids therein

Early deadline miss (EDM) policy drops all unscheduledpackets -e SDRCS circumvents wrong packet dropsdue to EDM since it adjusts to network changes easily

42 Sensor Fuzzy-Based Image Transport (SUIT) -is pro-tocol regulates congestion using a scheme based on fuzzy

24 Journal of Computer Networks and Communications

logic It transmits packets with reduced quality to allowablelevels if there is congestion enabling it to transmit lower-quality packetsframes without being dropped-e numberof packets delivered per second thus rises and while thisenhances video streaming apps it is more appropriatelymade for those that quickly transmit non-real-time videosuch as in video surveillance SUIT can stream JPEG pic-tures other than the traditional predictive video codingmethods that are not suitable for sensor devices because oftheir constraints of energy memory and computationalspeed Progressive JPEG-PJPEG performs better via packettransmission rate and delay though it degrades the imagequality because of congestion [65] -is protocol employscross-layer methods for interlayer data exchange though itstill employs some layering making it still possible topreserve all advantages of the layered design since it hasapplication and transport layers but has noMAC or routinglayer

For congestion detection and control SUIT bases on thepercentage number of in-bound and out-bound frames perwindow number of contenders and buffer occupancy of thenext-hop node Since buffer management has an effect oneffectiveness and QoS inWMSNs it is useful in the transportlayer so the protocol prioritizes packets for better QoS SUITreorders packets by tracking the sequence numbers of thedelivered packets since they all have a source ID framenumber and sequence number that it uses to detect anymissing packets -e protocol however does not providepacket reliability [65]

Advantages

(i) -e protocol prioritizes packets for better QoS(ii) It can transmit lower-quality packetsframes

without being dropped

Disadvantages

(i) It does not provide packet reliability

43 Network Layer QoS Support Enforced by a Cross-LayerController (NLQS) -is scheme permits packet-level servicedifferentiation as a function of throughput end-to-endpacket error rate and delay [66] (see Figure 8) -is en-hances network layer QoS It comprises a cross-layer controlunit (XLCU) for configuring and controlling networkingfunctionalities at physical MAC and network layers [18]-is depends upon the unified logic which affects choices forapplication layer needs and status of functional blocks whichdo implementation of networking functions [66]

Advantages

(i) Cross-layer interactions can be controlled withoutweakening the upgradability modularity andsimplicity of designing the system

44 Cross-Layer Signaling Shortcuts (CLASS) CLASS has ahigh level of effectiveness flexibility and comprehensive-ness According to [67] it has the following features

(i) Direct signaling among remote layers leading tofaster transmission

(ii) Lightweighted internal message format standard-ized protocols are not suitable for use in internalsignaling since they are not usually lightweight forinstance transmitting against faults in the networkCLASS requires only three fields namely destina-tion address event type and event contents

(iii) -e setup of its exterior messages is standardconsidering external signaling ICMP and TCPIPare suitable for general messages and short notifi-cations respectively

(iv) Message control protocol should ensure optimizedand organized exchange of dense simultaneousmessages across layers for high efficiency andavoidance of possible conflicts -e protocol appliesto many cross-layer signaling situations

Table 3 Differences and problems with application layer protocols

Protocol Architecture Transport QoS options Security Weaknesses

CoAP Requestresponse UDP DTLS No in-built security features DTLS does not supportmulticast

MQTT Publishsubscribe TCP TLSSSL Insufficient security at the protocol level sincepayload values are not encrypted

SMQTT Publishsubscribe TCP CPKP-ABE Key revocation and group pubsub for distributedSMQTT are still a challenge

MQTT-SN Publishsubscribe UDP TLSSSL SSLTLS suffers from attacks like BEAST CRIMERC4 and Heartbleed

XMPP RequestresponsePublishsubscribe TCP X TLSSSL Susceptible to spamming attacks and lacks worldwide

supportAMQP Publishsubscribe TCP TLSSSL Inappropriate for real-time applications

DDS Publishsubscribe TCPUDP TLSSSL Memory intensive and no open-source libraries forconstrained devices

REST Requestresponse HTTP X HTTPS Hard to implement Uses HTTP so incompatiblewith constrained apps

WebSockets PublishsubscribeClientserver TCP X TLSSSL Bad user experience for IoT apps since it runs the

clientserver architectureSTOMP Clientserver TCP HTTP -e broker can act as a bottleneck

Journal of Computer Networks and Communications 25

Advantages

(i) e protocol is scalable and has a very lowpropagation delay

(ii) It is very ecient and exible since it uses directsignaling between any two layers

(iii) It is very comprehensive and eective

45 Cross-Layer and Multipath-Based Video Transmission(CMVT) CMVT is a hybrid of application and networklayers At the application layer it carries out the encoding ofvideo streams to video data frames (I- P- and B-frames) bymeans of the MPEG-4 encoding format At the networklayer route discovery and data transfer take place Twoalgorithms greedy forwarding and rollback are used todiscover a number of paths from the source to the sink nodevia the route discovery method [20] A specied node icalculates the evaluation of its adjacent node j using thefollowing equation

fij (1 minus α)d2(j D) minus d2min(i)d2max(i) minus d

2min(i)

+ αeinit(j) minus eres(j)

einit(j)

(19)

where fij evaluation value for the node i to j d2(j D) isthe distance from the node j to the destination node D

d2min(i) and d2max(i) are the minimum and maximum dis-

tances of neighbors of the node i to D respectively einit(j) isthe initial energy of the node j eres(j) is the current residualenergy of the node j and α is the energy coecient given asfollows

α emax(i) minus emin(i)

emax(i) (20)

where emax(i) and emin(i) are respectively the maximumand minimum energy left of all node irsquos neighbors

e network layer further transmits video streams withCMVT doing status evaluation choosing an appropriatecommunication path for any packets and the QoS guaranteelevel for the path i is calculated as

fi (1 minus ω)hisum hi

+ ωnisum ni

(21)

where fi is the evaluation value for the path i hi are the hopsfor the path i ni is the summation of packets sent via isum ni isthe summation of packets sent by sources and ω is theenergy consumption factor [18]

Advantages

(i) e CMVT protocol is very superior in mediatransmission particularly large-scale WMSNs

To network peers

Flow requirements

-

Atimes

Admissioncontrol

Capacityassignment

Next-hopselection

Coding rate

Data rate

QoScontracts

Cross-layer controlunit (XLCU)

Localization subsystem

Synchronizationsubsystem

Networkcoordinationunit (NCU)

UWB modulator

Channel coder

QoS scheduler

Georouting

Source coder

Rate control

Figure 8 Cross-layer controller architecture

26 Journal of Computer Networks and Communications

46 Cross-Layer Cooperative MAC (CoopMAC)CoopMAC [68] is based on the IEEE 80211 DCF approach-ere is a reduction in interference of neighboring cellsleading to uniform coverage in densely deployed networks[68] A connection is established by two-way authenticationbetween the sender and the receiver When a Clear-to-Send(CTS) packet and short interframe space (SIFS) are obtainedby the sender packets are transmitted right to the receiverwith no valuable cooperation If available the two nodes findout whether they can exchange data -is is done by the useof a helper identification signal to establish how feasiblecooperative communication is It is established in thepresence of a signal and if it is not available direct com-munication is activated For distributed systems helper-initiated cooperation is preferable because of the presence ofthe RTSCTS packets [69]

Advantages

(i) CoopMAC has an advantage over others due to thespatial diversity amongst the three nodes [70]

(ii) It has the potential to achieve significantthroughput and delay with no complexity in thedesign of the system

47 Cooperative MAC Protocol for Multihop Networks (M-CMAC) Just as CoopMAC M-CMAC is made in such away that high data rate stations help low data rate stations toforward traffic for broadcasting [71] -ere are also helperschosen in such a way that 2 fast-hop transmissions replace aslow hop transmission -e helper via whom there existsleast delay from the sender to the receiver is considered bestbecause of theminimum two-hop transmission rate and so isthe chosen neighbor to the duo Here the working as-sumption is that all nodes have their position coordinatesknown and thus the Euclidean distances between all pairscan be calculated and converted to the data rate for that linkAll nodes have cooperative tables (CT) of potential helperswith destination and helper MAC address Euclidean dis-tance and total distance through the helper Any sender withdata to transmit checks out for the existence of any helper inthe CT for the destination and if available forwards RTS forchannel reservation for single-hop duration M-CMAC hasgot an RTS format having a shape with five fields similar tothe one given as follows

Framecontrol Duration Source

addressDestination

addressHelperaddress

-e destination nodersquos address is kept by the source in thepart for the helper address that is in turn saved as the desti-nation address Nodes inspect helper and destination addressfields on receipt of RTS whereby the node acts as the othernodersquos helper in case the helper address field is not the sameand the helper transmits CTS back to the source if at all it wantsto forward data On delivery of the CTS packet the sourcetransmits to the helper which in turn sends the packets to thedestination that also sends ACK to the helper on receipt of thepacket In M-CMAC there exists a higher level of channelreuse (parallel transmissions) due to the increased number of

nodes therein that subsequently raises availability of helpers fordata forwarding -is results in increased throughput incomparison with CoopMAC and IEEE 80211 DCF [71]

Advantages

(i) It promises a higher throughput compared toCoopMAC and IEEE 80211 DCF

48 Cluster-Based Cooperative Routing (CBCR) Protocol-e CBCR protocol has a multihop data-forwarding func-tion realized at the link layer with cooperative links that useM-CMAC -is protocol encompasses two stages namelythe routing relay selection phase and data forwarding phase

(i) Routing relay selection phase all nodes announcetheir presence to their neighbors by broadcastingperiodical beacon messages that carry the MAC ad-dress of the node Each of these constructs a relaytable that contains all neighboring nodes with whichit is able to communicate -e node further broad-casts its neighbor list in case of any change to itsentries since the previous broadcast -e MAC ad-dresses of nodes next to the node X are contained inits relay tablersquos column one and the neighbor nodersquosrow has MAC addresses for neighbors of the adjacentnode Based on its relay table each node in-dependently chooses routing relays Choosing a nodeas a relay node depends on the number of nodes itconnectsmdashit should be the highest number [71]

(ii) Data forwarding phase a node with packets totransmit needs to first verify if the receiver is in thesame cluster and if that receiver has a helper to whichthe packet is thence forwarded but in whose absencethe packet is delivered right to the receiver In casethe intended receiver is in a different cluster therelay table will be checked to see whether that re-ceiver is reachable via other routing relays If it isreachable packets will be sent to the routing relaydirectly or via the helper if at all the relay has one Incase of destination unreachable via relays packetsare multicast to all relays by the node [71]Advantages

(i) It enables multicasting

49 MAC-PHY Cross-Layer Protocol A cooperative cross-layer standard for cooperation at the physical layer in next-generation WMSNs is developed in [70] It provides acomplete MAC layer algorithm that gives a supportive shellfor the PHY-MAC layer Like CoopMAC its scheme de-pends on the middle node for communication betweennodes-eMAC layer protocol has been changed to regulateinformation exchanges at the physical layer -e receivernode receives duplicate packets from source and helpernodes to decode the data [70]

410 MAC-Centric Approach -is cross-layer protocoltargets multimedia applications by using the MPEG-4

Journal of Computer Networks and Communications 27

scheme [21] It is characterized by 4 access categories(AC3ndashAC0) according to their priority in transmission It ismeant for supporting different QoS needs in upcomingvideo applications and enabling MAC layer differentiationfor H264 partitioning [18] Because of low bandwidth delayand other QoS challenges that cause inefficiency in trans-mittingmultimedia data inWMSNs a number of algorithmsdepending on IEEE 80211e have been proposed to supporttransmission of quality videos [72] Selecting an AC dependson the measures for QoS such as the loss rate and delay-us AC3 having the highest priority is mapped to theparameter set concept since the stream is sensitive to bit lossespecially in video transmission [21]

411 Adaptive Cross-Layer Forward Error Correction(ACFEC) In the ACFEC [73] model data packets are ex-changed among nodes through the access point (AP) whichoperates the infrastructure mode in which it adds FEC to videodata [72] -ese data are dealt by encapsulation using astreaming server to the receiver through thewireless AP as RTPpackets An adaptive FEC controller senses the category ofpackets out of the RTP header and recovers the header ofpackets from UDP -e encoder generates some error-cor-recting packets whose number is determined by the sourcepacket number of the block Multimedia transmissions aremonitored by the controller using MAC failure data and itscounter is increased by one if there is a failed transmission-econtroller uses the failure counter to change the packet numberproduced after transmitting a block [18] In case there are lostpackets it changes the redundancy rates and produces extrapackets to replace the lost ones and satisfy the requirements ofthe receiving node [72] -e FEC packet number is enhancedor reduced to satisfy the receiverrsquos requirements and stoppacket losses-is is done through accurate detection of packetlosses and adjustment of redundancy rates If all video datapackets are well received there will not be generation of FECpackets [73]

Advantages

(i) It promises a good QoS through packet loss re-duction and redundancy rate adjustment

412 Balanced Cross-Layer Fuzzy Logic (BCFL) DesignRouting Algorithm in WSNs A new fuzzy logic-basedrouting algorithm (BCFL) [74] was designed using disper-sion of the cross-layer parameter as the fuzzy logic inferencesystem input Each cross-layer parameter has a dynamicweight depending on the value of dispersion -e designcomes with some innovations as follows (i) for fuzzy logicinference system input the absolute parameter value issubstituted by parameter dispersion thereby significantlyreducing algorithmic complexity (ii) dispersion does notchange with the order of magnitude according to the dis-persion formula and (iii) the weight of the parameter de-pends on the size of its dispersion and the two are inverselyproportional to each other -is enables BCFL to have somenobility unlike other algorithms as per the following dis-tinguishing properties making it advantageous

(i) It has got simple if-then rules which remain con-stant even when the constraints increase

(ii) It is capable of dealing with many constraints withno increment in complexity

(iii) It has capacity to yield a more balanced solutionthan other algorithms

(iv) It is easily adaptable to changes in network con-ditions and topology even when the changes arefrequent like in underwater WSNs -e algorithm isuseable in choosing the CH in cluster-based routingprotocols [74]

413 Minimum Hop Disjoint Multipath Routing Algorithmwith Time Slice Load-Balancing Congestion Control SchemeMHDMwTS is a two-phased routing protocol comprisingpath build-up and path acknowledgment phases It consists ofmultiple sources each of which has three build-up disjointpaths namely primary alternate and back-up paths Onactivation the source node requests to build up a route to thenearest hop neighbor which is the path build-up phase [19] Inthis phase step one has source activation in which therequested node for path building adds its number and time-stamp and sends to the least hop-count neighboring node-isgoes on until the least time latency sink having the informationneeded to construct the primary route is reached [75] In steptwo there is path extraction when the new package fromanother path arrives whereby the extracted path is comparedwith the primary path If the node is shared the package will berejected or an alternative path will be searched for to get abackup through comparison of the preceding two paths Phasetwo is path acknowledgment where we have the third step inwhich the sink returns the ACK packet to the sender with pathinformation having nodes and their time information after ithas computed it using the timestamp [1819]

Advantages

(i) -e protocol minimizes end-to-end latency(ii) It enhances congestion control

414 Cross-Layer Optimal Design (CLOD) Authors in [76]come up with a CLOD for scheduling at the data link layerrouting at the network layer and controlling congestion atthe transport layer with an assumption of fixed link capacity-rough congestion control energy efficiency is improvedTransport layer congestion at the nodes is minimized bycompressed sensing (CS) in which transmitted bits are re-duced whereas optimally allocating resources decreasescongestion on the links at the data link layer CLODpromises minimized computational complexity and betterperformance in light traffic scenarios [77] Generating andstoring CLOD is more efficient compared to Gaussianrandom matrices It prolongs network lifetime and savesenergy

Advantages

(i) It does congestion control subsequently increasingenergy efficiency

28 Journal of Computer Networks and Communications

(ii) It reduces computational complexity(iii) It improves performance in light traffic scenarios

415 Challenges and Opportunities of the Cross-Layer Design(CLD)

(i) Different layers in a real network assume differentfunctions andor services A layer only communi-cates with its neighboring layers -e layered modeldegrades system performance because of many

features in wireless transmission QoS parameters inthe layers of WMSNs are shared by routing pro-tocols to optimize the performance Neverthelessthere is a need to design cross-layer models forincreased efficiency in routing [1]

(ii) -e physical layer plays a highly vital role in CLDRate adaptation and channel allocation take place atthe physical layer via signal processing to enhanceQoS End-to-end performance in wireless media isaffected by the changes therein given the effect on

Table 4 A summary of some of the reviewed cross-layer models

Protocol Layers Aims Comments QoS parameters

SDRCS MACPHY Routing real-time traffic Transmits multimedia traffic viacross-layer packet forwarding

-roughputlatency

SUIT Applicationtransport Transmission of non-real-time video Uses fuzzy logic to regulate

congestionTransmission rate

delay

CLASS Any twoDesign serves as a framework fordifferent implementations ofdifferent application scenarios

Scalable very efficient flexible andhas very low propagation delay

Propagation delayjitter

NLQS PHYMACnetwork

Permits packet-level servicedifferentiation as a function of

throughput packet error rate anddelay

Network layer QoS is enhanced -roughput delaypacket error rate

CMVT Applicationnetwork

Encodes video streams to video dataframes via MPEG-4 encoding anddoes route discovery and data

transfer

Greedy forwarding and rollback areused to find source-to-sink paths Energy

CoopMAC PHYMAC Offers spatial diversity among thethree nodes

Helper ID signal finds out howfeasible cooperative communication

is

-roughput ratedelay

M-CMAC PHYMAC Increases end-to-end throughputand packet delivery ratio

Euclidean distances between nodesare calculated and converted to the

data rate for a link

-roughputpacket delivery

ratio

CBCR PHYMACMinimizes control overhead andtime consumed in establishing thecooperative paths than M-CMAC

Energy consumption is moreuniformly distributed in a network

enhancing network lifetime

-roughputpacket deliveryratio energy

MAC-PHY MACPHYMeant to enable PHY layer

cooperation and maximize gains ofcooperation at the MAC layer

Leverages both spatial diversity andcoding gain -roughput delay

MAC-centric MACAPPL

Meant to support QoS needs in newvideo apps and enable MAC layer

differentiation for H264partitioning

Targets multimedia applications byusing the MPEG-4 scheme

Delay packet lossrate

ACFEC MACnetwork(UDP)

Meant to enhance the quality ofvideo streaming over 80211WLANs

and overcome packet losses

Adjusts redundancy rates toovercome channel fluctuations and

detect and reduce packet lossPacket loss rate

MHDMwTS mdash

Meant to provide reliable datatransfer with the multipath routingand load-balancing congestioncontrol method in WMSNs

More reliable than basic routingschemes for transport multimedia

data

Latency packagetransmit rate

BCFL mdash

Introduces dispersion into fuzzylogic-based routing and sets every

cross-layer parameter with adynamic weight

Can be used to select a CH in cluster-based routing protocols andproposes a dispersion formula

Node utilitydispersion

CLODDatalinknetworktransport

Prolongs network lifetime andachieves congestion control anddesigned for lightly loaded WSNs

Assumes fixed link capacity andintegrates compressed sensing

technology

-roughputaverage energy CS

error ratio

Journal of Computer Networks and Communications 29

the purposefulness of the protocols at the networklayer CLD offers solutions to conserving powerminimizing energy and controlling both the flowand congestion in the network as well as fault tol-erance so it is an opportunity for designers con-sidering other layers We further desire to developCLDs that shall make an impact on network oper-ation and get close attention [18]

(iii) CLD interface standardization the architecturemust deliver module functionality although thereexist some queries amongst modules about possibleinterfaces predicted by the necessity to share in-formation among remote protocol layers Technicalchallenges include developing designing andstandardizing of cross-layer interfaces and algo-rithms that satisfy the cross-layer optimization re-quirements among protocol layers

Table 4 summarizes these cross-layer models with a fewcomments

5 Future Research Direction

51 At theMACLayer In WMSNs achieving duty cycling istricky since video traffic is volatile Listening to adjacentnodes wastes energy We need to switch states from awake tosleep -is requires further research

Self-organizing networks and green communicationalthough most CRWSN energy-saving algorithms aredesigned to reduce node transmission power energyconsumption is inevitable during the operational mode ofBSs because of internal processing Scholars are trying withpolicy-makers to encourage a shift to greenWSNs to reduceoperational costs and carbon footprints With self-orga-nizing CRWSNs for automatic switch-off when idle energyis saved which calls for cognitive switching algorithms Sointegrating cognitive switching with radio resource allo-cation in CRWSNs is a promising research area [78]

52 At the Application Layer Considering a WMSN ar-chitecture end users access the network via the BS to andfrom which they send their sensed data We desire to getguaranteed QoS and energy efficiency by designing a newarchitecture in the form of hardware and software whichneeds to be investigated further for a fix of the same [1]

QoS and energy efficiency are very crucial in WMSNsand mostly real-time apps that require guaranteed band-width and throughput in their network lifetime Mostprotocols ignore base station (BS) mobility and WMSNnodes Traffic management telemedicine and battlefieldsurveillance apps require mobile nodes or BSs so designingdynamic routing protocols to be adaptable in these cir-cumstances is necessary

53 Cross-Layer Security of data transmitted on WMSNseg in military surveillance and e-commerce is key withQoS and energy efficiency WMSNs are vulnerable to attacks

like worms and sinkhole Moreover their computationpower is not that high complicating implementation ofstrong secure protocols -is needs further attention

For fairness and priority issues in CRWSNs nodes areassigned diverse priorities depending on importance andurgency of their data For better network performance thereis a need to develop fair resource allocation mechanismswith priority-based fairness in sensors-ere is still potentialin this research area [78]

We must develop multiobjective adaptive protocols tooptimize QoS metrics involved in routing for best trans-mission results to get a trade-off between different opti-mization metrics We need to design standardized cross-layer algorithms which are able to meet cross-layer opti-mization standards

6 Conclusion

We have reviewed the different QoS strategies for WSNs inthe context of IoT from the MAC layer and application layeras well as the cross-layer paradigm For the MAC layer wehave reviewed protocols for WSNs such as the hybrid of thecontention-free and contention-based MAC protocols andthose for WMSNs We further reviewed a number of ap-plication layer protocols including many machine-to-ma-chine requestresponse and publishsubscribe protocols Forsystem optimization cross-layer QoS strategies are veryimportant in wireless communication We have reviewed anumber of cross-layer strategies For all categories reviewedchallenges and opportunities are discussed Finally somepossible future directions are discussed for research andapplication showing a promising potential for future re-search in this relatively new area especially as more WMSNapplications emerge recently

Conflicts of Interest

-e authors hereby declare no conflicts of interest regardingthe publication of this work

Acknowledgments

-is research was funded by the National Natural ScienceFoundation of China (Nos 61602398 61672447 and61711540306)

References

[1] V Bhandary A Malik and S Kumar ldquoRouting in wirelessmultimedia sensor networks a survey of existing protocolsand open research issuesrdquo Journal of Engineering vol 2016Article ID 9608757 27 pages 2016

[2] I F Akyildiz T Melodia and K R Chowdhury ldquoWirelessmultimedia sensor networks applications and testbedsrdquoProceedings of the IEEE vol 96 no 10 pp 1588ndash1605 2008

[3] M Iqbal M Naeem A Ahmed M Awais A Anpalagan andA Ahmad ldquoSwarm intelligence based resource managementfor cooperative cognitive radio network in smart hospitalsrdquoWireless Personal Communications vol 98 no 1 pp 571ndash5922018

30 Journal of Computer Networks and Communications

[4] S A Alvi B Afzal G A Shah L Atzori and W MahmoodldquoInternet of multimedia things vision and challengesrdquo AdHoc Networks vol 33 pp 87ndash111 2015

[5] Q F Hassan A U R Khan and S A Madani Internet ofings Challenges Advances and Applications CRC PressTaylor amp Francis Group A Chapman amp Hall Group BocaRaton FL USA 2018

[6] Q Wang Y Zhao W Wang et al ldquoMultimedia IoT systemsand applicationsrdquo in Proceedings of the 2017 Global Internet ofings Summit (GIoTS) vol 2 Geneva Switzerland June 2017

[7] A Aslam and E Curry ldquoTowards a generalized approach fordeep neural network based event processing for the internet ofmultimedia thingsrdquo IEEE Access vol 6 pp 25573ndash25587 2018

[8] M Jridi T Chapel V Dorez L Bougeant and A Le BotlanldquoSoC-based edge computing gateway in the context of theinternet of multimedia things experimental platformrdquoJournal of Low Power Electronics and Applications vol 8no 1 p 1 2018

[9] H Noura A Chehab L Sleem andM N Rapha ldquoOne roundcipher algorithm for multimedia IoT devicesrdquo MultimediaTools and Applications vol 77 no 14 pp 18383ndash18413 2018

[10] T Balan D Robu and F Sandu ldquoMultihoming for mobileinternet of multimedia thingsrdquo Mobile Information Systemsvol 2017 Article ID 6965028 16 pages 2017

[11] H Shen and G Bai ldquoRouting in wireless multimedia sensornetworks a survey and challenges aheadrdquo Journal of Networkand Computer Applications vol 71 pp 30ndash49 2016

[12] A Ahmad M H Rehmani H Tembine O A Mohammedand A Jamalipour ldquoIEEE access special section Editorial optimization for emerging wireless Networks IoT 5G andsmart grid communication networksrdquo IEEE Access vol 5pp 2096ndash2100 2017

[13] B A Muzakkari M A Mohamed M F A Kadir andZ Mohamad ldquoRecent advances in energy efficient-QoS awareMAC protocols for wireless sensor networksrdquo InternationalJournal of Advanced Computer Research vol 8 no 38pp 212ndash228 2018

[14] M A Yigitel O D Incel and C Ersoy ldquoQoS-aware MACprotocols for wireless sensor networks a surveyrdquo ComputerNetworks vol 55 no 8 pp 1982ndash2004 2011

[15] A M Abbas and O Kure ldquoQuality of Service in mobile ad hocnetworks a surveyrdquo International Journal of Ad Hoc andUbiquitous Computing vol 6 no 2 p 75 2010

[16] M Q Shatnawi ldquoApplication layer protocols for the internetof things a surveyrdquo in Proceedings of the InternationalConference on Engineering amp MIS (ICEMIS) Agadir Mo-rocco September 2016

[17] V Karagiannis P Chatzimisios F Vazquez-gallego andJ Alonso-zarate ldquoA survey on application layer protocols forthe internet of things research motivationrdquo InternationalJournal of Advanced Research in Computer Science vol 8no 3 pp 1ndash10 2015

[18] M S Bernard T Pei Z Li and K Li ldquoQoS strategies forwireless multimedia sensor networks in the context of IoTrdquoLecture Notes of the Institute for Computer Sciences SocialInformatics and Telecommunications Engineering vol 4pp 228ndash253 2019

[19] A AlAmri and M Abdullah ldquoCross-layer quality of serviceprotocols for wireless multimedia sensor networksrdquo in Pro-ceedings of the International Conference on CommunicationManagement and Information Technology pp 649ndash658Cosenza Italy April 2017

[20] J Guo L Sun and R Wang ldquoA cross-layer and multipathbased video transmission scheme for wireless multimedia

sensor networksrdquo Journal of Networks vol 7 no 9pp 1334ndash1340 2012

[21] A Ksentini A Naimi and M Gueroui ldquoToward an im-provement of H264 video transmission over IEEE 80211ethrough a cross-layer architecturerdquo IEEE CommunicationsMagazine vol 44 no 1 pp 107ndash114 2006

[22] W Ye J Heidemann and D Estrin ldquoAn energy-efficientMAC protocol for wireless sensor networksrdquo in Proceedings ofthe 21st International Annual Joint Conference of the IEEEComputer and Communications Societies (INFOCOM 2002)New York NY USA June 2002

[23] J Polastre J Hill and D Culler ldquoVersatile low power mediaaccess for wireless sensor networks categories and subjectdescriptorsrdquo in Proceedings of the Second InternationalConference on Embedded Networked Sensor Systems Sen-Sysrsquo04 pp 95ndash107 Baltimore MA USA November 2004

[24] T Danmanee K N Nakorn and K Rojviboonchai ldquoCU-MAC a duty-cycle MAC protocol for internet of things inwireless sensor networksrdquo ECTI Transactions on ElectricalEngineering Electronics and Communications vol 16 no 2pp 30ndash43 2018

[25] T Van Dam and K Langendoen ldquoAn adaptive energy-efficientMAC protocol for wireless sensor networksrdquo in Proceedings ofthe 1st International Conference on Embedded NetworkedSensor Systems Los Angeles CA USA November 2003

[26] P Xie and J Cui ldquoR-MAC an energy-efficient MAC protocolfor underwater sensor networksrdquo in International Conferenceon Wireless Algorithms Systems and Applications (WASA2007) Chicago IL USA August 2007

[27] A Rahim N Javaid M Aslam U Qasim and Z A KhanldquoAdaptive-reliable medium access control protocol forwireless body area networksrdquo in Proceedings of the AnnualIEEE Communications Society Conference on Sensor Meshand Ad Hoc Communications and Networks (SECON) SeoulKorea June 2012

[28] H S Kwak S Ullah D H Kwak and C H Lee ldquoA powerefficient MAC protocol for wireless body area networksrdquoJournal of the Korean Institute of Intelligent Systems vol 8no 6 pp 131ndash140 2009

[29] C Omeni A J Burdett and M Park ldquoEnergy efficientmedium access protocol for wireless medical body area sensornetworksrdquo in Proceedings of the 4th IEEEEMBS InternationalSummer School and Symposium on Medical Devices andBiosensors pp 29ndash32 Cambridge UK August 2007

[30] L Sitanayah C J Sreenan and K N Brown ldquoER-MAC ahybrid MAC protocol for emergency response wireless sensornetworksrdquo in Proceedings of the Fourth International Con-ference on Sensor Technologies and Applications pp 244ndash249Mestre Italy July 2010

[31] Y Sun and D B Johnson ldquoRI-MAC a receiver-initiatedasynchronous duty cycle MAC protocol for dynamic trafficloads in wireless sensor networksrdquo in Proceedings of the 6thACM conference on Embedded network sensorsystemsmdashSenSysrsquo08 pp 1ndash14 Raleigh NC USA November2008

[32] M Buettner G V Yee E Anderson and R Han ldquoX-MAC ashort preamble MAC protocol for duty-cycled wireless sensornetworksrdquo in Proceedings of the 4th International Conferenceon Embedded Networked Sensor Systems SenSys 2006pp 307ndash320 Boulder Colorado USA October 2006

[33] S Boulfekhar and M Benmohammed ldquoSynchronous receiverinitiated MAC protocol for long-lived sensor networksrdquoComputers amp Electrical Engineering vol 40 no 2 pp 504ndash516 2014

Journal of Computer Networks and Communications 31

[34] S Mehta and K S Kwak ldquoH-MAC a hybrid Mac protocol forwireless sensor networksrdquo International Journal of ComputerNetworks amp Communications vol 2 no 2 pp 108ndash117 2010

[35] I Rhee A Warrier M Aia J Min and M L Sichitiu ldquoZ-MAC a hybrid MAC for wireless sensor networksrdquo IEEEACM Transactions on Networking vol 16 no 3 pp 511ndash5242008

[36] S Arshad A Al-sadi and A Barnawi ldquoZ-MAC performanceevaluation and enhancementsrdquo Procedia Computer Sciencevol 21 pp 485ndash490 2013

[37] Y Liu I Elhanany and H Qi ldquoAn energy-efficient QoS-aware media access control protocol for wireless sensornetworksrdquo in Proceedings of the IEEE International Confer-ence on Mobile ad hoc and Sensor Systems pp 9ndash11 Wash-ington DC USA December 2005

[38] J Decotignie ldquoWiseMAC an ultra low power MAC protocolfor the downlink of infrastructure wireless sensor networksrdquoin Proceedings of the Ninth International Symposium onComputers And Communications pp 244ndash251 AlexandriaEgypt June 2004

[39] R A Rashid ldquoDevelopment of energy aware TDMA-basedMAC protocol for wireless sensor network systemrdquo EuropeanJournal of Scientific Research vol 30 no 4 pp 571ndash578 2009

[40] H P Van Hoesel ldquoA lightweight medium access protocol(LMAC) for wireless sensor networksrdquo in Proceedings of theInternational Workshop on Networked Sensing Systems Bal-timore MD USA November 2004

[41] C Li P Wang H Chen andM Guizani ldquoA cluster based on-demand multi-channel MAC protocol for wireless multi-media sensor networksrdquo in Proceedings of the IEEE In-ternational Conference on Communications pp 2371ndash2376Beijing China May 2008

[42] J L Sobrinho and A S Krishnakumar ldquoQuality-of-service inad hoc carrier sense multiple access wireless networksrdquo IEEEJournal on Selected Areas in Communications vol 17 no 8pp 1353ndash1368 1999

[43] J P John ldquoProviding MAC QoS for multimedia traffic in80211e based multi-hop ad hoc wireless networksrdquo ComputerNetworks vol 51 no 1 pp 153ndash176 2007

[44] S Lofty and N Padmavati ldquoA survey on mobility basedprotocols in WSNsrdquo in Proceedings of the InternationalConference Communication and Manufacturing pp 12ndash15Hangzhou China July 2014

[45] S Deng L Shen and J Li ldquoMobility-based clustering pro-tocol for wireless sensor networks with mobile nodesrdquo IETWireless Sensor Systems vol 1 no 1 pp 39ndash47 2011

[46] M Khalil A Khalid F U Khan and A Shabbir ldquoA review ofrouting protocol selection for wireless sensor networks insmart citiesrdquo in Proceedings of the 24th Asia-Pacific Confer-ence on Communications (APCC) Ningbo China November2014

[47] O Message Q Telemetry and T Mqtt Committee Specifi-cation Draft 02 April 2014

[48] M Collina M Bartolucci A Vanelli-coralli andG E Corazza ldquoInternet of things application layer protocolanalysis over error and delay prone linksrdquo in Proceedings ofthe 7th Advanced Satellite Multimedia Systems Conference andthe 13th Signal Processing for Space CommunicationsWorkshop Livorno Italy September 2014

[49] A Talaminos-barroso M A Estudillo-valderrama L M RoaJ Reina-tosina and F Ortega-ruiz ldquoA machine-to-machineprotocol benchmark for e health applicationsmdashuse case re-spiratory rehabilitationrdquo Computer Methods and Programs inBiomedicine vol 129 pp 1ndash11 2016

[50] V Goyal O Pandey A Sahai and B Waters ldquoAttribute-based encryption for fine-grained access control of encrypteddatardquo in Proceedings of the 13th ACM Conference on Com-puter and Communications Security pp 89ndash98 AlexandriaVA USA October 2006

[51] M Singh M A Rajan V L Shivraj and P BalamuralidharldquoSecure MQTT for internet of things (IoT)rdquo in Proceedings ofthe Fifth International Conference on Communication Systemsand Network Technologies Gwalior India April 2015

[52] A Stanford-Clark and H L Truong MQTT for Sensor Net-works (MQTT-SN) Protocol Specification httpmqttorgnewwp-contentuploads200906MQTT-SN_spec_v12pdf 2013

[53] RFC ldquoExtensible messaging and presence protocol (XMPP)rdquoin Proceedings of the IETF pp 1ndash9 San Diego CA USAAugust 2005

[54] S Vinoski and I Technologies ldquoQueuing protocolrdquo IEEEInternet Computing vol 10 no 6 pp 87ndash89 2006

[55] J L Fernandes I C Lopes J J P C Rodrigues and S UllahldquoPerformance evaluation of restful web services and AMQPprotocolrdquo in Proceedings of the Fifth International Conferenceon Ubiquitous and Future Networks (ICUFN) pp 2ndash7 DaNang Vietnam July 2013

[56] M Asim ldquoA survey on application layer protocols for internetof things (IoT)rdquo International Journal of Advanced Researchin Computer Science vol 8 no 3 pp 996ndash1000 2017

[57] K An A Gokhale D Schmidt S Tambe P Pazandak andG Pardo-castellote ldquoContent-based filtering discovery pro-tocol (CFDP) scalable and efficient OMG DDS discoveryprotocolrdquo in Proceedings of the 8th ACM InternationalConference on Distributed Event-Based Systems New YorkNY USA July 2014

[58] E Web What is a web service pp 52ndash57 2010[59] M Laine RESTful Web Services for the Internet of ings

pp 2ndash4 2012 httpmediaTkk[60] IETF -e WebSocket protocol pp 1ndash71 2011[61] V Pimentel ldquoCommunicating and displaying real-time data

with websocketrdquo IEEE Internet Computing vol 16 no 4pp 45ndash53 2012

[62] Wikipedia 2019 httpsenwikipediaorgwikiStreaming_Text_Oriented_Messaging_Protocol2019

[63] K Raghavendra and S Nireshwalya ldquoApplication layer se-curity issues and its solutionsrdquo IJCSET vol 2 no 6pp 1266ndash1269 2012

[64] Y Xue B Ramamurthy andM C Vuran ldquoSDRCS a service-differentiated real-time communication scheme for eventsensing in wireless sensor networksrdquo Computer Networksvol 55 no 15 pp 3287ndash3302 2011

[65] C Sonmez S Isik M Y Donmez O D Incel and C ErsoyldquoSUIT a cross layer image transport protocol with fuzzy logicbased congestion control for wireless multimedia sensornetworksrdquo in Proceedings of the 5th International Conferenceon New Technologies Mobility and Security (NTMS) pp 1ndash6IEEE Istanbul Turkey May 2012

[66] T Melodia and I Akyildiz ldquoCross-layer QoS-aware com-munication for ultra wide band wireless multimedia sensornetworksrdquo IEEE Journal on Selected Areas in Communica-tions vol 28 no 5 pp 653ndash663 2010

[67] Q Wang and M A Abu-Rgheff ldquoCross-layer signalling fornext-generation wireless systemsrdquo in Proceedings of the 2003IEEE Wireless Communications and Networking 2003WCNC 2003 vol 2 pp 1084ndash1089 IEEE New Orleans LAUSA March 2003

[68] P Liu Z Tao S Narayanan T Korakis and S PanwarldquoCoopMAC a cooperative MAC for wireless LANsrdquo IEEE

32 Journal of Computer Networks and Communications

Journal on Selected Areas in Communications vol 25 no 2pp 340ndash354 2007

[69] H Shan H T Cheng and W Zhuang ldquoCross-layer co-operative MAC protocol in distributed wireless networksrdquoIEEE Transactions onWireless Communications vol 10 no 8pp 2603ndash2615 2011

[70] F Liu T Korakis Z Tao and S Panwar ldquoAMAC-PHY cross-layer protocol for wireless Ad-Hoc networksrdquo in Proceedingsof the 2008 IEEE Wireless Communications and NetworkingConference pp 1792ndash1797 IEEE Las Vegas NV USA March2008

[71] L Jacob and H R Shamna ldquoEfficient cooperative MAC androuting in wireless networksrdquo Transactions on Networks andCommunications vol 3 no 5 2015

[72] S Rao and K Shama ldquoCross layer protocols for multimediatransmission in wireless networksrdquo International Journal ofComputer Science amp Engineering Survey vol 3 no 3pp 15ndash28 2012

[73] L Han S Park S Kang and H Peter ldquoAn adaptive cross-layer FEC mechanism for video transmission over 80211WLANsrdquo in Proceedings of the 2009 International Conferenceon Internet pp 209ndash215 Las Vegas NV USA December2009

[74] N Li J Martınez and V H Dıaz ldquo-e balanced cross-layerdesign routing algorithm in wireless sensor networks usingfuzzy logicrdquo Sensors vol 15 no 8 pp 19541ndash19559 2015

[75] G Sun J Qi Z Zang and Q Xu ldquoA reliable multipathrouting algorithm with related congestion control scheme inwireless multimedia sensor networksrdquo in Proceedings of the2011 3rd International Conference on Computer Research andDevelopment vol 4 pp 229ndash233 IEEE Shanghai ChinaMarch 2011

[76] M Li Y Jing and C Li ldquoA robust and efficient cross-layeroptimal design in wireless sensor networksrdquoWireless PersonalCommunications vol 72 no 4 pp 1889ndash1902 2013

[77] J Yan M Zhou and Z Ding ldquoRecent advances in energy-efficient routing protocols for wireless sensor networks areviewrdquo IEEE Access vol 4 pp 5673ndash5686 2016

[78] A Ahmad S Ahmad M H Rehmani and N U Hassan ldquoAsurvey on radio resource allocation in cognitive radio sensornetworksrdquo IEEE Communications Surveys amp Tutorials vol 17no 2 pp 888ndash917 2015

Journal of Computer Networks and Communications 33

International Journal of

AerospaceEngineeringHindawiwwwhindawicom Volume 2018

RoboticsJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Active and Passive Electronic Components

VLSI Design

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Shock and Vibration

Hindawiwwwhindawicom Volume 2018

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawiwwwhindawicom

Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Control Scienceand Engineering

Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Journal ofEngineeringVolume 2018

SensorsJournal of

Hindawiwwwhindawicom Volume 2018

International Journal of

RotatingMachinery

Hindawiwwwhindawicom Volume 2018

Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Navigation and Observation

International Journal of

Hindawi

wwwhindawicom Volume 2018

Advances in

Multimedia

Submit your manuscripts atwwwhindawicom

Page 9: ReviewArticle - downloads.hindawi.com · ReviewArticle QoSStrategiesforWirelessMultimediaSensorNetworksinthe ContextofIoTattheMACLayer,ApplicationLayer,and Cross-LayerAlgorithms MuwongeSsajjabbiBernard

a new synchronization scheme When a data packetarrives the current and expected packet arrival timeswith allowable delay (D) are compared by the CN and adrift value (DV) is computed based on the differenceΔT -e DV is sent to the node in the ACK packet forchanging the slot in subsequent transmissions thoughthe value is based on allowable delay and F In caseΔTgtD the CN will send the DV in the SYNC-ACKpacket for subsequent SYNC to nodes or else a mereACK packet is sent by the CN for the received packetFor any forthcoming data transmissions the node willchange its wake-up time plan based on the DV Withsuch a sync approach a sleeping node will not lose syncfor all cycles N Acceptable delay (D) is given as

D min TS1 TSn( 1113857XF

100 (4)

where F is the guard-band factor

ΔT expected arrival time minus current arrival time

DV 0 if |ΔT|ltD

ΔT if |ΔT|gtD1113896

(5)

Frame format the AR-MAC protocol employs twoforms of packets namely (i) data packets in which anode transmits in its assigned time slot and employsCAP for emergency data and (ii) control packets thatinclude channel packet Time Slot Request (TSR)packet Time Slot Request Reply (TSRR) packet Syn-chronization-Acknowledgment (SYNC-ACK) packetData Request (DR) packet and Acknowledgment(ACK) packet [27]Energy consumption if N is the number of cyclesenergy consumption for these is measured as follows

ETotal 1113944N

k1ESleepk

+ 1113944N

k1EActivek

ESleep TSleep x ISleep x V

TSleep Tframe minus Tactive

(6)

where ISleep current from the voltage source V whenthe node is asleep If Esw is the switching energy Etransthe transmission energy Erec the receiving energy andETout the time-out energy then

EActive 2Esw + Etrans + Erec + ETout (7)

Advantages

(i) It reduces energy consumption(ii) It minimizes overhearing and idle listening based

on the needs of the node via periodic sleep andwakeup

217 ER-MAC It is a combination of TDMA and CSMAmethods making it flexible and adaptable to topological andtraffic changes and hence applicable in rapid response

situations inWSNs Collision-free slots can be planned usingTDMA [30] ER-MAC is designed in such a way that thefollowing are met

(a) It has a higher ratio of packet delivery and low delayunder lower energy consumption levels out-performing Z-MAC

(b) It keeps two priority queues in order to distinguishhigh from least priority packets

(c) It permits contention in TDMA slots so as to handleenormous traffic volumes

(d) It has a harmonized slot shape and the nodes areable to locally change their plans enabling themconnect to or exit the network with easeProtocol design it discovers the topology in such away that it uses the simple flooding mechanism tobuild the tree as initiated by the base station (BS) -etopology discovery aims at neighbor discovery andchange tracking in addition to setting up a routingtree A TOPOLOGY_DISCOVERY message com-prising hop_count new_parent_id and old_-parent_id is generated by the BS A node broadcasts itto discover if it has any potential children and aresponse to the parent In case of the desire to changethe parent the former parent is notified For all nodesthe hop count to the BS parent ID children and one-hop neighbor are registered with the BS at this stageTDMA slot assignment when a BS sends a SYNCmessage it switches to the TDMAmode which whenreceived by the child uses the broadcast slot for itschildrenrsquos synchronization Neighboring nodescannot share a slot since slots are assigned andschedules are exchanged accordingly Slot assign-ment begins with a leaf node in a bottom-up modesuch that the flow of messages to the BS may besmooth Before getting the child-node schedules anonleaf node will not transmit its data via a unicastslot descendantrsquos data via many unicast slots andchild synchronization via a broadcast slot Childnodes send SCHEDULE_NOTIFICATION mes-sages to the BS to end this phase

Local time synchronization is done via parent-childbroadcast sync since they both need to operate the same clockso that when one is transmitting the other is ready to receivePacket prioritization happens in such a way that a highpriority queue is first emptied before a lower priority queue isallowed to transmit In case a queue is filled up the shortestslack packet is dropped and then fairness over source isconsidered as queue modification is done to enable the BSbalance info from different nodes Energy conservation is bysender turning off the radio in absence of any info to transmitand the receiver will switch to the sleep mode in case nopackets are received after a certain timeout [30]

Advantages

(i) It is flexible and adaptable to topological and trafficchanges

Journal of Computer Networks and Communications 9

(ii) It is applicable in rapid response situations inWSNs

(iii) It has a higher ratio of packet delivery and lowdelay under lower energy consumption levels

218 RI-MAC (Receiver-Initiated MAC) -is is an asyn-chronous duty-cycle protocol that tries to reduce timeoccupied by the communicating nodes on the medium toagree on time for data exchange-e sending node waits forthe receiving node to signify the beginning of the com-munication session through a short beacon frame that itsends -e channel is only occupied by the beacon and datatransmission reducing channel occupancy and thus en-abling data exchange for other nodes too -is implies animprovement in channel utilization that subsequently leadsto increment in throughput packet delivery ratio andpower efficiency over a wide-ranging traffic load [31] Beinga receiver-initiated model it ensures substantial reductionin overhearing and lowers the probability of colliding andthe cost of recovery compared to B-MAC [23] or X-MAC[32] RI-MAC is further capable of using a beacon withACK and Ready-to-Receive packets though it has aproblem if a sending node with Ready packets needs to havethe radio on till a ready-to-receive node is awake [13] whichthen sends it a beacon and transmission begins immediatelyand later is acknowledged by another beacon Nodes in-termittently wake up to listen to any possible frames meantfor them depending on their schedules A node announcesits readiness to receive a frame through a broadcast beaconon waking up when it switches on the radio When a dataframe has been received the receiver will spend some moretime known as ldquodwell timerdquo in the active state so as toreceive queued packets which is the time in RI-MAC thatdepends on the number of contending senders If there areseveral contending senders that might not have beenpredicted there is a challenge in trying to reduce the re-ceiverrsquos active time to optimize power efficiency and reducethe cost of collision detection and lost data recovery even inthe case of hidden senders For this protocol this can beaddressed by a beacon frame from the receiving node forcoordinating data frame transmissions of contendingsending nodes-e protocol also beats B-MAC and X-MACin minimizing the cost of collision detection and datarecovery through coordination of the receiversrsquo dataframes -is is because the receiving node is aware of themaximum delay prior to the arrival of the frame-e abilityto broadcast is supported by RI-MAC when it sends data asunicast to each senderrsquos neighboring node or continuouslytransmits the frame in a back-to-back mode for as long asthe sleep interval [31]

Advantages

(i) It improves channel utilization leading to in-crement in throughput packet delivery ratio andpower efficiency over a wide-ranging traffic load

(ii) It ensures substantial reduction in overhearing andlowers the probability of colliding and the cost ofrecovery

219 SRI-MAC (Synchronous Receiver-Initiated MAC)-is is a synchronous duty-cycle protocol that implementsthe method of receiver-initiated data transmission with anaim of enhancing network lifetime through energy wastageavoidance via collision overhearing and idle listening To dothis and also minimize duty cycle it uses a sequence ofadaptive beacons and RTSCTS packets In this protocolsending nodes do not go to sleep till they have got CTSpackets from receiving nodes to begin transmitting exceptfor nonsending nodes that return into the sleep mode rightaway as if the channel was idle [33] Since sending nodeshave separate schedules it reduces the contention proba-bility leading to more energy conservation A time frame isdivided into the information period the allocation periodand the communication period which are discussed asfollows

Information period at frame initiation a beacon is sentby the receiving node declaring its return from sleepand readiness to accept data packets in case the channelis not busy -e beacon is a small packet alerting otherlistening nodes of the transmitterrsquos readiness to get arequest for channel It consists of the receiverrsquos ID andthe duration allocation period (DAP) that relies onnumber of the receiving nodersquos neighbors [13]Allocation period here the receiving node stays awake forsome little time after broadcasting its beacon to confirm ifthere is any node having anRTS packet to transmitWhenthe beacon has been received an RTS packet (with 3fieldsmdashsender ID receiver ID and data size) is sent by thesender prompting the receiver to send a CTS for SYNC asthere is neighbor information exchangeCommunication period SRI-MAC is capable of or-ganizing and scheduling transmissions by means ofsensor node IDs -e receiver assigns order andtransmitting time for every sender such that they switchoff their radios till their time to transmit is dueHowever the receiving node remains in the wake-upmode to get all sending nodersquos data SRI-MAC canenable sensors have lengthy sleeping times duringtransmission such that energy can be saved as long as itis neither sending nor receiving data -e protocolrsquoscommunication period is premised on the TDMAtechnique which conserves energy since the radiorsquosduty cycle is lowered and collision is eliminated [33]

Advantages

(i) Since it employs the TDMA principle it eliminatescollisions while maintaining energy conservationtoo

(ii) Noncommunicating nodes switch off their radioand get into the sleep mode which eliminatesoverhearing

(iii) Since receiving nodes broadcast SYNC signals toall nodes in the form of CTS packets overemittingis avoided

(iv) It decreases duty cycling as sleep time increasesand lowers idle listening

10 Journal of Computer Networks and Communications

2110 H-MAC -is protocol is built on the power savingmechanism (PSM) of IEEE 80211 and slotted Aloha pro-tocols to enhance performance using multiple slots Timehere comprises large frames each having an active and asleeping part [34] If any node has got data to send it will askfor slots from the receiver node during active time and latertransmit the packets in the prenegotiated slots Nodes gointo the sleep mode during sleep-time slots if they do nothave data to transmit

Latency in IEEE 80211 if N is the number of hops nthe value for the current hop tcsn the back-off time andttx the transmission delay the total latency is calculatedusing the following equation

D(N) 1113944N

n1tcsn + ttx1113872 1113873 (8)

and the average latency is given by the followingequation

E[D(N)] N tcsn + ttx1113872 1113873 (9)

-roughput in H-MAC is given by the followingequation

Th npnm

tact + Cts (10)

where np is the number of transmitted packets nm is themaximum number of nodes connected in a time frameC is the number of slots with the same length tact is thelistening time and ts is the sleep delayAdvantages

(i) It enhances performance using multiple slots

2111 Z-MAC (ZebraMAC) -is rides on the strengths andoffsets weaknesses of TDMA and CSMA It yields very goodchannel usage and low delay under low contention and lowerscollision among two-hop neighbors at a reduced cost Itsperformance is strong via error synchronization failure inassigning slots and channel conditions that change with timeIn the set-up phase the protocol operates through certainsteps including neighbor discovery slot assignment localframe exchange and global time synchronization whichunless the topology of the network changes significantly donot run again When the data are being transmitted there isimprovement in network throughput and energy efficiencywhich covers up for the initial upfront operational costs [35]

Disadvantages

(i) Initial slot assignment is quite a cumbersome taskhere since it is done at the beginning yet this maynot be at the same time in reality It further as-sumes static links until setup happens again whichtoo is impractical

(ii) It also suffers from the hidden-terminal problemsince slot owners have a small contention windowand priority for slots allocated to them due to theECN messages

(iii) Synchronizing time also poses a challenge in thisprotocol due to a large clock drift between nodes

(iv) Source and destination nodes too are not wellcoordinated in Z-MAC [36]

2112 Q-MAC (QoS-Aware Media Access Control) -isprotocol offers high QoS with improved energy efficiencyand accesses the wireless channel using MACAW as theunderlying protocol [37] Traffic from different nodes isprioritized for QoS satisfaction depending on how criticalthe data are It mainly has two scheduling algorithms inWSNs (i) intranode and (ii) internode scheduling -eintranode scheduling scheme adopts a first-in first-out-(FIFO-) based queuing algorithm based on application andMAC layer abstraction for data packet classification In-ternode scheduling reduces idle listening and collision on thechannel for improved energy consumption

Advantages

(i) It gives a high QoS with improved energy efficiency

2113 WiseMAC (Wireless Sensor MAC) WiseMAC isbased on the Aloha protocol and employs preamble sam-pling in achieving low-power transmissions in infrastructuresensor networks by occasionally sampling the channel tocheck for activity [3238] -e protocol employs the samemethod as B-MAC except that the sender shortens theextended preamblersquos length by scheduling its transmissionsaccordingly after learning the receiverrsquos awake schedules-is is done by the receiver node putting its subsequentawake time on the ACK packet such that when the senderneeds to send to it again it starts the preamble shortly beforeit wakes up so as to avoid energy wastage in transmitting thepreamble [32] -e data frame is repeated instead of theextended preamble for less traffic transmissions that have alengthier preamble than the data frame in WiseMAC Afterprocessing the frame a node that is not the target recipientwill go back into the sleep mode while the recipient staysawake till the session ends thereby sending an ACK packetFor an idle channel the power consumption is rather lowwith this method Nevertheless there is a problem of longwake-up preamble limiting throughput and causing anenormous power consumption overhead in reception [38]All nodes overhearing a transmission bear the overheadtogether with the recipient node -e protocol can addressissues concerning low-power transmission but has nomechanism of nodes adapting to varying traffic patterns[32] To compensate for drift between clock at the accesspoint and on the sensor node there is a need to compute thewake-up preamble duration -e drift varies directly withtime since previous resynchronization and is given by thefollowing equation

TP min 4θl TW( 1113857 (11)

where θ is the time-base quartz frequency tolerance and l isthe interval between two transmissions Increase in trafficvolume naturally reduces overhearing

Journal of Computer Networks and Communications 11

Radiomodel for protocols with lower power transitiondelays between transceiver states and their powerconsumption are modeled Some of the states includeDOZE where the transceiver neither transmits norreceives but is prepared to power on to the receivetransmit state very fast RXwhere the transceiver listensto the channel receives data or demodulates data fromthe noisyidle channel and TX which is the trans-mission state by the transceiver Let TS be the set-uptime needed to turn the transceiver from the DOZE tothe RX or TX state TT the turn-around time needed toswitch between RX and TX states and B the trans-ceiverrsquos bit rate PZ PR and PT are the values of powerconsumed in the DOZE RX and TX statesrespectively

1113954PR PR minus PZ

1113954PT PT minus PZ(12)

where 1113954PR and 1113954PT are the changes in power as a result ofbeing in RX and TX states respectivelyTraffic model let N be the number of sensor nodesconnected to an access point (AP) and λ be the global rateat which the downlink Poisson traffic arrives at the APfrom the fixed network Traffic toward each sensor node isthe same and is given by λN Average packet interarrivaltime Lwith which data packets arrive at a node is given byL Nλ If TC and TD are the control packets and datapackets respectively assuming a small traffic volume theglobal interarrival 1λ far exceeds the summation ofdurations for the data packet and control packet and theturn-around time Mathematically it is given as

1λ≫TD + TC + TT (13)

-e average power consumed by WiseMAC andtransmission delay incurred are given by equations (14)and (15) [38]

PW PZ +1113954PR TS +(1B)( 1113857

TW+

1113954PR X + TD + TT( 1113857

L

+ 1113954PR(N minus 1)Y

L

(14)

where

X 2θL 1 minus eminus TD4θL( )1113874 1113875

Y T2D + 12TDθL

2TW1 minus e

minus TW4θL( )1113874 1113875

DW TD +TW

21 minus e

minus TW4θL( )1113874 1113875

+ 2θL 2 minus eminus TD4θL( ) minus e

minus TW4θL( )1113874 1113875

(15)

where PW and DW are the power consumed byWiseMAC and the transmission delay for WiseMACrespectivelyAdvantages

(i) It achieves low-power transmissions in in-frastructure sensor networks through preamblesampling while checking for activity

(ii) It reduces overhearing due to increased trafficvolume

Disadvantages

(i) It has a problem of long wake-up preamble limitingthroughput and causing enormous power con-sumption overhead

2114 X-MAC It is a simple minimal transmission powerMAC protocol that uses a short preamble with decouplingfor the sending and receiving node sleep plans inWSNs-isleads to a significant reduction in energy usage at sendingand receiving nodes and decrease in per-hop latency andgives flexibility in adapting to bursty and periodic sensordata sources [32]

Protocol design X-MACrsquos design is aimed at increasingenergy efficiency reducing data latency and increasingthroughput -is enables application in different kindsof packetized and bit stream digitized radios as well assimplified low-overhead distributed application -edesign is as such in order to address the low-powerlistening problems including overhearing excessivepreamble and incompatibility with packetizing radiosas in [32]Asynchronous duty cycling a node with information totransmit will send the packets after sending a preamblewhereas the remaining nodes will preserve their sleepschedules that are not synchronized -e receiver willsample the link on waking up and stays awake if it detectsa preamble and if it is the target it will return to sleep if itis not the target after getting the whole preambleEmbedding the target ID in the preamble to avoidoverhearing the lengthy preamble is divided into manyshortened packets embedded with the receiver nodersquosID so as to eliminate overhearing One lengthy pre-amble is constituted by the shortened preamble packetstream Any node on receiving the short preamble afterwaking up will examine the target ID on the packet Itgoes back into the sleepmode right away resuming dutycycling if it is not the targeted recipient or stay awakefor successive packets if it is the one -is approachchecks energy wastage due to network density It canalso be used on several types of radiosReducing excessive preamble using strobing it ispossible to enable low-power transmission and alsosave energy by reducing time for preamble trans-mission through the use of enhanced preambles inaddition to preamble sampling X-MAC does not sendan endless preamble packet stream but puts slight

12 Journal of Computer Networks and Communications

pauses in between packets to allow the sending node topause and listen to the channel and allow the receivingnode to resend an early ACK packet during that pause-is makes the sending node begin sending datapackets and the receiving node cut excessive preamblesand thus decrease per-hop latency and energy wastageon unnecessary waiting and transmissionPacketizing radios low-power listening is limited incapability for supporting packetizing radios butX-MACrsquos short strobed preambles can support all typesof digital radiosAdaptation to traffic load while a number of WSNapplications are reliable in traffic production they stillmust adapt to changing traffic Nodes are going to haveseparate sleep schedules depending on traffic loadConsidering systems that have time-fluctuating trafficall predetermined static schedules are suboptimalOptimality Buettner et al [32] model the expectationof energy as follows

Es (preamble energy + energy perACK listen)

lowast(expected preamble minus listen iterations)

+(energy to send parket)

Es PTxSp + PRxSal1113872 1113873 Rl + Rs( 1113857

Rl minus Sp+ SdPTx

(16)

where Es expected energy to send a packet PTx andPRx are the transmission and receiver power re-spectively Sp Sal and Sd are the duration of thesenderrsquos preamble ACK listen and data transmissionperiods respectively and Rl and Rs are the receiverrsquoslisten and sleep periods respectively Expected energyto receive a packet is given as

Er (listen cycle energy + sleep cycle energy)

lowast(expected iterations for preamble arrival)

+(energy to sendACK) +(energy to receive packet)

Er PsRs + PRxRl( 1113857

1 minus 1 minus Pd(t)( 1113857 Rl + Rs( 1113857+ RaPTx + RdPRx

(17)

And the one-hop expected latency is given as

Lat (preamble duration + ACK listen)

lowast(expected number of iterations)

+ duration to send packet

Lat Sp + Sal1113872 1113873 Rl + Rs( 1113857

Rl minus Sp

(18)

Advantages

(i) It increases energy efficiency(ii) It reduces data latency(iii) It increases throughput

2115 A-MAC (Advanced MAC) A-MAC is a TDMA-based protocol whose design is aimed at enhancing thelifetime of the network for low rate and reliability of datatransmission [39] To assign time slots it does not depend ona central controller like most protocols but it gathers in-formation from its neighbors so as to allocate the slotsthrough a distributed algorithm Energy is controlled byscheduling power when there is not a single event Its designcomprises many frames with each having many time slotswhich at the beginning have got a beacon transmitted tosynchronize and exchange information with neighborsNodes to participate in the subsequent session are selectedby the controlled node so they do not go into the sleep modewith others [13] -is protocol operates under four time slotdivisions namely initial wait discover and active states Anode in the initial state begins by listening to the medium forthe beacon packet from other nodes for network synchro-nization which when received makes it change the timerthrough subtraction of beacon transmission time from re-ception time To avoid synchronization problems due todrift and allow for continuity the node has to get thestrongest beacon signal It then enters the wait state afterchoosing a number of waiting frames so as to reduce theprobability that nodes get into the discover state simulta-neously In case the synchronization beacon is lost while stillin the wait state it will return to the initial state for anotherone On expiry of the waiting counter it enters the discoverstate and collects information from neighbors by listening totheir beacon messages On successful selection of a time slotit enters the active state where it endlessly transmits beaconsat the initiation of the slot A node goes to sleep when abeacon has been transmitted and there are no more data fortransmission or in case the neighborrsquos beacon received in-dicates no data packets are coming in [39]

Advantages

(i) It can enable multicasting(ii) It enhances the lifetime of the network(iii) It offers reliable data transmission

2116 L-MAC (Lightweight MAC) Based on TDMA thisprotocol is aimed at minimizing the number of transceiverswitches adapting the nodal sleep interval to data trafficamount and making implementation less complex [40] Apacket is transmitted comprising a control and a data partControl packets have a fixed size and the time slot con-trollerrsquos ID -ey show the distance from the transmittingnode for simplified routing to the network gateway as well asaddressing target node and reporting data unit length andinternode synchronization Adjacent nodes are more in-terested in getting control packets from nearby nodes -enodersquos transceivers are switched off until the subsequenttime slot unless the node is also addressed or the message isomnicast If so it listens to the data unit which may not fillthe remaining time slot in totality-e transceivers on eitherside are turned off when the transfer of packets is successfulEnergy wastage through idle listening in uncontrolled timeslots is avoided via a short time-out interval Network nodes

Journal of Computer Networks and Communications 13

are able to communicate without colliding-is saves energyand increases the lifetime of the network A node in L-MACcan only transmit one message per frame [13] -e networkis set up in such a way that all nodes are not synchronized onbeing switched on-e gateway takes control of a timeslot soas to have them synchronized and its immediate neighborgets the control messages thereby synchronizing their timerswith the gatewayrsquos Neighboring nodes learn of multiplegatewaysrsquo time slots in their vicinity just after one frame-ejust synchronized nodes randomly choose time slots forcontrolling as long as they are not occupied -e nodes thuspreserve a table for the neighboring nodes which makes atimeslot reusable after a minimum of 3 hops and preventsmessage collision All nodes keep their slots before thebattery is down or learn of an impending collision All nodescan track their hop distances to a selected gateway node andthe information is communicated in control messages [40]

Advantages

(i) It minimizes the number of transceiver switches(ii) It adapts the nodal sleep interval to data traffic

amount(ii) It makes implementation less complex(iii) It offers improved network lifetime(iv) Energy wastage through idle listening in un-

controlled time slots is avoided via a short time-out interval

Disadvantages

(i) L-MAC is not very good for mobile SNs since slotsare calculated only once

(ii) It suffers idle listening because nodes keep listeningto slotsrsquo control parts so as to get data and enableother nodes to join the network all the time [13]

2117 C-MAC -is was designed for WSNs with 3 seg-ments namely (i) aggressive RTS having double channelcheck to assess the channel (ii) anycast to initialize the flowand (iii) convergent packet forwarding to stabilize the flowIt promises an improvement in energy efficiency lowerlatency and enhanced throughput while avoiding syn-chronization overhead though it operates only in low duty-cycling apps especially where there is less traffic But whenthere is more traffic the protocol employs anycast-basedpacket forwarding for waking the nodes up or finding aforwarder quickly and converges from route-suboptimal toroute-optimal unicast After this energy could be preservedby the nodes using a synchronized wake-up plan [13]

Advantages

(i) It improves energy efficiency(ii) It lowers latency(iii) It enhances throughput avoiding synchronization

overhead

Disadvantages

(i) It operates only in low duty-cycling applicationsespecially where there is less traffic

22 MAC Protocols for WMSNs

221 Diff-MAC -is is a CSMACA-based QoS-awarehybrid-priority-based MAC protocol for WMSNs It targetsenhancing the channel utilization while making use of ef-ficient service differentiation to do coordination for mediumaccess of each traffic class while giving fair quick datatransfer It offers QoS using the following features [14]

(i) It balances energy consumption and delay throughsensor node duty-cycling adaptation depending onthe dominant traffic class

(ii) It has a feature for fragmentation and messagepassing that breaks lengthy video frames intosmaller packets that are transmitted as a burst thusreducing retransmission costs if there are any MACfailures

(iii) -ere is fair data delivery amongst the nodes andtraffic classes respectively resulting from intranodeand intraqueue prioritization reducing unbearableperformance

(iv) -e number of collisions can be reduced and packetlatencies are kept within limits by adjusting the sizeof the congestion window in this protocol as per thetraffic needsAdvantages

(i) Diff-MAC is very suitable for WMSNs since itadapts quickly to changes in networkconditions

(ii) It is a fair protocol(iii) It is scalable

Disadvantages

(i) It is quite difficult to monitor the statistics ofthe network

(ii) Adapting dynamically to the network condi-tions is not an easy task

(iii) It exhibits some degree of latency in packetdelivery

-e implementation of Diff-MAC on Imote2 indicatesthat resource requirements can be met on the sensorhardware Results of extensive simulation runs showed thatDiff-MAC outperforms its competitors

222 Cluster-Based On-Demand Multichannel MAC Pro-tocol for WMSN (COM-MAC) -is is an energy-efficientprotocol in WMSNs with high throughput and reliable datatransmission It achieves energy efficiency by schedulingmultichannel media access to reduce channel contentionand remove collision idle listening and overhearing [41]

Design it has a three-phased operation with the (i)request phase (ii) scheduling phase and (iii) datatransmission phaseRequest phase here all nodes have two protocols eachfor sending requests to the cluster head (CH) ie thecontention-based and contention-free TDMAFDMA

14 Journal of Computer Networks and Communications

protocols -e contention-based protocol allocates achannel per node for sending the request message andfree channels are useable as control channels Forcontention-free protocols a control slot is allocatedwhen a network is deployed Here a slot per channel isallocated to a different node enabling requests to be sentwithout interfering with transmissions from othernodes -e contention-based protocol operates undertwo steps (i) Control channel assignment phase achannel is assigned per node for transmitting a request(REQ) message A channel is assigned to multiplenodes and nodes are evenly distributed to availablechannels for congestion control (ii) Request trans-mission phase nodes with data send a REQmessage viathe assigned control channel to notify the CH whichresponds with an ACK message -e contention-freeTDMAFDMA protocol too has two phases includingthe control slot assignment and request transmissionphases For unreliable channel conditions and lowtraffic load the contention-based protocol is fit for usebecause of less collisions and congestion whereas thecontention-free protocol is preferable in heavy trafficload and reliable channel applicationsScheduling phase from the above phase a schedule isgenerated by the CH for coordination and this enablesnodes to transmit and broadcast their data via controlchannels-e schedule for broadcasting only comprisesdata for nodes assigned to a channel -e schedule isrebroadcast for transmission reliability enhancementbut for energy efficiency the nodersquos transceiver isswitched off when all the schedule is completely de-livered till it is the time for transmitting again-e timeslot and data communication radio channel are allincluded in the schedule generated for each nodeData transmission phase on getting the schedule thesensor node sends its data as per the time slot and thechannel assigned -e time slot is divided into datatransmission andACK sections ACK supports link layererror control Using implicit selective repeat ARQ theCH sends an ACK message for all received packets Ifthere is no ACK the packet is assumed lost and hence aretransmission during the subsequent interval is done-is enhances reliability of transmissions though excessdelay is possible here which is also intolerable for real-time critical apps like video or audio -is gives rise to ahybrid MAC protocol to exploit the unused spectrumduring data transmission sessions [41]Advantages

(i) It achieves energy efficiency by scheduling mul-tichannel media access to reduce channel con-tention and remove collision idle listening andoverhearing

(ii) It maximizes network throughput by dynamicallyallocating time slots and channels for nodes byexecution of a traffic-adaptive QoS-aware sched-uling algorithm depending on needs for QoS andtraffic data information

(iii) It increases transmission reliability by in-corporating spectrum-aware ARQ to avoid spec-trum wastage

223 EQoSA -is protocol is developed for video andimage transfer across WMSNs in a hybrid mode withprovision of QoS Concerning the active sensor nodes andtheir traffic loads it employs dynamic session sizes andchanges bitmap-assisted (BMA) MACrsquos fixed session sizeNodes declare any data available for transmission duringcontention time -is forces the cluster head to make aschedule of how slots have been assigned It communicatesthis schedule to the nodes giving each node a given numberof slots per session and so enabling the busty traffic to beaccommodated [14] -e protocol does not perform wellwith DiffServe mechanisms but does well with schemessupporting busty traffic loads and strong cluster heads arenecessary in WMSNs that can assign and broadcast slots

224 QoS-Supported Energy-Efficient MAC (QE-MAC)Based on the IEEE 80211e standard this protocol ensuresfairness and has low latency and jitter and better efficiencyon energy usage than other QoS-aware MAC protocols forWMSNs It operates under a two-phased routine

-e first phase introduces the innovating prioritymechanism in IEEE 80211e QoS and several data types areassigned priorities Its coordination function is a combi-nation of distributed contention-based channel allocationand centralized polling-based channel access mechanisms

-e second phase preserves the nodesrsquo energy usingdynamic duty cycling -ere is an exchange of RTSCTSpackets to supplement CSMACA by muting nodes near thesenderreceiver during packet duration [42]

Advantages

(i) It ensures fairness(ii) It has low latency and jitter and better efficiency on

energy usage than other QoS-awareMAC protocolsfor WMSNs

225 Black-Burst (BB) Contention A contention-basedMAC scheme for enhancing instantaneous QoS access tocarrier-sense WSNs is proposed in [42] In this schemenodes contend for the channel until it is free Real-timepackets are given higher priority over non-real-time packetsand if a node is transmitting real-time packets it is prior-itized over other nodes [15] Voice and video are some of thereal-time apps in consideration here which periodicallyaccess the common radio channel when transmitting -eseneed constrained end-to-end delay of packets at the MAClayer -is protocol majorly does the following [42]

(i) Prioritization of real-time traffic(ii) Implementation of the round-robin mechanism in

real-time nodes(iii) Bounded access delays among real-time packets

Journal of Computer Networks and Communications 15

(iv) Implementation of real-time communication withvarying needs of bandwidth good for multimediatraffic

Amechanism for dynamically assigning packet prioritiesbased on their deadlines and the traversed hops is proposedin [43] -e ReAllocative Priority (ReAP) mechanism ismeant to provide QoS at the MAC layer to video traffic -isscheme introduces adaptive TXOP by modifying the TXOPdynamically based on the queue length [15] Other protocolsin this category include dynamic duty cycle and adaptivecontention window-based MAC protocol frame sharing(FRASH) MAC protocol and real-time independentchannels (RICH) MAC protocol

Advantages

(i) It can implement real-time communication withvarying needs of bandwidth

23 Challenges and Opportunities(a) One of the challenges in satellite transmission is

cross-talk leading to interference in different fre-quency bands It is necessary to increase the rise andfall times as well as protect sensitive nodes as wecircumvent floating ones if we are to regulate cross-talk as in satellite communication therefore Wefurther need to investigate the possibility of usingdifferential signaling -is constitutes an opportu-nity for further research

(b) Another challenge is that the hidden and exposedterminal problems are still noticeable despite the useof RTSCTS solutions in MAC since collisions arestill possible from transmissions of separate nodesAt the MAC layer ACKs might block efficientchannel usage by prohibiting exposed terminalsfrom reusing the channel and may be eliminatedleaving no proof of sender data packet receptionUsing a directional antenna is one of the proposedtechniques of high-speed WSNs like IEEE 80211-is constitutes an opportunity for further study intothese issues for better solutions though

(c) Another challenge is that since nodes in contention-based MAC protocols are mobile there are manycollisions by the packets and schedule in-consistencies in schedule-based protocols in the two-hop neighborhood information when nodes areentering or leaving [44] Deng et al [45] propose amobility-based clustering (MBC) protocol for mo-bile nodes to enhance the packet delivery rate -isprotocol in comparison with CBR has 25 chancesof decreasing collision-induced packet losses and50 chances in comparison with low-energyadaptive clustering hierarchy (LEACH)

(d) -e process of analyzing and evaluating routingperformance for WMSNs is quite complex becauseof the bursty multimedia traffic Because of the wayencoding scheme frames are structured as well as thechanges in between the scenes that occur naturally

the compressed video frames display some substantialburstiness on several time scales -ere will also befrequent route recalculation that is as well energy in-tensive for nodes frequently capturing and transmittingenergy -e solution here could as well possibly be theuse of the MBC protocol Route recalculation tooshould be dynamic to cater for the mobile nodes whichmight be in and out of the range at different instants oftime Recently the use of smart dumpsters for effectivewaste management in smart cities has also been pro-posed as a possible solution [46]

(e) For WSNs their routing protocols do not look atmultimedia applications which require enormousbandwidth and processing energy and should betransmitted in real time with utmost fidelity Extrachallenges in routing are imposed in WMSNs indeployment of heterogeneous sensors since therecould be audio video and still pictures in addition toscalar data during transmission Heterogeneous datahandled also have different QoS requirements andwith many different business needs different ser-vices with varying requirements give rise to a bigchallenge for the routing design -ere is a need todesign protocols that can support WMSNs such asEQoSA [14] for this purpose

Table 2 summarizes the reviewed MAC protocols forWSNs and the ones designed for suitability in WMSNs

3 QoS Strategies at the Application Layer

We discuss many protocols used to solve different needs ofcommunication between machines

31 ConstrainedApplication Protocol (CoAP) It is a requestresponse protocol designed by the Internet Engineering TaskForce (IETF) for constrained devices such as those havinginsufficient RAM or CPU and WPANs with low-powerconstraints resulting in poor transfer of packets andenormous overhead -e protocol was designed for ma-chine-to-machine (M2M) apps and system automation toreduce bandwidth requirements increase packet transferdecrease the overhead and make work simpler leading tolightweight implementation HTTP commands like GETPOST PUT and DELETE are used in a client-server ar-chitecture Using the publishsubscribe architecture CoAPcan support multiple users and hence yield better perfor-mance via multicasting enabling asynchronous messageexchange -is protocol can support both multicasting andunicasting via UDP and hence is a reliable and simpleprotocol -is is done via the messaging sublayer (for re-liability based on UDP) and the requestresponse sublayer(for communication and interaction) [16]

It uses the following messages

(i) Confirmable message this guarantees reliabilityusing ACK messages sent by the receiver node-ese could be synchronous or asynchronous ifthere is a need for more computational time

16 Journal of Computer Networks and Communications

(ii) Nonconfirmable message here no ACK messagesare sent back

(iii) Acknowledgment (ACK) message this confirmsreceipt of a confirmable message

(iv) Reset message this confirms receipt of an un-processed message in case any critical parts nec-essary for interpreting the message are missing

(v) Piggybacked response the receiver directly re-sponds while getting the ACK message

(vi) Separate response the reply by the receiver is dis-crete from the ACK message

CoAP has no in-built security features It is secured bythe datagram transport layer security (DTLS) that runs ontop of UDP though it was not designed for the IoT -ismakes its suitability debatable since DTLS offers no supportfor multicasting It also needs more packets for handshakingthereby increasing the traffic volume and computationalresources and so shortening the lifespan of mobile devicessince they use batteries [17]

Advantages

(i) It can support multiple users and hence yieldbetter performance via multicasting

Table 2 Comparison of MAC layer protocols used in QoS provisioning

Protocol Features QoS parameter Comments Access Priorityassignment Energy eff Designed for

S-MAC[22] Collision avoidance Energy latency Requires periodic

synchronization CSMA Hybrid Medium WSNs

B-MAC[23] Collision avoidance Energy delay

throughput Flexible interface CSMA Dynamic High WSNs

T-MAC[25] Collision avoidance Energy Complex nonscalable CSMA Dynamic High WSNs

C-MAC[13]

Dynamic channelallocation Energy Multichannel MAC CSMA Dynamic Medium WSNs

CU-MAC[24]

Multichannelmechanism

-roughputlatency High packet delivery ratio TDMA Dynamic mdash WSNs

A-MAC[39]

Per flow serviceguarantees Delay m-ary tree model

dynamic prioritiesTDMACSMA Dynamic Medium WSNs

ER-MAC[30]

Emergencyresponse

Energy delaydelivery ratio

Adapts to traffic andtopology changes

TDMACSMA Hybrid Higher than Z-

MAC WSNs

RI-MAC[31] Collision detection Delay packet

delivery ratioHigh packet delivery ratio

and throughput CSMA Dynamic Medium WSNs

SRI-MAC[33] Collision avoidance Energy delay Good data transfer and

energy consumption TDMA Dynamic High WSNs

H-MAC[34] Slotted sleep time -roughput

latencyData transmission only in

sleep timeCSMAAloha Dynamic Medium WSNs

Z-MAC[36] Collision avoidance Energy delay

throughputVery good channel usage

and error syncCSMATDMA Dynamic Medium WSNs

Q-MAC[37]

Intrainternodescheduling Energy High QoS improved

energy efficiency CSMA Hybrid High WSNs

X-MAC[32]

Overhearingavoidance

Energy delaythroughput

Reduces excessivepreamble by strobing CSMA Adaptive

static Medium WSNs

L-MAC[40] Collision avoidance Network

lifetime energyReduces the number oftransceiver switches TDMA Static Medium WSNs

WiseMAC[38]

No downlinkchannel collision Power delay Significantly low-power

consumptionAlohaCSMA Static Higher than

ZigbeeInfrastructure

WSNsAR-MAC[27] Collision avoidance Energy delay Uses data packets and

control packets TDMA Static Higher thanIEEE 802154 WBANs

BB-MAC[42]

Real-time trafficprioritization Delay Employs round robin in

real-time nodesCSMATDMA Dynamic High WMSNs

COM-MAC [41] Collision avoidance Packet delay

throughputBalances reliability and

retransmissionTDMAFDMA Dynamic High WMSNs

EQoSA[14]

Busty traffic loadaccommodation Energy Provides QoS support for

video and imagesTDMACDMA mdash High WMSNs

QE-MAC[42] Collision avoidance Energy jitter

latencyRuns central polling-like

channel access CSMA Dynamic High WMSNs

R-MAC[26] Collision avoidance Energy delay Reservation based mdash Static High Underwater

SNs

Journal of Computer Networks and Communications 17

(ii) It enables asynchronous message exchange(iii) It is a reliable and simple protocol

32 Message Queue Telemetry Transport (MQTT)Developed by IBM and later improved by OASIS with anintention of decreasing bandwidth MQTT is a simple open-source publishsubscribe protocol analogous to the client-server version It is appropriate for lightweight IoT appli-cations and M2M transmissions in constrained situationslike limited power computation capability and memory orbandwidth -e network bandwidth and the computationalresource usage are decreasing since clients are not updatedmaking publishsubscribe protocols outperform requestresponse protocols via IoT requirements [16] -is protocolcan run over TCPIP and so has the following advantages

(i) Delivering of packets with reliable guarantee(ii) Enabling multicasting(iii) Remote device session establishment(iv) Reduction of transport overhead and protocol ex-

change to minimize network traffic(v) Provision of the notification mechanism when

anomalies occur [47]

MQTTrsquos design targets economical use of bandwidthand battery so it has a major application in FacebookMessenger MQTT ensures QoS and reliability of messagedelivery by providing the following three QoS levels

(1) Fire and forget a message is sent once depending onthe networkrsquos best effort and there is no need forACK (lowest QoS level)

(2) Delivered at least once a message is sent at least onceand an ACK message is needed to avoid duplicates

(3) Delivered exactly once by a four-way handshake itensures a message is sent exactly one time (highestQoS level)-e architecture it has three key components (i)publishersmdashlightweight nodes sending data to thebroker before returning to sleep (ii) broker whichorganizes sensory data in topics and sends them tointerested subscribers and (iii) subscribersmdashIoTapps interested in data sent by sensors through thebroker as shown in Figure 3

MQTT is a simple lightweight protocol that is betterthan CoAP via low and high packet loss probability interms of throughput and exhibits lower latency in lowsampling rates [48] It supports a number of imple-mentations for embedded devices and mobile applicationsand offers web-socket support too Nevertheless it hassome shortcomings

(i) It is only employed in simple data types(ii) It exhibits high latency in high sampling rates(iii) It is unsuitable for real-time apps in such cases(iv) It does not provide automatic discovery andor

sufficient security at the protocol level [49]

33 Secure Message Queue Telemetry Transport (SMQTT)-is is an extension of MQTT that is based on lightweightattribute-based encryption using elliptic curve cryptogra-phy -e protocol has three entities as shown in Figure 4(i) publisher that publishes information under a certaintopic (ii) subscriber that obtains the same topic in-formation via a broker and (iii) public key generator(PKG) or broker that acts as the trusted third party -eprotocol has four phases namely (a) set-up phase (b)encrypt phase (c) publish phase and (d) decrypt phase

(a) Set-up phase publisher and subscriber provide aunique ID with attributes for registration with thePKG -is generates a secret master key set andpublic parameters that it publishes with set U A1A2 An For ciphertext policy-attribute-basedencryption (CP-ABE) the PKG gives a key set todevices with attributes [50]

(b) Encrypt phase based on the access tree the publishercreates an access policy An access tree is sent by thepublisher to the PKG for generating the key-gen-eration policy in the key policy- (KP-) based ABEalgorithm -e publishers know beforehand thetopics and subscribers to access as well as the rulesfor access Subscribers are issued with private keysfor all communication sessions Publishers in theCP-ABE algorithm come up with access trees andrules where the payload is encrypted by the pub-lisher who also issues more info needed for de-cryption and the rules With the help of public keycryptography data are encrypted giving rise to ci-phertext on the basis of KPCP-ABE and the receivergets SUBACK from the broker [51]

(c) Publish phase ciphertext is embedded in theSPublish command as payload and topic name isdefined in the header by the publisher -e brokerreceives the SPublish packet and replies withPUBACK prompting the publisher to respond withthe PUBREL packet -e broker then broadcasts totopic subscribers and deletes the data forwarding thePUBCOMP packet to the sender node

(d) Decrypt phase the encrypted information isdecrypted by the subscriber with the help of a secretkey Considering KP-ABE on verification that itsatisfies the access policy the subscriber invokes thePKG for a corresponding key which it sends afterverifying the request With CP-ABE subscribers usethe private key for decryption after authenticationthereby enabling offline interaction [51]Advantages

(i) -e information being encrypted makes it moresecure

34 MQTT for Sensor Networks (MQTT-SN) -is is a ver-sion of MQTT designed to operate under environments oflow bandwidth high link failures and short message lengthIt is a publishsubscribe (pubsub) protocol designed for

18 Journal of Computer Networks and Communications

UDP-based WSNs to visualize communication for low-power devices [5152] It comprises a publisher device actingas the MQTT-SN client to forward communications to agateway that also changes the MQTT-SN message to theMQTTmessage before delivering it to the broker that sendsit to MQTT-SN subscribers [51]

It has a gateway for changing protocols from MQTT toMQTT-SN (Figure 5) [52] -is protocol supports nodes in

the sleeping mode using an offline keep-alive procedure thatcan be used by battery-operated devices to get into thesleeping mode To support sleeping clients MQTT-SN has anew offline keep-alive procedure with which battery-oper-ated devices can go to the sleeping state in which theirmessages are buffered at the server until they are awake InMQTT-SN clients with no preconfigured gateway addresscan figure out the real address of their operational gateway

Publisher

Subscriber

Subscriber

Broker

Publisher

4G

Figure 3 Architecture of the MQTT protocol

Register by mentioning URI Register by mentioning URI

Access tree

Set-up

Publisher ReceiverPKG(broker)

Key management Key management

Encrypt data using public parametersand random key This is done using

CP-ABE Publish it as

Subscribe (topics)SUBACK

S-Publish (TopicEncrypt (data))S-Publish (TopicEncrypt (data))

Encryptand publish

PUBACK

PUBACK

PUBRELPUBRELPUBCOMP

PUBCOMP

Figure 4 SMQTT protocol

Journal of Computer Networks and Communications 19

within the network through a discovery procedure Simul-taneously we may have several gateways in a single wirelessnetwork that can be cooperative via load sharing

MQTT-SN architecture it is composed of MQTT-SN clients gateways and forwarders (Figure 5)rough the MQTT-SN gateway (GW) the MQTT-SN protocol is used by the MQTT-SN clients toconnect to the MQTT server Integration betweenthe MQTT-SN GW and the MQTT-SN server mightbe possible or not Translating between the MQTT-SN and the MQTT for a stand-alone GW is doneusing the MQTT-SN protocol If a GW is detachedfrom the network it can be accessed by the MQTT-SN clients through a forwarder is just encapsu-lates the received MQTT-SN frames that are on thewireless side before forwarding them to the GW un-altered It then decapsulates the received frames tooresending them to the client as sent by the GW [52]ere are 2 types of gateways (a) transparent gatewayand (b) aggregating gateway as shown in Figure 6

(i) Transparent gateway this launches and sustains adirect MQTT connection between the MQTT-SNclient and the MQTT server which is entirely re-served for end-to-end client-server messaging enumber of servers and that of MQTT-SN clientsconnected to the GW are the same is transparent

GW translates syntax between these protocols and iseasier to implement compared to the aggregatingGW But the MQTT server must support dierentsessions for all connected clients Some MQTTservers are congured to enable few parallel sessions

(ii) Aggregating gateway this maintains a single MQTTconnection to the server on behalf of all clients Allexchanged messages from the MQTT-SN client to theaggregating GW are directed to the GW which sortsthe information to forward to the server e aggre-gating GW decreases the number of MQTT sessionssimultaneously supported by the server in WSNs withmany battery-operated sensors and actuators (SAs)having limited storage and processing capabilities It ishowever more complex to implement compared to atransparent GWe general format of the MQTT-SNprotocol message has two or four message headeroctets and n octets of variable header that is optionaldepending on the message in question MQTT-SN has256 message types (MsgType) with those from0X1Eminus 0XFD 0X19 and 0XFF reserved for futureusage [51]

Advantages

(i) e protocol can operate under environmentsof low bandwidth high link failures and shortmessage length

MQTT-SN gateway

MQTT-SN

MQTT

MQTT-SN

MQTT-SN

MQTT-SN

MQTT-SN client

MQTT-SN clients

MQTT broker

MQTT-SN client

MQTT-SN client

MQTT-SN gateway

MQTT-SN forwarder

Figure 5 MQTT-SN architecture

20 Journal of Computer Networks and Communications

(ii) It is a pubsub protocol designed for UDP-basedWSNs to visualize communication for low-power devices

(iii) It can be optimal for low-cost battery-operatednodes having low processing power and lessstorage

(iv) It supports nodes in the sleeping mode using theobrvbarine keep-alive procedure used by battery-operated devices

35e Extensible Messaging and Presence Protocol (XMPP)e standardization of this protocol by the IETF made itpopular for usage over the Internet especially in instantmessaging and chatting platforms It is however susceptibleto spamming attacks and its failure to have worldwidesupport has made Google too withdraw its support of thestandard [17] e protocol is mainly useable in the client-server architecture but can also be applicable in client-clientXMPP communication [53] Among the application layerprotocols XMPP is the only one that supports both asyn-chronous (publishsubscribe) and synchronous (requestresponse) services while running over TCP but can still runover other protocols

Given the simplicity of its implementation and ease ofusability in XMPP instant messaging (IM) is suitable forXMPP and is implemented using the message stanzaltmessagegt ese stanzas are usable in one-to-one ormultiuser sessions or chatting or in error reporting andsending headlines Attribute type of the stanza denes themessage type e attribute type could be a chat group chatheadline normal or error An instant messaging network inXMPP can support many XMPP clients and servers with agateway congured to enable other IM network users toconnect to other XMPP users e XMPP server mainlycontrols sessions and XML streams to clients and servers inaddition to routing XML stanzas in the streams [53] eTCP ports 5269 and 5222 have been respectively registeredfor the XMPP server-to-server connections and XMPP clientconnections by the Internet Assigned Numbers Authority(IANA) XMPP enables specication of XMPP ExtensionProtocols (XEP) to enhance its usefulness [17]

XMPP security clients connect to the server usingtransport layer security (TLS) and simple authentication andsecurity layer (SASL) protocols which are at the same timeused by the servers for interdomain communications Fordata condentiality and integrity TLS will encrypt thestreams of XML for all sessions To conrm the identity of anyclient that needs to access the server the SASL protocol doesthe authentication of the XML stream For any communi-cating session with a server to start the client needs to resolvethe serverrsquos DNS hostname For protection of the server of theclient from third-party attacks the server maintains the au-thentication details secret ie IP address and access methodand only original connections of the server are required

In server-to-server communications the SASL protocolis employed for authentication and condentiality Server-to-server communications may as well be deactivated from aserver by an administrator depending on the organizationalpolicy To guard against domain-spoong attacks serverdialback can be a good solution since servers support itsecuring more the XML stanzas Nevertheless it cannot beused to authenticate secure or encrypt in-server streamsand the server identication resulting therein is not strongenough to be relied on It further cannot secure against DNSpoisoning attacks and IP session hijacking for remote do-mains [53] High-security domains can use TLS and SASL

Advantages

(i) It is popular for usage over the Internet especially ininstant messaging and chatting platforms

Disadvantages

(i) It is susceptible to spamming attacks(ii) It has failed to have worldwide support making

Google withdraw support of the standard(iii) It is not secure against DNS poisoning attacks and

IP session hijacking for remote domains

36 Advanced Message Queuing Protocol (AMQP) isprotocol can be utilized by a number of platforms formessage exchange since it is open source and written inmany languages It is a publishsubscribe model depending

MQTT-SNclients Broker

MQTT

(a)

MQTT-SNclients Broker

(b)

Figure 6 (a) Transparent gateway and (b) aggregating gateway

Journal of Computer Networks and Communications 21

on a reliable and efficient messaging queue with capacity forinstant messaging or e-mail for different applications es-pecially in IoT environments It differs from others in thesense that message types can be specified and have theirsource traceable and how to do trade-off between perfor-mance security and reliability For reliability of messagetransmission AMQPmakes use of some delivery guaranteeslike at most once at least once and exactly once It can alsohowever utilize the TCP transport layer for guaranteeingreliability -e AMQPrsquos publishsubscribe method has twocomponents namely (i) exchange queue meant for routingof messages to an appropriate order in the queue and (ii)message queue that acts as storage pending their delivery toreceivers [16] In this protocol model the broker can makerouting decisions contrary to othermessaging systems whereapps using the queue have in them embedded the decision-making logic for recipients or senders -e developer mayneed to check each of the affected apps so as to change thelogic for routing and delivery of the messages [54]

Advantages

(i) It is convenient(ii) It has capacity for interoperability between dif-

ferent vendors(iii) It can use open standards to enable connection

between business partners(iv) It can enable innovations based on AMQP foun-

dations [55](v) It can enable supporting of mission-critical ap-

plications in e-commerce(vi) It can guarantee reliable delivery of messages

Disadvantages

(i) It is not very appropriate for real-timeapplications

(ii) It may not offer automatic discovery(iii) Interoperability is not guaranteed(iv) It lacks open-source libraries for constrained

gadgets [49]

37 Data Distribution Service (DDS) A product of theObject Management Group (OMG) DDS is a pubsubprotocol meant for M2M communications with two sub-layers -ese are the data-centric pubsub sublayer re-sponsible for connecting anonymous data publishers tosubscribers and the data-local reconstruction sublayer that isoptional and integrates DDS into the application layer [56]-e DDS architecture connects system components asshown in Figure 7 Publishers and subscribers are discon-nected with respect to (i) time (ii) space (iii) flow andbehavior (iv) platforms and (v) programming languages[57] A DDS data publisher generates topics and publisher-subscriber coupling is represented as topic name data typeschema and publishers and subscribersrsquo QoS attributesDelivering of data is done by the publisher layer -e datawriter and publishers together make a decision about anynecessary alterations for subsequent forwarding to thesubscribers who need to send data to IoT applications Data

readers are responsible for delivering topics to the sub-scribers after reading through them [56]

Some of the major components of the DDS architectureinclude the following [57]

(i) Topic this is a logical path specifying publicationand subscriptionrsquos data type between DataWritersand DataReaders For a successful session to begintopic names types and DataWriters and Data-Readersrsquo QoS should correspond

(ii) Domain this is a logical setting for transmissionthat can be utilized for separating and optimizingnetwork transmission in a setup of similar appli-cations If DDS apps share a domain ID then theycan exchange data among themselves

(iii) DataWriter and DataReader these are terminalobjects for writing and reading messages re-spectively to and from a universal data space

(iv) Participant this is an object that stands for apublisher or subscriber of the DDS app in a do-main acting like a container for other objects

Advantages

(i) It is highly reliable and offers QoS support(ii) It is highly scalable(iii) It is appropriate for real-time apps(iv) It is fault tolerant(v) It has high level of interoperability(vi) It does not need brokers that may act as

bottlenecks(vii) It offers automatic discovery(viii) It has a user-defined data structure for topic(ix) It supports publisher and subscriber decoupling

Disadvantages

(i) It is originally developed for only stand-aloneLANs

(ii) A lot of memory is needed(iii) It has no open-source libraries to support con-

strained devices(iv) It is hard to design and configure compared to

other protocols [49]

38 RESTful Services REpresentational State Transfer (REST)-is is an architectural style based on a set of principles thatdefine and address networked resources It offers web servicesthat enable transmission from the HTTP device to the devicewithin the IoT architecture and enables clients to reach theserver [16]-is protocol utilizes HTTP commands to backupthe requestresponse messaging model HTTP is a popularand secure protocol in the WWW using TLSSSL [16] ARESTful application has the following characteristics

(i) State and functionality are separated into distrib-uted resources

(ii) All resources are distinctively named by the use ofunique HTTP commands (GET POST PUT orDELETE over the Internet)

22 Journal of Computer Networks and Communications

(iii) It is a clientserver stateless layered protocol thatsupports caching REST uses the same methods asHTTP for all requests or responses with POST andGET for creating and retrieving resources PUT forupdating and changing the state of resources andDELETE for removing resourcesAdvantages of REST over arbitrary web services(SOAP) in the IoT context

(i) It reduces overhead(ii) It reduces parsing complexity(iii) It is stateless(iii) It integrates more tightly with HTTP [58](iv) It is easy to implement and learnis together

with its natural t over HTTP makes it apreferred method for Web 20 applications toshowcase their data and secure M2M sessionsin the IoT

(v) Applications that support RESTful web ser-vices exhibit better performance than the re-source-constrained WSN nodes [59]

(vi) REST is important for the IoT since it supportscommercial M2M cloud platforms

(vii) Its implementation in smart phones and tabletapps is also easy since it needs only an HTTPlibrary that is accessible to all operating systemplatforms [56]

Disadvantages

(i) Although it enjoys wide usage in commercialM2M platforms it is not likely to dominatebecause it is hard to implement

(ii) Its HTTP usage makes it incompatible withconstrained M2M devices leaving its usage toend applications

39 WebSockets It is a web-based protocol operating on asingle TCP channel and providing full-duplex transmissions

It permits secure duplex communication between a clientexecuting unreliable code in a restricted environment and adistant node communicating from that code while using theorigin-based security model employed by web browsers overTCP [60] e protocol is not a requestresponse model andis not a publishsubscribe model So unlike other protocolsa web-socket session is established by the client initiating ahandshake to the server to launch the session is onceestablished a full-duplex client-to-server connection beginsasynchronously and runs until terminated by either party[56] It gives a model for browser-based apps utilizing bi-directional communication with no need of many HTTPconnections to servers

WebSocket comprises two parts ie (i) the handshakehaving a clientrsquos message and the serverrsquos handshake responseand (ii) the data transfer HTTP headers should be repeated byapplications in all client requests and server responses arisingfrom the endless polling which might increase the commu-nication overhead subject to the application [61] WebSocketcan be used to come up with scalable real-time web appli-cations since it can enable full-duplex transmission via onesocket across the Web Its security can be ensured with thehelp of TLSSSL is protocol is suitable for resource-con-strained devices and IoTapplications will not have a good userexperience with it since it operates a clientserver architectureNevertheless it can outcompete all TCP-based protocolsbecause of the following advantages

(i) It is t for real-time transmission(ii) It has improved security(iii) It minimizes transmission overhead(iv) It can provide ecient messaging systems using

WebSocket ApplicationMessaging Protocol (WAMP)[17]

310 Streaming (Simple) Text-Oriented Messaging Protocol(STOMP) is is a TCP-based protocol designed for textmessage-oriented middleware (MOM) with the use of

Topic

Datareader

Datareader

Datareader

Datawriter

Datawriter

Datawriter

Subscriber PublisherPublisher

Data bus (DDS domain)

Subscriber

Topic Topic

Figure 7 DDS architecture

Journal of Computer Networks and Communications 23

HTTP-like commands for interoperability among platformslanguages and brokers Data communication takes placebetween the client and the broker in line with commands(like CONNECT DISCONNECT ACK NACK SUB-SCRIBE UNSUBSCRIBE SEND BEGIN COMMIT orABORT) then headers in the form ltkeygt ltvaluegt (one perline) then a blank line and then body content terminatingin a null character Server-client communication takes placeover a MESSAGE RECEIPT or ERROR frame with headerand body content format being analogous MOM productssupporting STOMP include Apache ActiveMQ or FuseMessage Broker HornetQ Open Message Queue(OpenMQ) RabbitMQ and syslog-ng through its STOMPdestination plugin [62] In addition to these other appli-cation layer protocols include simple media control protocol(SMCP) lightweight local automation protocol (LLAP)simple sensor interface (SSI) lightweight M2M (LWM2M)XMPP-IOT and simple object access protocol (SOAP)

Advantages

(i) It is lightweight for constrained networks(ii) It has flexibility to choose quality of services with

the given functionality(iii) It is standardized by OASIS Technical Committee(iv) It is easy and quick to implement

Challenges and opportunities

Here security threats are imminent because of the open-endedness of the application layer [63] -ese could includethe following

(i) -ere is poor or lack of a security design for anapplicationrsquos function Coming up with a gooddesign to address this could be a good ideaopportunity

(ii) Some of the programs might have backdoorsevading otherwise secure controls to allow un-authorized access to network resources We requiremechanisms to patch up such loopholes

(iii) Applications that lack or have weak authenticationmechanisms can be easily targeted by unlicensedusers to abuse the network Such mechanisms needto be strengthened further

(iv) Excessively complicated access control rules maybecause of their misunderstanding or poor writinggive unauthorized access allowing dubious activityrather than protecting the network they purport toprotect -ese should be simple and clear to un-derstand by all users

(v) In booting workstations without disks and managingnetwork devices we usually use the TFTP protocolalthough access does not require authenticating theuser by usernamepassword -is makes an intruderable to gain easy access to configaccess info as longas he can guess the filenames -is needs to be fixed

A few steps can be used to secure the application layerTo protect network data transmitted across the network

applications need to strengthen their encryption and au-thentication mechanisms Apps too need to enable strictcontrols of privileges for data access using good mechanismsto balance between usability and effectiveness [63] Appli-cations handling sensitive data need detailed logging andaudit capability as well as testing and reviewing

Table 3 summarizes the application layer protocols dis-cussed with their weaknesses and differences among them

4 Cross-Layer QoS Strategies

41 Service-Differentiated Real-Time Communication Scheme(SDRCS) -is is an event-driven routing protocol that routesreal-time traffic by featuring a cross-layer packet-forwardingdesign in which it integrates the real-time routing functionalitywith a new priority-based MAC scheme -e protocol ap-proximates distributed packet traversal speed for traffic clas-sification and admission control on the basis of this design Itfurther localizes decision-making by prioritizing packet for-warding to maximize the speed of packets [64]

Advantages

(i) It improves bandwidth utilization bandwidthdegradation due to unschedulable packets can beavoided

(ii) It offers a lightweight packet schedulability estima-tion mechanism Here it uses received signalstrength and admission control and then earlymissed-deadline packet-dropping policies are made

(iii) It delivers high end-to-end throughput via highersource data rates alongside strict end-to-end latencyneeds

(iv) Nodes which meet real-time requirements of lesstraffic load and better channel quality are givenhighest priority when it comes to packet forwarding

-e SDRCS operates via 5 mechanisms namely (1) RSS-(Rich Site SummaryReally Simple Syndication-) basedgrouping (2) admission control (3) prioritized queuing(4) real-time MAC and (5) dynamic forwarding -isprotocol has a queuing policy that employs per-hopdeadline-based queues and the nodes schedule theirpackets using FIFO priority-based queues [1] -e prin-cipal features of this protocol are mentioned as follows

(i) It is event driven and is easily adaptable to anynetwork changes

(ii) It does not need additional hardware for locali-zation or multichannel transmission

(iii) -e cross-layer packet-forwarding strategy can beused when transmitting multimedia traffic

(iv) It is able to efficiently circumvent any voids therein

Early deadline miss (EDM) policy drops all unscheduledpackets -e SDRCS circumvents wrong packet dropsdue to EDM since it adjusts to network changes easily

42 Sensor Fuzzy-Based Image Transport (SUIT) -is pro-tocol regulates congestion using a scheme based on fuzzy

24 Journal of Computer Networks and Communications

logic It transmits packets with reduced quality to allowablelevels if there is congestion enabling it to transmit lower-quality packetsframes without being dropped-e numberof packets delivered per second thus rises and while thisenhances video streaming apps it is more appropriatelymade for those that quickly transmit non-real-time videosuch as in video surveillance SUIT can stream JPEG pic-tures other than the traditional predictive video codingmethods that are not suitable for sensor devices because oftheir constraints of energy memory and computationalspeed Progressive JPEG-PJPEG performs better via packettransmission rate and delay though it degrades the imagequality because of congestion [65] -is protocol employscross-layer methods for interlayer data exchange though itstill employs some layering making it still possible topreserve all advantages of the layered design since it hasapplication and transport layers but has noMAC or routinglayer

For congestion detection and control SUIT bases on thepercentage number of in-bound and out-bound frames perwindow number of contenders and buffer occupancy of thenext-hop node Since buffer management has an effect oneffectiveness and QoS inWMSNs it is useful in the transportlayer so the protocol prioritizes packets for better QoS SUITreorders packets by tracking the sequence numbers of thedelivered packets since they all have a source ID framenumber and sequence number that it uses to detect anymissing packets -e protocol however does not providepacket reliability [65]

Advantages

(i) -e protocol prioritizes packets for better QoS(ii) It can transmit lower-quality packetsframes

without being dropped

Disadvantages

(i) It does not provide packet reliability

43 Network Layer QoS Support Enforced by a Cross-LayerController (NLQS) -is scheme permits packet-level servicedifferentiation as a function of throughput end-to-endpacket error rate and delay [66] (see Figure 8) -is en-hances network layer QoS It comprises a cross-layer controlunit (XLCU) for configuring and controlling networkingfunctionalities at physical MAC and network layers [18]-is depends upon the unified logic which affects choices forapplication layer needs and status of functional blocks whichdo implementation of networking functions [66]

Advantages

(i) Cross-layer interactions can be controlled withoutweakening the upgradability modularity andsimplicity of designing the system

44 Cross-Layer Signaling Shortcuts (CLASS) CLASS has ahigh level of effectiveness flexibility and comprehensive-ness According to [67] it has the following features

(i) Direct signaling among remote layers leading tofaster transmission

(ii) Lightweighted internal message format standard-ized protocols are not suitable for use in internalsignaling since they are not usually lightweight forinstance transmitting against faults in the networkCLASS requires only three fields namely destina-tion address event type and event contents

(iii) -e setup of its exterior messages is standardconsidering external signaling ICMP and TCPIPare suitable for general messages and short notifi-cations respectively

(iv) Message control protocol should ensure optimizedand organized exchange of dense simultaneousmessages across layers for high efficiency andavoidance of possible conflicts -e protocol appliesto many cross-layer signaling situations

Table 3 Differences and problems with application layer protocols

Protocol Architecture Transport QoS options Security Weaknesses

CoAP Requestresponse UDP DTLS No in-built security features DTLS does not supportmulticast

MQTT Publishsubscribe TCP TLSSSL Insufficient security at the protocol level sincepayload values are not encrypted

SMQTT Publishsubscribe TCP CPKP-ABE Key revocation and group pubsub for distributedSMQTT are still a challenge

MQTT-SN Publishsubscribe UDP TLSSSL SSLTLS suffers from attacks like BEAST CRIMERC4 and Heartbleed

XMPP RequestresponsePublishsubscribe TCP X TLSSSL Susceptible to spamming attacks and lacks worldwide

supportAMQP Publishsubscribe TCP TLSSSL Inappropriate for real-time applications

DDS Publishsubscribe TCPUDP TLSSSL Memory intensive and no open-source libraries forconstrained devices

REST Requestresponse HTTP X HTTPS Hard to implement Uses HTTP so incompatiblewith constrained apps

WebSockets PublishsubscribeClientserver TCP X TLSSSL Bad user experience for IoT apps since it runs the

clientserver architectureSTOMP Clientserver TCP HTTP -e broker can act as a bottleneck

Journal of Computer Networks and Communications 25

Advantages

(i) e protocol is scalable and has a very lowpropagation delay

(ii) It is very ecient and exible since it uses directsignaling between any two layers

(iii) It is very comprehensive and eective

45 Cross-Layer and Multipath-Based Video Transmission(CMVT) CMVT is a hybrid of application and networklayers At the application layer it carries out the encoding ofvideo streams to video data frames (I- P- and B-frames) bymeans of the MPEG-4 encoding format At the networklayer route discovery and data transfer take place Twoalgorithms greedy forwarding and rollback are used todiscover a number of paths from the source to the sink nodevia the route discovery method [20] A specied node icalculates the evaluation of its adjacent node j using thefollowing equation

fij (1 minus α)d2(j D) minus d2min(i)d2max(i) minus d

2min(i)

+ αeinit(j) minus eres(j)

einit(j)

(19)

where fij evaluation value for the node i to j d2(j D) isthe distance from the node j to the destination node D

d2min(i) and d2max(i) are the minimum and maximum dis-

tances of neighbors of the node i to D respectively einit(j) isthe initial energy of the node j eres(j) is the current residualenergy of the node j and α is the energy coecient given asfollows

α emax(i) minus emin(i)

emax(i) (20)

where emax(i) and emin(i) are respectively the maximumand minimum energy left of all node irsquos neighbors

e network layer further transmits video streams withCMVT doing status evaluation choosing an appropriatecommunication path for any packets and the QoS guaranteelevel for the path i is calculated as

fi (1 minus ω)hisum hi

+ ωnisum ni

(21)

where fi is the evaluation value for the path i hi are the hopsfor the path i ni is the summation of packets sent via isum ni isthe summation of packets sent by sources and ω is theenergy consumption factor [18]

Advantages

(i) e CMVT protocol is very superior in mediatransmission particularly large-scale WMSNs

To network peers

Flow requirements

-

Atimes

Admissioncontrol

Capacityassignment

Next-hopselection

Coding rate

Data rate

QoScontracts

Cross-layer controlunit (XLCU)

Localization subsystem

Synchronizationsubsystem

Networkcoordinationunit (NCU)

UWB modulator

Channel coder

QoS scheduler

Georouting

Source coder

Rate control

Figure 8 Cross-layer controller architecture

26 Journal of Computer Networks and Communications

46 Cross-Layer Cooperative MAC (CoopMAC)CoopMAC [68] is based on the IEEE 80211 DCF approach-ere is a reduction in interference of neighboring cellsleading to uniform coverage in densely deployed networks[68] A connection is established by two-way authenticationbetween the sender and the receiver When a Clear-to-Send(CTS) packet and short interframe space (SIFS) are obtainedby the sender packets are transmitted right to the receiverwith no valuable cooperation If available the two nodes findout whether they can exchange data -is is done by the useof a helper identification signal to establish how feasiblecooperative communication is It is established in thepresence of a signal and if it is not available direct com-munication is activated For distributed systems helper-initiated cooperation is preferable because of the presence ofthe RTSCTS packets [69]

Advantages

(i) CoopMAC has an advantage over others due to thespatial diversity amongst the three nodes [70]

(ii) It has the potential to achieve significantthroughput and delay with no complexity in thedesign of the system

47 Cooperative MAC Protocol for Multihop Networks (M-CMAC) Just as CoopMAC M-CMAC is made in such away that high data rate stations help low data rate stations toforward traffic for broadcasting [71] -ere are also helperschosen in such a way that 2 fast-hop transmissions replace aslow hop transmission -e helper via whom there existsleast delay from the sender to the receiver is considered bestbecause of theminimum two-hop transmission rate and so isthe chosen neighbor to the duo Here the working as-sumption is that all nodes have their position coordinatesknown and thus the Euclidean distances between all pairscan be calculated and converted to the data rate for that linkAll nodes have cooperative tables (CT) of potential helperswith destination and helper MAC address Euclidean dis-tance and total distance through the helper Any sender withdata to transmit checks out for the existence of any helper inthe CT for the destination and if available forwards RTS forchannel reservation for single-hop duration M-CMAC hasgot an RTS format having a shape with five fields similar tothe one given as follows

Framecontrol Duration Source

addressDestination

addressHelperaddress

-e destination nodersquos address is kept by the source in thepart for the helper address that is in turn saved as the desti-nation address Nodes inspect helper and destination addressfields on receipt of RTS whereby the node acts as the othernodersquos helper in case the helper address field is not the sameand the helper transmits CTS back to the source if at all it wantsto forward data On delivery of the CTS packet the sourcetransmits to the helper which in turn sends the packets to thedestination that also sends ACK to the helper on receipt of thepacket In M-CMAC there exists a higher level of channelreuse (parallel transmissions) due to the increased number of

nodes therein that subsequently raises availability of helpers fordata forwarding -is results in increased throughput incomparison with CoopMAC and IEEE 80211 DCF [71]

Advantages

(i) It promises a higher throughput compared toCoopMAC and IEEE 80211 DCF

48 Cluster-Based Cooperative Routing (CBCR) Protocol-e CBCR protocol has a multihop data-forwarding func-tion realized at the link layer with cooperative links that useM-CMAC -is protocol encompasses two stages namelythe routing relay selection phase and data forwarding phase

(i) Routing relay selection phase all nodes announcetheir presence to their neighbors by broadcastingperiodical beacon messages that carry the MAC ad-dress of the node Each of these constructs a relaytable that contains all neighboring nodes with whichit is able to communicate -e node further broad-casts its neighbor list in case of any change to itsentries since the previous broadcast -e MAC ad-dresses of nodes next to the node X are contained inits relay tablersquos column one and the neighbor nodersquosrow has MAC addresses for neighbors of the adjacentnode Based on its relay table each node in-dependently chooses routing relays Choosing a nodeas a relay node depends on the number of nodes itconnectsmdashit should be the highest number [71]

(ii) Data forwarding phase a node with packets totransmit needs to first verify if the receiver is in thesame cluster and if that receiver has a helper to whichthe packet is thence forwarded but in whose absencethe packet is delivered right to the receiver In casethe intended receiver is in a different cluster therelay table will be checked to see whether that re-ceiver is reachable via other routing relays If it isreachable packets will be sent to the routing relaydirectly or via the helper if at all the relay has one Incase of destination unreachable via relays packetsare multicast to all relays by the node [71]Advantages

(i) It enables multicasting

49 MAC-PHY Cross-Layer Protocol A cooperative cross-layer standard for cooperation at the physical layer in next-generation WMSNs is developed in [70] It provides acomplete MAC layer algorithm that gives a supportive shellfor the PHY-MAC layer Like CoopMAC its scheme de-pends on the middle node for communication betweennodes-eMAC layer protocol has been changed to regulateinformation exchanges at the physical layer -e receivernode receives duplicate packets from source and helpernodes to decode the data [70]

410 MAC-Centric Approach -is cross-layer protocoltargets multimedia applications by using the MPEG-4

Journal of Computer Networks and Communications 27

scheme [21] It is characterized by 4 access categories(AC3ndashAC0) according to their priority in transmission It ismeant for supporting different QoS needs in upcomingvideo applications and enabling MAC layer differentiationfor H264 partitioning [18] Because of low bandwidth delayand other QoS challenges that cause inefficiency in trans-mittingmultimedia data inWMSNs a number of algorithmsdepending on IEEE 80211e have been proposed to supporttransmission of quality videos [72] Selecting an AC dependson the measures for QoS such as the loss rate and delay-us AC3 having the highest priority is mapped to theparameter set concept since the stream is sensitive to bit lossespecially in video transmission [21]

411 Adaptive Cross-Layer Forward Error Correction(ACFEC) In the ACFEC [73] model data packets are ex-changed among nodes through the access point (AP) whichoperates the infrastructure mode in which it adds FEC to videodata [72] -ese data are dealt by encapsulation using astreaming server to the receiver through thewireless AP as RTPpackets An adaptive FEC controller senses the category ofpackets out of the RTP header and recovers the header ofpackets from UDP -e encoder generates some error-cor-recting packets whose number is determined by the sourcepacket number of the block Multimedia transmissions aremonitored by the controller using MAC failure data and itscounter is increased by one if there is a failed transmission-econtroller uses the failure counter to change the packet numberproduced after transmitting a block [18] In case there are lostpackets it changes the redundancy rates and produces extrapackets to replace the lost ones and satisfy the requirements ofthe receiving node [72] -e FEC packet number is enhancedor reduced to satisfy the receiverrsquos requirements and stoppacket losses-is is done through accurate detection of packetlosses and adjustment of redundancy rates If all video datapackets are well received there will not be generation of FECpackets [73]

Advantages

(i) It promises a good QoS through packet loss re-duction and redundancy rate adjustment

412 Balanced Cross-Layer Fuzzy Logic (BCFL) DesignRouting Algorithm in WSNs A new fuzzy logic-basedrouting algorithm (BCFL) [74] was designed using disper-sion of the cross-layer parameter as the fuzzy logic inferencesystem input Each cross-layer parameter has a dynamicweight depending on the value of dispersion -e designcomes with some innovations as follows (i) for fuzzy logicinference system input the absolute parameter value issubstituted by parameter dispersion thereby significantlyreducing algorithmic complexity (ii) dispersion does notchange with the order of magnitude according to the dis-persion formula and (iii) the weight of the parameter de-pends on the size of its dispersion and the two are inverselyproportional to each other -is enables BCFL to have somenobility unlike other algorithms as per the following dis-tinguishing properties making it advantageous

(i) It has got simple if-then rules which remain con-stant even when the constraints increase

(ii) It is capable of dealing with many constraints withno increment in complexity

(iii) It has capacity to yield a more balanced solutionthan other algorithms

(iv) It is easily adaptable to changes in network con-ditions and topology even when the changes arefrequent like in underwater WSNs -e algorithm isuseable in choosing the CH in cluster-based routingprotocols [74]

413 Minimum Hop Disjoint Multipath Routing Algorithmwith Time Slice Load-Balancing Congestion Control SchemeMHDMwTS is a two-phased routing protocol comprisingpath build-up and path acknowledgment phases It consists ofmultiple sources each of which has three build-up disjointpaths namely primary alternate and back-up paths Onactivation the source node requests to build up a route to thenearest hop neighbor which is the path build-up phase [19] Inthis phase step one has source activation in which therequested node for path building adds its number and time-stamp and sends to the least hop-count neighboring node-isgoes on until the least time latency sink having the informationneeded to construct the primary route is reached [75] In steptwo there is path extraction when the new package fromanother path arrives whereby the extracted path is comparedwith the primary path If the node is shared the package will berejected or an alternative path will be searched for to get abackup through comparison of the preceding two paths Phasetwo is path acknowledgment where we have the third step inwhich the sink returns the ACK packet to the sender with pathinformation having nodes and their time information after ithas computed it using the timestamp [1819]

Advantages

(i) -e protocol minimizes end-to-end latency(ii) It enhances congestion control

414 Cross-Layer Optimal Design (CLOD) Authors in [76]come up with a CLOD for scheduling at the data link layerrouting at the network layer and controlling congestion atthe transport layer with an assumption of fixed link capacity-rough congestion control energy efficiency is improvedTransport layer congestion at the nodes is minimized bycompressed sensing (CS) in which transmitted bits are re-duced whereas optimally allocating resources decreasescongestion on the links at the data link layer CLODpromises minimized computational complexity and betterperformance in light traffic scenarios [77] Generating andstoring CLOD is more efficient compared to Gaussianrandom matrices It prolongs network lifetime and savesenergy

Advantages

(i) It does congestion control subsequently increasingenergy efficiency

28 Journal of Computer Networks and Communications

(ii) It reduces computational complexity(iii) It improves performance in light traffic scenarios

415 Challenges and Opportunities of the Cross-Layer Design(CLD)

(i) Different layers in a real network assume differentfunctions andor services A layer only communi-cates with its neighboring layers -e layered modeldegrades system performance because of many

features in wireless transmission QoS parameters inthe layers of WMSNs are shared by routing pro-tocols to optimize the performance Neverthelessthere is a need to design cross-layer models forincreased efficiency in routing [1]

(ii) -e physical layer plays a highly vital role in CLDRate adaptation and channel allocation take place atthe physical layer via signal processing to enhanceQoS End-to-end performance in wireless media isaffected by the changes therein given the effect on

Table 4 A summary of some of the reviewed cross-layer models

Protocol Layers Aims Comments QoS parameters

SDRCS MACPHY Routing real-time traffic Transmits multimedia traffic viacross-layer packet forwarding

-roughputlatency

SUIT Applicationtransport Transmission of non-real-time video Uses fuzzy logic to regulate

congestionTransmission rate

delay

CLASS Any twoDesign serves as a framework fordifferent implementations ofdifferent application scenarios

Scalable very efficient flexible andhas very low propagation delay

Propagation delayjitter

NLQS PHYMACnetwork

Permits packet-level servicedifferentiation as a function of

throughput packet error rate anddelay

Network layer QoS is enhanced -roughput delaypacket error rate

CMVT Applicationnetwork

Encodes video streams to video dataframes via MPEG-4 encoding anddoes route discovery and data

transfer

Greedy forwarding and rollback areused to find source-to-sink paths Energy

CoopMAC PHYMAC Offers spatial diversity among thethree nodes

Helper ID signal finds out howfeasible cooperative communication

is

-roughput ratedelay

M-CMAC PHYMAC Increases end-to-end throughputand packet delivery ratio

Euclidean distances between nodesare calculated and converted to the

data rate for a link

-roughputpacket delivery

ratio

CBCR PHYMACMinimizes control overhead andtime consumed in establishing thecooperative paths than M-CMAC

Energy consumption is moreuniformly distributed in a network

enhancing network lifetime

-roughputpacket deliveryratio energy

MAC-PHY MACPHYMeant to enable PHY layer

cooperation and maximize gains ofcooperation at the MAC layer

Leverages both spatial diversity andcoding gain -roughput delay

MAC-centric MACAPPL

Meant to support QoS needs in newvideo apps and enable MAC layer

differentiation for H264partitioning

Targets multimedia applications byusing the MPEG-4 scheme

Delay packet lossrate

ACFEC MACnetwork(UDP)

Meant to enhance the quality ofvideo streaming over 80211WLANs

and overcome packet losses

Adjusts redundancy rates toovercome channel fluctuations and

detect and reduce packet lossPacket loss rate

MHDMwTS mdash

Meant to provide reliable datatransfer with the multipath routingand load-balancing congestioncontrol method in WMSNs

More reliable than basic routingschemes for transport multimedia

data

Latency packagetransmit rate

BCFL mdash

Introduces dispersion into fuzzylogic-based routing and sets every

cross-layer parameter with adynamic weight

Can be used to select a CH in cluster-based routing protocols andproposes a dispersion formula

Node utilitydispersion

CLODDatalinknetworktransport

Prolongs network lifetime andachieves congestion control anddesigned for lightly loaded WSNs

Assumes fixed link capacity andintegrates compressed sensing

technology

-roughputaverage energy CS

error ratio

Journal of Computer Networks and Communications 29

the purposefulness of the protocols at the networklayer CLD offers solutions to conserving powerminimizing energy and controlling both the flowand congestion in the network as well as fault tol-erance so it is an opportunity for designers con-sidering other layers We further desire to developCLDs that shall make an impact on network oper-ation and get close attention [18]

(iii) CLD interface standardization the architecturemust deliver module functionality although thereexist some queries amongst modules about possibleinterfaces predicted by the necessity to share in-formation among remote protocol layers Technicalchallenges include developing designing andstandardizing of cross-layer interfaces and algo-rithms that satisfy the cross-layer optimization re-quirements among protocol layers

Table 4 summarizes these cross-layer models with a fewcomments

5 Future Research Direction

51 At theMACLayer In WMSNs achieving duty cycling istricky since video traffic is volatile Listening to adjacentnodes wastes energy We need to switch states from awake tosleep -is requires further research

Self-organizing networks and green communicationalthough most CRWSN energy-saving algorithms aredesigned to reduce node transmission power energyconsumption is inevitable during the operational mode ofBSs because of internal processing Scholars are trying withpolicy-makers to encourage a shift to greenWSNs to reduceoperational costs and carbon footprints With self-orga-nizing CRWSNs for automatic switch-off when idle energyis saved which calls for cognitive switching algorithms Sointegrating cognitive switching with radio resource allo-cation in CRWSNs is a promising research area [78]

52 At the Application Layer Considering a WMSN ar-chitecture end users access the network via the BS to andfrom which they send their sensed data We desire to getguaranteed QoS and energy efficiency by designing a newarchitecture in the form of hardware and software whichneeds to be investigated further for a fix of the same [1]

QoS and energy efficiency are very crucial in WMSNsand mostly real-time apps that require guaranteed band-width and throughput in their network lifetime Mostprotocols ignore base station (BS) mobility and WMSNnodes Traffic management telemedicine and battlefieldsurveillance apps require mobile nodes or BSs so designingdynamic routing protocols to be adaptable in these cir-cumstances is necessary

53 Cross-Layer Security of data transmitted on WMSNseg in military surveillance and e-commerce is key withQoS and energy efficiency WMSNs are vulnerable to attacks

like worms and sinkhole Moreover their computationpower is not that high complicating implementation ofstrong secure protocols -is needs further attention

For fairness and priority issues in CRWSNs nodes areassigned diverse priorities depending on importance andurgency of their data For better network performance thereis a need to develop fair resource allocation mechanismswith priority-based fairness in sensors-ere is still potentialin this research area [78]

We must develop multiobjective adaptive protocols tooptimize QoS metrics involved in routing for best trans-mission results to get a trade-off between different opti-mization metrics We need to design standardized cross-layer algorithms which are able to meet cross-layer opti-mization standards

6 Conclusion

We have reviewed the different QoS strategies for WSNs inthe context of IoT from the MAC layer and application layeras well as the cross-layer paradigm For the MAC layer wehave reviewed protocols for WSNs such as the hybrid of thecontention-free and contention-based MAC protocols andthose for WMSNs We further reviewed a number of ap-plication layer protocols including many machine-to-ma-chine requestresponse and publishsubscribe protocols Forsystem optimization cross-layer QoS strategies are veryimportant in wireless communication We have reviewed anumber of cross-layer strategies For all categories reviewedchallenges and opportunities are discussed Finally somepossible future directions are discussed for research andapplication showing a promising potential for future re-search in this relatively new area especially as more WMSNapplications emerge recently

Conflicts of Interest

-e authors hereby declare no conflicts of interest regardingthe publication of this work

Acknowledgments

-is research was funded by the National Natural ScienceFoundation of China (Nos 61602398 61672447 and61711540306)

References

[1] V Bhandary A Malik and S Kumar ldquoRouting in wirelessmultimedia sensor networks a survey of existing protocolsand open research issuesrdquo Journal of Engineering vol 2016Article ID 9608757 27 pages 2016

[2] I F Akyildiz T Melodia and K R Chowdhury ldquoWirelessmultimedia sensor networks applications and testbedsrdquoProceedings of the IEEE vol 96 no 10 pp 1588ndash1605 2008

[3] M Iqbal M Naeem A Ahmed M Awais A Anpalagan andA Ahmad ldquoSwarm intelligence based resource managementfor cooperative cognitive radio network in smart hospitalsrdquoWireless Personal Communications vol 98 no 1 pp 571ndash5922018

30 Journal of Computer Networks and Communications

[4] S A Alvi B Afzal G A Shah L Atzori and W MahmoodldquoInternet of multimedia things vision and challengesrdquo AdHoc Networks vol 33 pp 87ndash111 2015

[5] Q F Hassan A U R Khan and S A Madani Internet ofings Challenges Advances and Applications CRC PressTaylor amp Francis Group A Chapman amp Hall Group BocaRaton FL USA 2018

[6] Q Wang Y Zhao W Wang et al ldquoMultimedia IoT systemsand applicationsrdquo in Proceedings of the 2017 Global Internet ofings Summit (GIoTS) vol 2 Geneva Switzerland June 2017

[7] A Aslam and E Curry ldquoTowards a generalized approach fordeep neural network based event processing for the internet ofmultimedia thingsrdquo IEEE Access vol 6 pp 25573ndash25587 2018

[8] M Jridi T Chapel V Dorez L Bougeant and A Le BotlanldquoSoC-based edge computing gateway in the context of theinternet of multimedia things experimental platformrdquoJournal of Low Power Electronics and Applications vol 8no 1 p 1 2018

[9] H Noura A Chehab L Sleem andM N Rapha ldquoOne roundcipher algorithm for multimedia IoT devicesrdquo MultimediaTools and Applications vol 77 no 14 pp 18383ndash18413 2018

[10] T Balan D Robu and F Sandu ldquoMultihoming for mobileinternet of multimedia thingsrdquo Mobile Information Systemsvol 2017 Article ID 6965028 16 pages 2017

[11] H Shen and G Bai ldquoRouting in wireless multimedia sensornetworks a survey and challenges aheadrdquo Journal of Networkand Computer Applications vol 71 pp 30ndash49 2016

[12] A Ahmad M H Rehmani H Tembine O A Mohammedand A Jamalipour ldquoIEEE access special section Editorial optimization for emerging wireless Networks IoT 5G andsmart grid communication networksrdquo IEEE Access vol 5pp 2096ndash2100 2017

[13] B A Muzakkari M A Mohamed M F A Kadir andZ Mohamad ldquoRecent advances in energy efficient-QoS awareMAC protocols for wireless sensor networksrdquo InternationalJournal of Advanced Computer Research vol 8 no 38pp 212ndash228 2018

[14] M A Yigitel O D Incel and C Ersoy ldquoQoS-aware MACprotocols for wireless sensor networks a surveyrdquo ComputerNetworks vol 55 no 8 pp 1982ndash2004 2011

[15] A M Abbas and O Kure ldquoQuality of Service in mobile ad hocnetworks a surveyrdquo International Journal of Ad Hoc andUbiquitous Computing vol 6 no 2 p 75 2010

[16] M Q Shatnawi ldquoApplication layer protocols for the internetof things a surveyrdquo in Proceedings of the InternationalConference on Engineering amp MIS (ICEMIS) Agadir Mo-rocco September 2016

[17] V Karagiannis P Chatzimisios F Vazquez-gallego andJ Alonso-zarate ldquoA survey on application layer protocols forthe internet of things research motivationrdquo InternationalJournal of Advanced Research in Computer Science vol 8no 3 pp 1ndash10 2015

[18] M S Bernard T Pei Z Li and K Li ldquoQoS strategies forwireless multimedia sensor networks in the context of IoTrdquoLecture Notes of the Institute for Computer Sciences SocialInformatics and Telecommunications Engineering vol 4pp 228ndash253 2019

[19] A AlAmri and M Abdullah ldquoCross-layer quality of serviceprotocols for wireless multimedia sensor networksrdquo in Pro-ceedings of the International Conference on CommunicationManagement and Information Technology pp 649ndash658Cosenza Italy April 2017

[20] J Guo L Sun and R Wang ldquoA cross-layer and multipathbased video transmission scheme for wireless multimedia

sensor networksrdquo Journal of Networks vol 7 no 9pp 1334ndash1340 2012

[21] A Ksentini A Naimi and M Gueroui ldquoToward an im-provement of H264 video transmission over IEEE 80211ethrough a cross-layer architecturerdquo IEEE CommunicationsMagazine vol 44 no 1 pp 107ndash114 2006

[22] W Ye J Heidemann and D Estrin ldquoAn energy-efficientMAC protocol for wireless sensor networksrdquo in Proceedings ofthe 21st International Annual Joint Conference of the IEEEComputer and Communications Societies (INFOCOM 2002)New York NY USA June 2002

[23] J Polastre J Hill and D Culler ldquoVersatile low power mediaaccess for wireless sensor networks categories and subjectdescriptorsrdquo in Proceedings of the Second InternationalConference on Embedded Networked Sensor Systems Sen-Sysrsquo04 pp 95ndash107 Baltimore MA USA November 2004

[24] T Danmanee K N Nakorn and K Rojviboonchai ldquoCU-MAC a duty-cycle MAC protocol for internet of things inwireless sensor networksrdquo ECTI Transactions on ElectricalEngineering Electronics and Communications vol 16 no 2pp 30ndash43 2018

[25] T Van Dam and K Langendoen ldquoAn adaptive energy-efficientMAC protocol for wireless sensor networksrdquo in Proceedings ofthe 1st International Conference on Embedded NetworkedSensor Systems Los Angeles CA USA November 2003

[26] P Xie and J Cui ldquoR-MAC an energy-efficient MAC protocolfor underwater sensor networksrdquo in International Conferenceon Wireless Algorithms Systems and Applications (WASA2007) Chicago IL USA August 2007

[27] A Rahim N Javaid M Aslam U Qasim and Z A KhanldquoAdaptive-reliable medium access control protocol forwireless body area networksrdquo in Proceedings of the AnnualIEEE Communications Society Conference on Sensor Meshand Ad Hoc Communications and Networks (SECON) SeoulKorea June 2012

[28] H S Kwak S Ullah D H Kwak and C H Lee ldquoA powerefficient MAC protocol for wireless body area networksrdquoJournal of the Korean Institute of Intelligent Systems vol 8no 6 pp 131ndash140 2009

[29] C Omeni A J Burdett and M Park ldquoEnergy efficientmedium access protocol for wireless medical body area sensornetworksrdquo in Proceedings of the 4th IEEEEMBS InternationalSummer School and Symposium on Medical Devices andBiosensors pp 29ndash32 Cambridge UK August 2007

[30] L Sitanayah C J Sreenan and K N Brown ldquoER-MAC ahybrid MAC protocol for emergency response wireless sensornetworksrdquo in Proceedings of the Fourth International Con-ference on Sensor Technologies and Applications pp 244ndash249Mestre Italy July 2010

[31] Y Sun and D B Johnson ldquoRI-MAC a receiver-initiatedasynchronous duty cycle MAC protocol for dynamic trafficloads in wireless sensor networksrdquo in Proceedings of the 6thACM conference on Embedded network sensorsystemsmdashSenSysrsquo08 pp 1ndash14 Raleigh NC USA November2008

[32] M Buettner G V Yee E Anderson and R Han ldquoX-MAC ashort preamble MAC protocol for duty-cycled wireless sensornetworksrdquo in Proceedings of the 4th International Conferenceon Embedded Networked Sensor Systems SenSys 2006pp 307ndash320 Boulder Colorado USA October 2006

[33] S Boulfekhar and M Benmohammed ldquoSynchronous receiverinitiated MAC protocol for long-lived sensor networksrdquoComputers amp Electrical Engineering vol 40 no 2 pp 504ndash516 2014

Journal of Computer Networks and Communications 31

[34] S Mehta and K S Kwak ldquoH-MAC a hybrid Mac protocol forwireless sensor networksrdquo International Journal of ComputerNetworks amp Communications vol 2 no 2 pp 108ndash117 2010

[35] I Rhee A Warrier M Aia J Min and M L Sichitiu ldquoZ-MAC a hybrid MAC for wireless sensor networksrdquo IEEEACM Transactions on Networking vol 16 no 3 pp 511ndash5242008

[36] S Arshad A Al-sadi and A Barnawi ldquoZ-MAC performanceevaluation and enhancementsrdquo Procedia Computer Sciencevol 21 pp 485ndash490 2013

[37] Y Liu I Elhanany and H Qi ldquoAn energy-efficient QoS-aware media access control protocol for wireless sensornetworksrdquo in Proceedings of the IEEE International Confer-ence on Mobile ad hoc and Sensor Systems pp 9ndash11 Wash-ington DC USA December 2005

[38] J Decotignie ldquoWiseMAC an ultra low power MAC protocolfor the downlink of infrastructure wireless sensor networksrdquoin Proceedings of the Ninth International Symposium onComputers And Communications pp 244ndash251 AlexandriaEgypt June 2004

[39] R A Rashid ldquoDevelopment of energy aware TDMA-basedMAC protocol for wireless sensor network systemrdquo EuropeanJournal of Scientific Research vol 30 no 4 pp 571ndash578 2009

[40] H P Van Hoesel ldquoA lightweight medium access protocol(LMAC) for wireless sensor networksrdquo in Proceedings of theInternational Workshop on Networked Sensing Systems Bal-timore MD USA November 2004

[41] C Li P Wang H Chen andM Guizani ldquoA cluster based on-demand multi-channel MAC protocol for wireless multi-media sensor networksrdquo in Proceedings of the IEEE In-ternational Conference on Communications pp 2371ndash2376Beijing China May 2008

[42] J L Sobrinho and A S Krishnakumar ldquoQuality-of-service inad hoc carrier sense multiple access wireless networksrdquo IEEEJournal on Selected Areas in Communications vol 17 no 8pp 1353ndash1368 1999

[43] J P John ldquoProviding MAC QoS for multimedia traffic in80211e based multi-hop ad hoc wireless networksrdquo ComputerNetworks vol 51 no 1 pp 153ndash176 2007

[44] S Lofty and N Padmavati ldquoA survey on mobility basedprotocols in WSNsrdquo in Proceedings of the InternationalConference Communication and Manufacturing pp 12ndash15Hangzhou China July 2014

[45] S Deng L Shen and J Li ldquoMobility-based clustering pro-tocol for wireless sensor networks with mobile nodesrdquo IETWireless Sensor Systems vol 1 no 1 pp 39ndash47 2011

[46] M Khalil A Khalid F U Khan and A Shabbir ldquoA review ofrouting protocol selection for wireless sensor networks insmart citiesrdquo in Proceedings of the 24th Asia-Pacific Confer-ence on Communications (APCC) Ningbo China November2014

[47] O Message Q Telemetry and T Mqtt Committee Specifi-cation Draft 02 April 2014

[48] M Collina M Bartolucci A Vanelli-coralli andG E Corazza ldquoInternet of things application layer protocolanalysis over error and delay prone linksrdquo in Proceedings ofthe 7th Advanced Satellite Multimedia Systems Conference andthe 13th Signal Processing for Space CommunicationsWorkshop Livorno Italy September 2014

[49] A Talaminos-barroso M A Estudillo-valderrama L M RoaJ Reina-tosina and F Ortega-ruiz ldquoA machine-to-machineprotocol benchmark for e health applicationsmdashuse case re-spiratory rehabilitationrdquo Computer Methods and Programs inBiomedicine vol 129 pp 1ndash11 2016

[50] V Goyal O Pandey A Sahai and B Waters ldquoAttribute-based encryption for fine-grained access control of encrypteddatardquo in Proceedings of the 13th ACM Conference on Com-puter and Communications Security pp 89ndash98 AlexandriaVA USA October 2006

[51] M Singh M A Rajan V L Shivraj and P BalamuralidharldquoSecure MQTT for internet of things (IoT)rdquo in Proceedings ofthe Fifth International Conference on Communication Systemsand Network Technologies Gwalior India April 2015

[52] A Stanford-Clark and H L Truong MQTT for Sensor Net-works (MQTT-SN) Protocol Specification httpmqttorgnewwp-contentuploads200906MQTT-SN_spec_v12pdf 2013

[53] RFC ldquoExtensible messaging and presence protocol (XMPP)rdquoin Proceedings of the IETF pp 1ndash9 San Diego CA USAAugust 2005

[54] S Vinoski and I Technologies ldquoQueuing protocolrdquo IEEEInternet Computing vol 10 no 6 pp 87ndash89 2006

[55] J L Fernandes I C Lopes J J P C Rodrigues and S UllahldquoPerformance evaluation of restful web services and AMQPprotocolrdquo in Proceedings of the Fifth International Conferenceon Ubiquitous and Future Networks (ICUFN) pp 2ndash7 DaNang Vietnam July 2013

[56] M Asim ldquoA survey on application layer protocols for internetof things (IoT)rdquo International Journal of Advanced Researchin Computer Science vol 8 no 3 pp 996ndash1000 2017

[57] K An A Gokhale D Schmidt S Tambe P Pazandak andG Pardo-castellote ldquoContent-based filtering discovery pro-tocol (CFDP) scalable and efficient OMG DDS discoveryprotocolrdquo in Proceedings of the 8th ACM InternationalConference on Distributed Event-Based Systems New YorkNY USA July 2014

[58] E Web What is a web service pp 52ndash57 2010[59] M Laine RESTful Web Services for the Internet of ings

pp 2ndash4 2012 httpmediaTkk[60] IETF -e WebSocket protocol pp 1ndash71 2011[61] V Pimentel ldquoCommunicating and displaying real-time data

with websocketrdquo IEEE Internet Computing vol 16 no 4pp 45ndash53 2012

[62] Wikipedia 2019 httpsenwikipediaorgwikiStreaming_Text_Oriented_Messaging_Protocol2019

[63] K Raghavendra and S Nireshwalya ldquoApplication layer se-curity issues and its solutionsrdquo IJCSET vol 2 no 6pp 1266ndash1269 2012

[64] Y Xue B Ramamurthy andM C Vuran ldquoSDRCS a service-differentiated real-time communication scheme for eventsensing in wireless sensor networksrdquo Computer Networksvol 55 no 15 pp 3287ndash3302 2011

[65] C Sonmez S Isik M Y Donmez O D Incel and C ErsoyldquoSUIT a cross layer image transport protocol with fuzzy logicbased congestion control for wireless multimedia sensornetworksrdquo in Proceedings of the 5th International Conferenceon New Technologies Mobility and Security (NTMS) pp 1ndash6IEEE Istanbul Turkey May 2012

[66] T Melodia and I Akyildiz ldquoCross-layer QoS-aware com-munication for ultra wide band wireless multimedia sensornetworksrdquo IEEE Journal on Selected Areas in Communica-tions vol 28 no 5 pp 653ndash663 2010

[67] Q Wang and M A Abu-Rgheff ldquoCross-layer signalling fornext-generation wireless systemsrdquo in Proceedings of the 2003IEEE Wireless Communications and Networking 2003WCNC 2003 vol 2 pp 1084ndash1089 IEEE New Orleans LAUSA March 2003

[68] P Liu Z Tao S Narayanan T Korakis and S PanwarldquoCoopMAC a cooperative MAC for wireless LANsrdquo IEEE

32 Journal of Computer Networks and Communications

Journal on Selected Areas in Communications vol 25 no 2pp 340ndash354 2007

[69] H Shan H T Cheng and W Zhuang ldquoCross-layer co-operative MAC protocol in distributed wireless networksrdquoIEEE Transactions onWireless Communications vol 10 no 8pp 2603ndash2615 2011

[70] F Liu T Korakis Z Tao and S Panwar ldquoAMAC-PHY cross-layer protocol for wireless Ad-Hoc networksrdquo in Proceedingsof the 2008 IEEE Wireless Communications and NetworkingConference pp 1792ndash1797 IEEE Las Vegas NV USA March2008

[71] L Jacob and H R Shamna ldquoEfficient cooperative MAC androuting in wireless networksrdquo Transactions on Networks andCommunications vol 3 no 5 2015

[72] S Rao and K Shama ldquoCross layer protocols for multimediatransmission in wireless networksrdquo International Journal ofComputer Science amp Engineering Survey vol 3 no 3pp 15ndash28 2012

[73] L Han S Park S Kang and H Peter ldquoAn adaptive cross-layer FEC mechanism for video transmission over 80211WLANsrdquo in Proceedings of the 2009 International Conferenceon Internet pp 209ndash215 Las Vegas NV USA December2009

[74] N Li J Martınez and V H Dıaz ldquo-e balanced cross-layerdesign routing algorithm in wireless sensor networks usingfuzzy logicrdquo Sensors vol 15 no 8 pp 19541ndash19559 2015

[75] G Sun J Qi Z Zang and Q Xu ldquoA reliable multipathrouting algorithm with related congestion control scheme inwireless multimedia sensor networksrdquo in Proceedings of the2011 3rd International Conference on Computer Research andDevelopment vol 4 pp 229ndash233 IEEE Shanghai ChinaMarch 2011

[76] M Li Y Jing and C Li ldquoA robust and efficient cross-layeroptimal design in wireless sensor networksrdquoWireless PersonalCommunications vol 72 no 4 pp 1889ndash1902 2013

[77] J Yan M Zhou and Z Ding ldquoRecent advances in energy-efficient routing protocols for wireless sensor networks areviewrdquo IEEE Access vol 4 pp 5673ndash5686 2016

[78] A Ahmad S Ahmad M H Rehmani and N U Hassan ldquoAsurvey on radio resource allocation in cognitive radio sensornetworksrdquo IEEE Communications Surveys amp Tutorials vol 17no 2 pp 888ndash917 2015

Journal of Computer Networks and Communications 33

International Journal of

AerospaceEngineeringHindawiwwwhindawicom Volume 2018

RoboticsJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Active and Passive Electronic Components

VLSI Design

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Shock and Vibration

Hindawiwwwhindawicom Volume 2018

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawiwwwhindawicom

Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Control Scienceand Engineering

Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Journal ofEngineeringVolume 2018

SensorsJournal of

Hindawiwwwhindawicom Volume 2018

International Journal of

RotatingMachinery

Hindawiwwwhindawicom Volume 2018

Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Navigation and Observation

International Journal of

Hindawi

wwwhindawicom Volume 2018

Advances in

Multimedia

Submit your manuscripts atwwwhindawicom

Page 10: ReviewArticle - downloads.hindawi.com · ReviewArticle QoSStrategiesforWirelessMultimediaSensorNetworksinthe ContextofIoTattheMACLayer,ApplicationLayer,and Cross-LayerAlgorithms MuwongeSsajjabbiBernard

(ii) It is applicable in rapid response situations inWSNs

(iii) It has a higher ratio of packet delivery and lowdelay under lower energy consumption levels

218 RI-MAC (Receiver-Initiated MAC) -is is an asyn-chronous duty-cycle protocol that tries to reduce timeoccupied by the communicating nodes on the medium toagree on time for data exchange-e sending node waits forthe receiving node to signify the beginning of the com-munication session through a short beacon frame that itsends -e channel is only occupied by the beacon and datatransmission reducing channel occupancy and thus en-abling data exchange for other nodes too -is implies animprovement in channel utilization that subsequently leadsto increment in throughput packet delivery ratio andpower efficiency over a wide-ranging traffic load [31] Beinga receiver-initiated model it ensures substantial reductionin overhearing and lowers the probability of colliding andthe cost of recovery compared to B-MAC [23] or X-MAC[32] RI-MAC is further capable of using a beacon withACK and Ready-to-Receive packets though it has aproblem if a sending node with Ready packets needs to havethe radio on till a ready-to-receive node is awake [13] whichthen sends it a beacon and transmission begins immediatelyand later is acknowledged by another beacon Nodes in-termittently wake up to listen to any possible frames meantfor them depending on their schedules A node announcesits readiness to receive a frame through a broadcast beaconon waking up when it switches on the radio When a dataframe has been received the receiver will spend some moretime known as ldquodwell timerdquo in the active state so as toreceive queued packets which is the time in RI-MAC thatdepends on the number of contending senders If there areseveral contending senders that might not have beenpredicted there is a challenge in trying to reduce the re-ceiverrsquos active time to optimize power efficiency and reducethe cost of collision detection and lost data recovery even inthe case of hidden senders For this protocol this can beaddressed by a beacon frame from the receiving node forcoordinating data frame transmissions of contendingsending nodes-e protocol also beats B-MAC and X-MACin minimizing the cost of collision detection and datarecovery through coordination of the receiversrsquo dataframes -is is because the receiving node is aware of themaximum delay prior to the arrival of the frame-e abilityto broadcast is supported by RI-MAC when it sends data asunicast to each senderrsquos neighboring node or continuouslytransmits the frame in a back-to-back mode for as long asthe sleep interval [31]

Advantages

(i) It improves channel utilization leading to in-crement in throughput packet delivery ratio andpower efficiency over a wide-ranging traffic load

(ii) It ensures substantial reduction in overhearing andlowers the probability of colliding and the cost ofrecovery

219 SRI-MAC (Synchronous Receiver-Initiated MAC)-is is a synchronous duty-cycle protocol that implementsthe method of receiver-initiated data transmission with anaim of enhancing network lifetime through energy wastageavoidance via collision overhearing and idle listening To dothis and also minimize duty cycle it uses a sequence ofadaptive beacons and RTSCTS packets In this protocolsending nodes do not go to sleep till they have got CTSpackets from receiving nodes to begin transmitting exceptfor nonsending nodes that return into the sleep mode rightaway as if the channel was idle [33] Since sending nodeshave separate schedules it reduces the contention proba-bility leading to more energy conservation A time frame isdivided into the information period the allocation periodand the communication period which are discussed asfollows

Information period at frame initiation a beacon is sentby the receiving node declaring its return from sleepand readiness to accept data packets in case the channelis not busy -e beacon is a small packet alerting otherlistening nodes of the transmitterrsquos readiness to get arequest for channel It consists of the receiverrsquos ID andthe duration allocation period (DAP) that relies onnumber of the receiving nodersquos neighbors [13]Allocation period here the receiving node stays awake forsome little time after broadcasting its beacon to confirm ifthere is any node having anRTS packet to transmitWhenthe beacon has been received an RTS packet (with 3fieldsmdashsender ID receiver ID and data size) is sent by thesender prompting the receiver to send a CTS for SYNC asthere is neighbor information exchangeCommunication period SRI-MAC is capable of or-ganizing and scheduling transmissions by means ofsensor node IDs -e receiver assigns order andtransmitting time for every sender such that they switchoff their radios till their time to transmit is dueHowever the receiving node remains in the wake-upmode to get all sending nodersquos data SRI-MAC canenable sensors have lengthy sleeping times duringtransmission such that energy can be saved as long as itis neither sending nor receiving data -e protocolrsquoscommunication period is premised on the TDMAtechnique which conserves energy since the radiorsquosduty cycle is lowered and collision is eliminated [33]

Advantages

(i) Since it employs the TDMA principle it eliminatescollisions while maintaining energy conservationtoo

(ii) Noncommunicating nodes switch off their radioand get into the sleep mode which eliminatesoverhearing

(iii) Since receiving nodes broadcast SYNC signals toall nodes in the form of CTS packets overemittingis avoided

(iv) It decreases duty cycling as sleep time increasesand lowers idle listening

10 Journal of Computer Networks and Communications

2110 H-MAC -is protocol is built on the power savingmechanism (PSM) of IEEE 80211 and slotted Aloha pro-tocols to enhance performance using multiple slots Timehere comprises large frames each having an active and asleeping part [34] If any node has got data to send it will askfor slots from the receiver node during active time and latertransmit the packets in the prenegotiated slots Nodes gointo the sleep mode during sleep-time slots if they do nothave data to transmit

Latency in IEEE 80211 if N is the number of hops nthe value for the current hop tcsn the back-off time andttx the transmission delay the total latency is calculatedusing the following equation

D(N) 1113944N

n1tcsn + ttx1113872 1113873 (8)

and the average latency is given by the followingequation

E[D(N)] N tcsn + ttx1113872 1113873 (9)

-roughput in H-MAC is given by the followingequation

Th npnm

tact + Cts (10)

where np is the number of transmitted packets nm is themaximum number of nodes connected in a time frameC is the number of slots with the same length tact is thelistening time and ts is the sleep delayAdvantages

(i) It enhances performance using multiple slots

2111 Z-MAC (ZebraMAC) -is rides on the strengths andoffsets weaknesses of TDMA and CSMA It yields very goodchannel usage and low delay under low contention and lowerscollision among two-hop neighbors at a reduced cost Itsperformance is strong via error synchronization failure inassigning slots and channel conditions that change with timeIn the set-up phase the protocol operates through certainsteps including neighbor discovery slot assignment localframe exchange and global time synchronization whichunless the topology of the network changes significantly donot run again When the data are being transmitted there isimprovement in network throughput and energy efficiencywhich covers up for the initial upfront operational costs [35]

Disadvantages

(i) Initial slot assignment is quite a cumbersome taskhere since it is done at the beginning yet this maynot be at the same time in reality It further as-sumes static links until setup happens again whichtoo is impractical

(ii) It also suffers from the hidden-terminal problemsince slot owners have a small contention windowand priority for slots allocated to them due to theECN messages

(iii) Synchronizing time also poses a challenge in thisprotocol due to a large clock drift between nodes

(iv) Source and destination nodes too are not wellcoordinated in Z-MAC [36]

2112 Q-MAC (QoS-Aware Media Access Control) -isprotocol offers high QoS with improved energy efficiencyand accesses the wireless channel using MACAW as theunderlying protocol [37] Traffic from different nodes isprioritized for QoS satisfaction depending on how criticalthe data are It mainly has two scheduling algorithms inWSNs (i) intranode and (ii) internode scheduling -eintranode scheduling scheme adopts a first-in first-out-(FIFO-) based queuing algorithm based on application andMAC layer abstraction for data packet classification In-ternode scheduling reduces idle listening and collision on thechannel for improved energy consumption

Advantages

(i) It gives a high QoS with improved energy efficiency

2113 WiseMAC (Wireless Sensor MAC) WiseMAC isbased on the Aloha protocol and employs preamble sam-pling in achieving low-power transmissions in infrastructuresensor networks by occasionally sampling the channel tocheck for activity [3238] -e protocol employs the samemethod as B-MAC except that the sender shortens theextended preamblersquos length by scheduling its transmissionsaccordingly after learning the receiverrsquos awake schedules-is is done by the receiver node putting its subsequentawake time on the ACK packet such that when the senderneeds to send to it again it starts the preamble shortly beforeit wakes up so as to avoid energy wastage in transmitting thepreamble [32] -e data frame is repeated instead of theextended preamble for less traffic transmissions that have alengthier preamble than the data frame in WiseMAC Afterprocessing the frame a node that is not the target recipientwill go back into the sleep mode while the recipient staysawake till the session ends thereby sending an ACK packetFor an idle channel the power consumption is rather lowwith this method Nevertheless there is a problem of longwake-up preamble limiting throughput and causing anenormous power consumption overhead in reception [38]All nodes overhearing a transmission bear the overheadtogether with the recipient node -e protocol can addressissues concerning low-power transmission but has nomechanism of nodes adapting to varying traffic patterns[32] To compensate for drift between clock at the accesspoint and on the sensor node there is a need to compute thewake-up preamble duration -e drift varies directly withtime since previous resynchronization and is given by thefollowing equation

TP min 4θl TW( 1113857 (11)

where θ is the time-base quartz frequency tolerance and l isthe interval between two transmissions Increase in trafficvolume naturally reduces overhearing

Journal of Computer Networks and Communications 11

Radiomodel for protocols with lower power transitiondelays between transceiver states and their powerconsumption are modeled Some of the states includeDOZE where the transceiver neither transmits norreceives but is prepared to power on to the receivetransmit state very fast RXwhere the transceiver listensto the channel receives data or demodulates data fromthe noisyidle channel and TX which is the trans-mission state by the transceiver Let TS be the set-uptime needed to turn the transceiver from the DOZE tothe RX or TX state TT the turn-around time needed toswitch between RX and TX states and B the trans-ceiverrsquos bit rate PZ PR and PT are the values of powerconsumed in the DOZE RX and TX statesrespectively

1113954PR PR minus PZ

1113954PT PT minus PZ(12)

where 1113954PR and 1113954PT are the changes in power as a result ofbeing in RX and TX states respectivelyTraffic model let N be the number of sensor nodesconnected to an access point (AP) and λ be the global rateat which the downlink Poisson traffic arrives at the APfrom the fixed network Traffic toward each sensor node isthe same and is given by λN Average packet interarrivaltime Lwith which data packets arrive at a node is given byL Nλ If TC and TD are the control packets and datapackets respectively assuming a small traffic volume theglobal interarrival 1λ far exceeds the summation ofdurations for the data packet and control packet and theturn-around time Mathematically it is given as

1λ≫TD + TC + TT (13)

-e average power consumed by WiseMAC andtransmission delay incurred are given by equations (14)and (15) [38]

PW PZ +1113954PR TS +(1B)( 1113857

TW+

1113954PR X + TD + TT( 1113857

L

+ 1113954PR(N minus 1)Y

L

(14)

where

X 2θL 1 minus eminus TD4θL( )1113874 1113875

Y T2D + 12TDθL

2TW1 minus e

minus TW4θL( )1113874 1113875

DW TD +TW

21 minus e

minus TW4θL( )1113874 1113875

+ 2θL 2 minus eminus TD4θL( ) minus e

minus TW4θL( )1113874 1113875

(15)

where PW and DW are the power consumed byWiseMAC and the transmission delay for WiseMACrespectivelyAdvantages

(i) It achieves low-power transmissions in in-frastructure sensor networks through preamblesampling while checking for activity

(ii) It reduces overhearing due to increased trafficvolume

Disadvantages

(i) It has a problem of long wake-up preamble limitingthroughput and causing enormous power con-sumption overhead

2114 X-MAC It is a simple minimal transmission powerMAC protocol that uses a short preamble with decouplingfor the sending and receiving node sleep plans inWSNs-isleads to a significant reduction in energy usage at sendingand receiving nodes and decrease in per-hop latency andgives flexibility in adapting to bursty and periodic sensordata sources [32]

Protocol design X-MACrsquos design is aimed at increasingenergy efficiency reducing data latency and increasingthroughput -is enables application in different kindsof packetized and bit stream digitized radios as well assimplified low-overhead distributed application -edesign is as such in order to address the low-powerlistening problems including overhearing excessivepreamble and incompatibility with packetizing radiosas in [32]Asynchronous duty cycling a node with information totransmit will send the packets after sending a preamblewhereas the remaining nodes will preserve their sleepschedules that are not synchronized -e receiver willsample the link on waking up and stays awake if it detectsa preamble and if it is the target it will return to sleep if itis not the target after getting the whole preambleEmbedding the target ID in the preamble to avoidoverhearing the lengthy preamble is divided into manyshortened packets embedded with the receiver nodersquosID so as to eliminate overhearing One lengthy pre-amble is constituted by the shortened preamble packetstream Any node on receiving the short preamble afterwaking up will examine the target ID on the packet Itgoes back into the sleepmode right away resuming dutycycling if it is not the targeted recipient or stay awakefor successive packets if it is the one -is approachchecks energy wastage due to network density It canalso be used on several types of radiosReducing excessive preamble using strobing it ispossible to enable low-power transmission and alsosave energy by reducing time for preamble trans-mission through the use of enhanced preambles inaddition to preamble sampling X-MAC does not sendan endless preamble packet stream but puts slight

12 Journal of Computer Networks and Communications

pauses in between packets to allow the sending node topause and listen to the channel and allow the receivingnode to resend an early ACK packet during that pause-is makes the sending node begin sending datapackets and the receiving node cut excessive preamblesand thus decrease per-hop latency and energy wastageon unnecessary waiting and transmissionPacketizing radios low-power listening is limited incapability for supporting packetizing radios butX-MACrsquos short strobed preambles can support all typesof digital radiosAdaptation to traffic load while a number of WSNapplications are reliable in traffic production they stillmust adapt to changing traffic Nodes are going to haveseparate sleep schedules depending on traffic loadConsidering systems that have time-fluctuating trafficall predetermined static schedules are suboptimalOptimality Buettner et al [32] model the expectationof energy as follows

Es (preamble energy + energy perACK listen)

lowast(expected preamble minus listen iterations)

+(energy to send parket)

Es PTxSp + PRxSal1113872 1113873 Rl + Rs( 1113857

Rl minus Sp+ SdPTx

(16)

where Es expected energy to send a packet PTx andPRx are the transmission and receiver power re-spectively Sp Sal and Sd are the duration of thesenderrsquos preamble ACK listen and data transmissionperiods respectively and Rl and Rs are the receiverrsquoslisten and sleep periods respectively Expected energyto receive a packet is given as

Er (listen cycle energy + sleep cycle energy)

lowast(expected iterations for preamble arrival)

+(energy to sendACK) +(energy to receive packet)

Er PsRs + PRxRl( 1113857

1 minus 1 minus Pd(t)( 1113857 Rl + Rs( 1113857+ RaPTx + RdPRx

(17)

And the one-hop expected latency is given as

Lat (preamble duration + ACK listen)

lowast(expected number of iterations)

+ duration to send packet

Lat Sp + Sal1113872 1113873 Rl + Rs( 1113857

Rl minus Sp

(18)

Advantages

(i) It increases energy efficiency(ii) It reduces data latency(iii) It increases throughput

2115 A-MAC (Advanced MAC) A-MAC is a TDMA-based protocol whose design is aimed at enhancing thelifetime of the network for low rate and reliability of datatransmission [39] To assign time slots it does not depend ona central controller like most protocols but it gathers in-formation from its neighbors so as to allocate the slotsthrough a distributed algorithm Energy is controlled byscheduling power when there is not a single event Its designcomprises many frames with each having many time slotswhich at the beginning have got a beacon transmitted tosynchronize and exchange information with neighborsNodes to participate in the subsequent session are selectedby the controlled node so they do not go into the sleep modewith others [13] -is protocol operates under four time slotdivisions namely initial wait discover and active states Anode in the initial state begins by listening to the medium forthe beacon packet from other nodes for network synchro-nization which when received makes it change the timerthrough subtraction of beacon transmission time from re-ception time To avoid synchronization problems due todrift and allow for continuity the node has to get thestrongest beacon signal It then enters the wait state afterchoosing a number of waiting frames so as to reduce theprobability that nodes get into the discover state simulta-neously In case the synchronization beacon is lost while stillin the wait state it will return to the initial state for anotherone On expiry of the waiting counter it enters the discoverstate and collects information from neighbors by listening totheir beacon messages On successful selection of a time slotit enters the active state where it endlessly transmits beaconsat the initiation of the slot A node goes to sleep when abeacon has been transmitted and there are no more data fortransmission or in case the neighborrsquos beacon received in-dicates no data packets are coming in [39]

Advantages

(i) It can enable multicasting(ii) It enhances the lifetime of the network(iii) It offers reliable data transmission

2116 L-MAC (Lightweight MAC) Based on TDMA thisprotocol is aimed at minimizing the number of transceiverswitches adapting the nodal sleep interval to data trafficamount and making implementation less complex [40] Apacket is transmitted comprising a control and a data partControl packets have a fixed size and the time slot con-trollerrsquos ID -ey show the distance from the transmittingnode for simplified routing to the network gateway as well asaddressing target node and reporting data unit length andinternode synchronization Adjacent nodes are more in-terested in getting control packets from nearby nodes -enodersquos transceivers are switched off until the subsequenttime slot unless the node is also addressed or the message isomnicast If so it listens to the data unit which may not fillthe remaining time slot in totality-e transceivers on eitherside are turned off when the transfer of packets is successfulEnergy wastage through idle listening in uncontrolled timeslots is avoided via a short time-out interval Network nodes

Journal of Computer Networks and Communications 13

are able to communicate without colliding-is saves energyand increases the lifetime of the network A node in L-MACcan only transmit one message per frame [13] -e networkis set up in such a way that all nodes are not synchronized onbeing switched on-e gateway takes control of a timeslot soas to have them synchronized and its immediate neighborgets the control messages thereby synchronizing their timerswith the gatewayrsquos Neighboring nodes learn of multiplegatewaysrsquo time slots in their vicinity just after one frame-ejust synchronized nodes randomly choose time slots forcontrolling as long as they are not occupied -e nodes thuspreserve a table for the neighboring nodes which makes atimeslot reusable after a minimum of 3 hops and preventsmessage collision All nodes keep their slots before thebattery is down or learn of an impending collision All nodescan track their hop distances to a selected gateway node andthe information is communicated in control messages [40]

Advantages

(i) It minimizes the number of transceiver switches(ii) It adapts the nodal sleep interval to data traffic

amount(ii) It makes implementation less complex(iii) It offers improved network lifetime(iv) Energy wastage through idle listening in un-

controlled time slots is avoided via a short time-out interval

Disadvantages

(i) L-MAC is not very good for mobile SNs since slotsare calculated only once

(ii) It suffers idle listening because nodes keep listeningto slotsrsquo control parts so as to get data and enableother nodes to join the network all the time [13]

2117 C-MAC -is was designed for WSNs with 3 seg-ments namely (i) aggressive RTS having double channelcheck to assess the channel (ii) anycast to initialize the flowand (iii) convergent packet forwarding to stabilize the flowIt promises an improvement in energy efficiency lowerlatency and enhanced throughput while avoiding syn-chronization overhead though it operates only in low duty-cycling apps especially where there is less traffic But whenthere is more traffic the protocol employs anycast-basedpacket forwarding for waking the nodes up or finding aforwarder quickly and converges from route-suboptimal toroute-optimal unicast After this energy could be preservedby the nodes using a synchronized wake-up plan [13]

Advantages

(i) It improves energy efficiency(ii) It lowers latency(iii) It enhances throughput avoiding synchronization

overhead

Disadvantages

(i) It operates only in low duty-cycling applicationsespecially where there is less traffic

22 MAC Protocols for WMSNs

221 Diff-MAC -is is a CSMACA-based QoS-awarehybrid-priority-based MAC protocol for WMSNs It targetsenhancing the channel utilization while making use of ef-ficient service differentiation to do coordination for mediumaccess of each traffic class while giving fair quick datatransfer It offers QoS using the following features [14]

(i) It balances energy consumption and delay throughsensor node duty-cycling adaptation depending onthe dominant traffic class

(ii) It has a feature for fragmentation and messagepassing that breaks lengthy video frames intosmaller packets that are transmitted as a burst thusreducing retransmission costs if there are any MACfailures

(iii) -ere is fair data delivery amongst the nodes andtraffic classes respectively resulting from intranodeand intraqueue prioritization reducing unbearableperformance

(iv) -e number of collisions can be reduced and packetlatencies are kept within limits by adjusting the sizeof the congestion window in this protocol as per thetraffic needsAdvantages

(i) Diff-MAC is very suitable for WMSNs since itadapts quickly to changes in networkconditions

(ii) It is a fair protocol(iii) It is scalable

Disadvantages

(i) It is quite difficult to monitor the statistics ofthe network

(ii) Adapting dynamically to the network condi-tions is not an easy task

(iii) It exhibits some degree of latency in packetdelivery

-e implementation of Diff-MAC on Imote2 indicatesthat resource requirements can be met on the sensorhardware Results of extensive simulation runs showed thatDiff-MAC outperforms its competitors

222 Cluster-Based On-Demand Multichannel MAC Pro-tocol for WMSN (COM-MAC) -is is an energy-efficientprotocol in WMSNs with high throughput and reliable datatransmission It achieves energy efficiency by schedulingmultichannel media access to reduce channel contentionand remove collision idle listening and overhearing [41]

Design it has a three-phased operation with the (i)request phase (ii) scheduling phase and (iii) datatransmission phaseRequest phase here all nodes have two protocols eachfor sending requests to the cluster head (CH) ie thecontention-based and contention-free TDMAFDMA

14 Journal of Computer Networks and Communications

protocols -e contention-based protocol allocates achannel per node for sending the request message andfree channels are useable as control channels Forcontention-free protocols a control slot is allocatedwhen a network is deployed Here a slot per channel isallocated to a different node enabling requests to be sentwithout interfering with transmissions from othernodes -e contention-based protocol operates undertwo steps (i) Control channel assignment phase achannel is assigned per node for transmitting a request(REQ) message A channel is assigned to multiplenodes and nodes are evenly distributed to availablechannels for congestion control (ii) Request trans-mission phase nodes with data send a REQmessage viathe assigned control channel to notify the CH whichresponds with an ACK message -e contention-freeTDMAFDMA protocol too has two phases includingthe control slot assignment and request transmissionphases For unreliable channel conditions and lowtraffic load the contention-based protocol is fit for usebecause of less collisions and congestion whereas thecontention-free protocol is preferable in heavy trafficload and reliable channel applicationsScheduling phase from the above phase a schedule isgenerated by the CH for coordination and this enablesnodes to transmit and broadcast their data via controlchannels-e schedule for broadcasting only comprisesdata for nodes assigned to a channel -e schedule isrebroadcast for transmission reliability enhancementbut for energy efficiency the nodersquos transceiver isswitched off when all the schedule is completely de-livered till it is the time for transmitting again-e timeslot and data communication radio channel are allincluded in the schedule generated for each nodeData transmission phase on getting the schedule thesensor node sends its data as per the time slot and thechannel assigned -e time slot is divided into datatransmission andACK sections ACK supports link layererror control Using implicit selective repeat ARQ theCH sends an ACK message for all received packets Ifthere is no ACK the packet is assumed lost and hence aretransmission during the subsequent interval is done-is enhances reliability of transmissions though excessdelay is possible here which is also intolerable for real-time critical apps like video or audio -is gives rise to ahybrid MAC protocol to exploit the unused spectrumduring data transmission sessions [41]Advantages

(i) It achieves energy efficiency by scheduling mul-tichannel media access to reduce channel con-tention and remove collision idle listening andoverhearing

(ii) It maximizes network throughput by dynamicallyallocating time slots and channels for nodes byexecution of a traffic-adaptive QoS-aware sched-uling algorithm depending on needs for QoS andtraffic data information

(iii) It increases transmission reliability by in-corporating spectrum-aware ARQ to avoid spec-trum wastage

223 EQoSA -is protocol is developed for video andimage transfer across WMSNs in a hybrid mode withprovision of QoS Concerning the active sensor nodes andtheir traffic loads it employs dynamic session sizes andchanges bitmap-assisted (BMA) MACrsquos fixed session sizeNodes declare any data available for transmission duringcontention time -is forces the cluster head to make aschedule of how slots have been assigned It communicatesthis schedule to the nodes giving each node a given numberof slots per session and so enabling the busty traffic to beaccommodated [14] -e protocol does not perform wellwith DiffServe mechanisms but does well with schemessupporting busty traffic loads and strong cluster heads arenecessary in WMSNs that can assign and broadcast slots

224 QoS-Supported Energy-Efficient MAC (QE-MAC)Based on the IEEE 80211e standard this protocol ensuresfairness and has low latency and jitter and better efficiencyon energy usage than other QoS-aware MAC protocols forWMSNs It operates under a two-phased routine

-e first phase introduces the innovating prioritymechanism in IEEE 80211e QoS and several data types areassigned priorities Its coordination function is a combi-nation of distributed contention-based channel allocationand centralized polling-based channel access mechanisms

-e second phase preserves the nodesrsquo energy usingdynamic duty cycling -ere is an exchange of RTSCTSpackets to supplement CSMACA by muting nodes near thesenderreceiver during packet duration [42]

Advantages

(i) It ensures fairness(ii) It has low latency and jitter and better efficiency on

energy usage than other QoS-awareMAC protocolsfor WMSNs

225 Black-Burst (BB) Contention A contention-basedMAC scheme for enhancing instantaneous QoS access tocarrier-sense WSNs is proposed in [42] In this schemenodes contend for the channel until it is free Real-timepackets are given higher priority over non-real-time packetsand if a node is transmitting real-time packets it is prior-itized over other nodes [15] Voice and video are some of thereal-time apps in consideration here which periodicallyaccess the common radio channel when transmitting -eseneed constrained end-to-end delay of packets at the MAClayer -is protocol majorly does the following [42]

(i) Prioritization of real-time traffic(ii) Implementation of the round-robin mechanism in

real-time nodes(iii) Bounded access delays among real-time packets

Journal of Computer Networks and Communications 15

(iv) Implementation of real-time communication withvarying needs of bandwidth good for multimediatraffic

Amechanism for dynamically assigning packet prioritiesbased on their deadlines and the traversed hops is proposedin [43] -e ReAllocative Priority (ReAP) mechanism ismeant to provide QoS at the MAC layer to video traffic -isscheme introduces adaptive TXOP by modifying the TXOPdynamically based on the queue length [15] Other protocolsin this category include dynamic duty cycle and adaptivecontention window-based MAC protocol frame sharing(FRASH) MAC protocol and real-time independentchannels (RICH) MAC protocol

Advantages

(i) It can implement real-time communication withvarying needs of bandwidth

23 Challenges and Opportunities(a) One of the challenges in satellite transmission is

cross-talk leading to interference in different fre-quency bands It is necessary to increase the rise andfall times as well as protect sensitive nodes as wecircumvent floating ones if we are to regulate cross-talk as in satellite communication therefore Wefurther need to investigate the possibility of usingdifferential signaling -is constitutes an opportu-nity for further research

(b) Another challenge is that the hidden and exposedterminal problems are still noticeable despite the useof RTSCTS solutions in MAC since collisions arestill possible from transmissions of separate nodesAt the MAC layer ACKs might block efficientchannel usage by prohibiting exposed terminalsfrom reusing the channel and may be eliminatedleaving no proof of sender data packet receptionUsing a directional antenna is one of the proposedtechniques of high-speed WSNs like IEEE 80211-is constitutes an opportunity for further study intothese issues for better solutions though

(c) Another challenge is that since nodes in contention-based MAC protocols are mobile there are manycollisions by the packets and schedule in-consistencies in schedule-based protocols in the two-hop neighborhood information when nodes areentering or leaving [44] Deng et al [45] propose amobility-based clustering (MBC) protocol for mo-bile nodes to enhance the packet delivery rate -isprotocol in comparison with CBR has 25 chancesof decreasing collision-induced packet losses and50 chances in comparison with low-energyadaptive clustering hierarchy (LEACH)

(d) -e process of analyzing and evaluating routingperformance for WMSNs is quite complex becauseof the bursty multimedia traffic Because of the wayencoding scheme frames are structured as well as thechanges in between the scenes that occur naturally

the compressed video frames display some substantialburstiness on several time scales -ere will also befrequent route recalculation that is as well energy in-tensive for nodes frequently capturing and transmittingenergy -e solution here could as well possibly be theuse of the MBC protocol Route recalculation tooshould be dynamic to cater for the mobile nodes whichmight be in and out of the range at different instants oftime Recently the use of smart dumpsters for effectivewaste management in smart cities has also been pro-posed as a possible solution [46]

(e) For WSNs their routing protocols do not look atmultimedia applications which require enormousbandwidth and processing energy and should betransmitted in real time with utmost fidelity Extrachallenges in routing are imposed in WMSNs indeployment of heterogeneous sensors since therecould be audio video and still pictures in addition toscalar data during transmission Heterogeneous datahandled also have different QoS requirements andwith many different business needs different ser-vices with varying requirements give rise to a bigchallenge for the routing design -ere is a need todesign protocols that can support WMSNs such asEQoSA [14] for this purpose

Table 2 summarizes the reviewed MAC protocols forWSNs and the ones designed for suitability in WMSNs

3 QoS Strategies at the Application Layer

We discuss many protocols used to solve different needs ofcommunication between machines

31 ConstrainedApplication Protocol (CoAP) It is a requestresponse protocol designed by the Internet Engineering TaskForce (IETF) for constrained devices such as those havinginsufficient RAM or CPU and WPANs with low-powerconstraints resulting in poor transfer of packets andenormous overhead -e protocol was designed for ma-chine-to-machine (M2M) apps and system automation toreduce bandwidth requirements increase packet transferdecrease the overhead and make work simpler leading tolightweight implementation HTTP commands like GETPOST PUT and DELETE are used in a client-server ar-chitecture Using the publishsubscribe architecture CoAPcan support multiple users and hence yield better perfor-mance via multicasting enabling asynchronous messageexchange -is protocol can support both multicasting andunicasting via UDP and hence is a reliable and simpleprotocol -is is done via the messaging sublayer (for re-liability based on UDP) and the requestresponse sublayer(for communication and interaction) [16]

It uses the following messages

(i) Confirmable message this guarantees reliabilityusing ACK messages sent by the receiver node-ese could be synchronous or asynchronous ifthere is a need for more computational time

16 Journal of Computer Networks and Communications

(ii) Nonconfirmable message here no ACK messagesare sent back

(iii) Acknowledgment (ACK) message this confirmsreceipt of a confirmable message

(iv) Reset message this confirms receipt of an un-processed message in case any critical parts nec-essary for interpreting the message are missing

(v) Piggybacked response the receiver directly re-sponds while getting the ACK message

(vi) Separate response the reply by the receiver is dis-crete from the ACK message

CoAP has no in-built security features It is secured bythe datagram transport layer security (DTLS) that runs ontop of UDP though it was not designed for the IoT -ismakes its suitability debatable since DTLS offers no supportfor multicasting It also needs more packets for handshakingthereby increasing the traffic volume and computationalresources and so shortening the lifespan of mobile devicessince they use batteries [17]

Advantages

(i) It can support multiple users and hence yieldbetter performance via multicasting

Table 2 Comparison of MAC layer protocols used in QoS provisioning

Protocol Features QoS parameter Comments Access Priorityassignment Energy eff Designed for

S-MAC[22] Collision avoidance Energy latency Requires periodic

synchronization CSMA Hybrid Medium WSNs

B-MAC[23] Collision avoidance Energy delay

throughput Flexible interface CSMA Dynamic High WSNs

T-MAC[25] Collision avoidance Energy Complex nonscalable CSMA Dynamic High WSNs

C-MAC[13]

Dynamic channelallocation Energy Multichannel MAC CSMA Dynamic Medium WSNs

CU-MAC[24]

Multichannelmechanism

-roughputlatency High packet delivery ratio TDMA Dynamic mdash WSNs

A-MAC[39]

Per flow serviceguarantees Delay m-ary tree model

dynamic prioritiesTDMACSMA Dynamic Medium WSNs

ER-MAC[30]

Emergencyresponse

Energy delaydelivery ratio

Adapts to traffic andtopology changes

TDMACSMA Hybrid Higher than Z-

MAC WSNs

RI-MAC[31] Collision detection Delay packet

delivery ratioHigh packet delivery ratio

and throughput CSMA Dynamic Medium WSNs

SRI-MAC[33] Collision avoidance Energy delay Good data transfer and

energy consumption TDMA Dynamic High WSNs

H-MAC[34] Slotted sleep time -roughput

latencyData transmission only in

sleep timeCSMAAloha Dynamic Medium WSNs

Z-MAC[36] Collision avoidance Energy delay

throughputVery good channel usage

and error syncCSMATDMA Dynamic Medium WSNs

Q-MAC[37]

Intrainternodescheduling Energy High QoS improved

energy efficiency CSMA Hybrid High WSNs

X-MAC[32]

Overhearingavoidance

Energy delaythroughput

Reduces excessivepreamble by strobing CSMA Adaptive

static Medium WSNs

L-MAC[40] Collision avoidance Network

lifetime energyReduces the number oftransceiver switches TDMA Static Medium WSNs

WiseMAC[38]

No downlinkchannel collision Power delay Significantly low-power

consumptionAlohaCSMA Static Higher than

ZigbeeInfrastructure

WSNsAR-MAC[27] Collision avoidance Energy delay Uses data packets and

control packets TDMA Static Higher thanIEEE 802154 WBANs

BB-MAC[42]

Real-time trafficprioritization Delay Employs round robin in

real-time nodesCSMATDMA Dynamic High WMSNs

COM-MAC [41] Collision avoidance Packet delay

throughputBalances reliability and

retransmissionTDMAFDMA Dynamic High WMSNs

EQoSA[14]

Busty traffic loadaccommodation Energy Provides QoS support for

video and imagesTDMACDMA mdash High WMSNs

QE-MAC[42] Collision avoidance Energy jitter

latencyRuns central polling-like

channel access CSMA Dynamic High WMSNs

R-MAC[26] Collision avoidance Energy delay Reservation based mdash Static High Underwater

SNs

Journal of Computer Networks and Communications 17

(ii) It enables asynchronous message exchange(iii) It is a reliable and simple protocol

32 Message Queue Telemetry Transport (MQTT)Developed by IBM and later improved by OASIS with anintention of decreasing bandwidth MQTT is a simple open-source publishsubscribe protocol analogous to the client-server version It is appropriate for lightweight IoT appli-cations and M2M transmissions in constrained situationslike limited power computation capability and memory orbandwidth -e network bandwidth and the computationalresource usage are decreasing since clients are not updatedmaking publishsubscribe protocols outperform requestresponse protocols via IoT requirements [16] -is protocolcan run over TCPIP and so has the following advantages

(i) Delivering of packets with reliable guarantee(ii) Enabling multicasting(iii) Remote device session establishment(iv) Reduction of transport overhead and protocol ex-

change to minimize network traffic(v) Provision of the notification mechanism when

anomalies occur [47]

MQTTrsquos design targets economical use of bandwidthand battery so it has a major application in FacebookMessenger MQTT ensures QoS and reliability of messagedelivery by providing the following three QoS levels

(1) Fire and forget a message is sent once depending onthe networkrsquos best effort and there is no need forACK (lowest QoS level)

(2) Delivered at least once a message is sent at least onceand an ACK message is needed to avoid duplicates

(3) Delivered exactly once by a four-way handshake itensures a message is sent exactly one time (highestQoS level)-e architecture it has three key components (i)publishersmdashlightweight nodes sending data to thebroker before returning to sleep (ii) broker whichorganizes sensory data in topics and sends them tointerested subscribers and (iii) subscribersmdashIoTapps interested in data sent by sensors through thebroker as shown in Figure 3

MQTT is a simple lightweight protocol that is betterthan CoAP via low and high packet loss probability interms of throughput and exhibits lower latency in lowsampling rates [48] It supports a number of imple-mentations for embedded devices and mobile applicationsand offers web-socket support too Nevertheless it hassome shortcomings

(i) It is only employed in simple data types(ii) It exhibits high latency in high sampling rates(iii) It is unsuitable for real-time apps in such cases(iv) It does not provide automatic discovery andor

sufficient security at the protocol level [49]

33 Secure Message Queue Telemetry Transport (SMQTT)-is is an extension of MQTT that is based on lightweightattribute-based encryption using elliptic curve cryptogra-phy -e protocol has three entities as shown in Figure 4(i) publisher that publishes information under a certaintopic (ii) subscriber that obtains the same topic in-formation via a broker and (iii) public key generator(PKG) or broker that acts as the trusted third party -eprotocol has four phases namely (a) set-up phase (b)encrypt phase (c) publish phase and (d) decrypt phase

(a) Set-up phase publisher and subscriber provide aunique ID with attributes for registration with thePKG -is generates a secret master key set andpublic parameters that it publishes with set U A1A2 An For ciphertext policy-attribute-basedencryption (CP-ABE) the PKG gives a key set todevices with attributes [50]

(b) Encrypt phase based on the access tree the publishercreates an access policy An access tree is sent by thepublisher to the PKG for generating the key-gen-eration policy in the key policy- (KP-) based ABEalgorithm -e publishers know beforehand thetopics and subscribers to access as well as the rulesfor access Subscribers are issued with private keysfor all communication sessions Publishers in theCP-ABE algorithm come up with access trees andrules where the payload is encrypted by the pub-lisher who also issues more info needed for de-cryption and the rules With the help of public keycryptography data are encrypted giving rise to ci-phertext on the basis of KPCP-ABE and the receivergets SUBACK from the broker [51]

(c) Publish phase ciphertext is embedded in theSPublish command as payload and topic name isdefined in the header by the publisher -e brokerreceives the SPublish packet and replies withPUBACK prompting the publisher to respond withthe PUBREL packet -e broker then broadcasts totopic subscribers and deletes the data forwarding thePUBCOMP packet to the sender node

(d) Decrypt phase the encrypted information isdecrypted by the subscriber with the help of a secretkey Considering KP-ABE on verification that itsatisfies the access policy the subscriber invokes thePKG for a corresponding key which it sends afterverifying the request With CP-ABE subscribers usethe private key for decryption after authenticationthereby enabling offline interaction [51]Advantages

(i) -e information being encrypted makes it moresecure

34 MQTT for Sensor Networks (MQTT-SN) -is is a ver-sion of MQTT designed to operate under environments oflow bandwidth high link failures and short message lengthIt is a publishsubscribe (pubsub) protocol designed for

18 Journal of Computer Networks and Communications

UDP-based WSNs to visualize communication for low-power devices [5152] It comprises a publisher device actingas the MQTT-SN client to forward communications to agateway that also changes the MQTT-SN message to theMQTTmessage before delivering it to the broker that sendsit to MQTT-SN subscribers [51]

It has a gateway for changing protocols from MQTT toMQTT-SN (Figure 5) [52] -is protocol supports nodes in

the sleeping mode using an offline keep-alive procedure thatcan be used by battery-operated devices to get into thesleeping mode To support sleeping clients MQTT-SN has anew offline keep-alive procedure with which battery-oper-ated devices can go to the sleeping state in which theirmessages are buffered at the server until they are awake InMQTT-SN clients with no preconfigured gateway addresscan figure out the real address of their operational gateway

Publisher

Subscriber

Subscriber

Broker

Publisher

4G

Figure 3 Architecture of the MQTT protocol

Register by mentioning URI Register by mentioning URI

Access tree

Set-up

Publisher ReceiverPKG(broker)

Key management Key management

Encrypt data using public parametersand random key This is done using

CP-ABE Publish it as

Subscribe (topics)SUBACK

S-Publish (TopicEncrypt (data))S-Publish (TopicEncrypt (data))

Encryptand publish

PUBACK

PUBACK

PUBRELPUBRELPUBCOMP

PUBCOMP

Figure 4 SMQTT protocol

Journal of Computer Networks and Communications 19

within the network through a discovery procedure Simul-taneously we may have several gateways in a single wirelessnetwork that can be cooperative via load sharing

MQTT-SN architecture it is composed of MQTT-SN clients gateways and forwarders (Figure 5)rough the MQTT-SN gateway (GW) the MQTT-SN protocol is used by the MQTT-SN clients toconnect to the MQTT server Integration betweenthe MQTT-SN GW and the MQTT-SN server mightbe possible or not Translating between the MQTT-SN and the MQTT for a stand-alone GW is doneusing the MQTT-SN protocol If a GW is detachedfrom the network it can be accessed by the MQTT-SN clients through a forwarder is just encapsu-lates the received MQTT-SN frames that are on thewireless side before forwarding them to the GW un-altered It then decapsulates the received frames tooresending them to the client as sent by the GW [52]ere are 2 types of gateways (a) transparent gatewayand (b) aggregating gateway as shown in Figure 6

(i) Transparent gateway this launches and sustains adirect MQTT connection between the MQTT-SNclient and the MQTT server which is entirely re-served for end-to-end client-server messaging enumber of servers and that of MQTT-SN clientsconnected to the GW are the same is transparent

GW translates syntax between these protocols and iseasier to implement compared to the aggregatingGW But the MQTT server must support dierentsessions for all connected clients Some MQTTservers are congured to enable few parallel sessions

(ii) Aggregating gateway this maintains a single MQTTconnection to the server on behalf of all clients Allexchanged messages from the MQTT-SN client to theaggregating GW are directed to the GW which sortsthe information to forward to the server e aggre-gating GW decreases the number of MQTT sessionssimultaneously supported by the server in WSNs withmany battery-operated sensors and actuators (SAs)having limited storage and processing capabilities It ishowever more complex to implement compared to atransparent GWe general format of the MQTT-SNprotocol message has two or four message headeroctets and n octets of variable header that is optionaldepending on the message in question MQTT-SN has256 message types (MsgType) with those from0X1Eminus 0XFD 0X19 and 0XFF reserved for futureusage [51]

Advantages

(i) e protocol can operate under environmentsof low bandwidth high link failures and shortmessage length

MQTT-SN gateway

MQTT-SN

MQTT

MQTT-SN

MQTT-SN

MQTT-SN

MQTT-SN client

MQTT-SN clients

MQTT broker

MQTT-SN client

MQTT-SN client

MQTT-SN gateway

MQTT-SN forwarder

Figure 5 MQTT-SN architecture

20 Journal of Computer Networks and Communications

(ii) It is a pubsub protocol designed for UDP-basedWSNs to visualize communication for low-power devices

(iii) It can be optimal for low-cost battery-operatednodes having low processing power and lessstorage

(iv) It supports nodes in the sleeping mode using theobrvbarine keep-alive procedure used by battery-operated devices

35e Extensible Messaging and Presence Protocol (XMPP)e standardization of this protocol by the IETF made itpopular for usage over the Internet especially in instantmessaging and chatting platforms It is however susceptibleto spamming attacks and its failure to have worldwidesupport has made Google too withdraw its support of thestandard [17] e protocol is mainly useable in the client-server architecture but can also be applicable in client-clientXMPP communication [53] Among the application layerprotocols XMPP is the only one that supports both asyn-chronous (publishsubscribe) and synchronous (requestresponse) services while running over TCP but can still runover other protocols

Given the simplicity of its implementation and ease ofusability in XMPP instant messaging (IM) is suitable forXMPP and is implemented using the message stanzaltmessagegt ese stanzas are usable in one-to-one ormultiuser sessions or chatting or in error reporting andsending headlines Attribute type of the stanza denes themessage type e attribute type could be a chat group chatheadline normal or error An instant messaging network inXMPP can support many XMPP clients and servers with agateway congured to enable other IM network users toconnect to other XMPP users e XMPP server mainlycontrols sessions and XML streams to clients and servers inaddition to routing XML stanzas in the streams [53] eTCP ports 5269 and 5222 have been respectively registeredfor the XMPP server-to-server connections and XMPP clientconnections by the Internet Assigned Numbers Authority(IANA) XMPP enables specication of XMPP ExtensionProtocols (XEP) to enhance its usefulness [17]

XMPP security clients connect to the server usingtransport layer security (TLS) and simple authentication andsecurity layer (SASL) protocols which are at the same timeused by the servers for interdomain communications Fordata condentiality and integrity TLS will encrypt thestreams of XML for all sessions To conrm the identity of anyclient that needs to access the server the SASL protocol doesthe authentication of the XML stream For any communi-cating session with a server to start the client needs to resolvethe serverrsquos DNS hostname For protection of the server of theclient from third-party attacks the server maintains the au-thentication details secret ie IP address and access methodand only original connections of the server are required

In server-to-server communications the SASL protocolis employed for authentication and condentiality Server-to-server communications may as well be deactivated from aserver by an administrator depending on the organizationalpolicy To guard against domain-spoong attacks serverdialback can be a good solution since servers support itsecuring more the XML stanzas Nevertheless it cannot beused to authenticate secure or encrypt in-server streamsand the server identication resulting therein is not strongenough to be relied on It further cannot secure against DNSpoisoning attacks and IP session hijacking for remote do-mains [53] High-security domains can use TLS and SASL

Advantages

(i) It is popular for usage over the Internet especially ininstant messaging and chatting platforms

Disadvantages

(i) It is susceptible to spamming attacks(ii) It has failed to have worldwide support making

Google withdraw support of the standard(iii) It is not secure against DNS poisoning attacks and

IP session hijacking for remote domains

36 Advanced Message Queuing Protocol (AMQP) isprotocol can be utilized by a number of platforms formessage exchange since it is open source and written inmany languages It is a publishsubscribe model depending

MQTT-SNclients Broker

MQTT

(a)

MQTT-SNclients Broker

(b)

Figure 6 (a) Transparent gateway and (b) aggregating gateway

Journal of Computer Networks and Communications 21

on a reliable and efficient messaging queue with capacity forinstant messaging or e-mail for different applications es-pecially in IoT environments It differs from others in thesense that message types can be specified and have theirsource traceable and how to do trade-off between perfor-mance security and reliability For reliability of messagetransmission AMQPmakes use of some delivery guaranteeslike at most once at least once and exactly once It can alsohowever utilize the TCP transport layer for guaranteeingreliability -e AMQPrsquos publishsubscribe method has twocomponents namely (i) exchange queue meant for routingof messages to an appropriate order in the queue and (ii)message queue that acts as storage pending their delivery toreceivers [16] In this protocol model the broker can makerouting decisions contrary to othermessaging systems whereapps using the queue have in them embedded the decision-making logic for recipients or senders -e developer mayneed to check each of the affected apps so as to change thelogic for routing and delivery of the messages [54]

Advantages

(i) It is convenient(ii) It has capacity for interoperability between dif-

ferent vendors(iii) It can use open standards to enable connection

between business partners(iv) It can enable innovations based on AMQP foun-

dations [55](v) It can enable supporting of mission-critical ap-

plications in e-commerce(vi) It can guarantee reliable delivery of messages

Disadvantages

(i) It is not very appropriate for real-timeapplications

(ii) It may not offer automatic discovery(iii) Interoperability is not guaranteed(iv) It lacks open-source libraries for constrained

gadgets [49]

37 Data Distribution Service (DDS) A product of theObject Management Group (OMG) DDS is a pubsubprotocol meant for M2M communications with two sub-layers -ese are the data-centric pubsub sublayer re-sponsible for connecting anonymous data publishers tosubscribers and the data-local reconstruction sublayer that isoptional and integrates DDS into the application layer [56]-e DDS architecture connects system components asshown in Figure 7 Publishers and subscribers are discon-nected with respect to (i) time (ii) space (iii) flow andbehavior (iv) platforms and (v) programming languages[57] A DDS data publisher generates topics and publisher-subscriber coupling is represented as topic name data typeschema and publishers and subscribersrsquo QoS attributesDelivering of data is done by the publisher layer -e datawriter and publishers together make a decision about anynecessary alterations for subsequent forwarding to thesubscribers who need to send data to IoT applications Data

readers are responsible for delivering topics to the sub-scribers after reading through them [56]

Some of the major components of the DDS architectureinclude the following [57]

(i) Topic this is a logical path specifying publicationand subscriptionrsquos data type between DataWritersand DataReaders For a successful session to begintopic names types and DataWriters and Data-Readersrsquo QoS should correspond

(ii) Domain this is a logical setting for transmissionthat can be utilized for separating and optimizingnetwork transmission in a setup of similar appli-cations If DDS apps share a domain ID then theycan exchange data among themselves

(iii) DataWriter and DataReader these are terminalobjects for writing and reading messages re-spectively to and from a universal data space

(iv) Participant this is an object that stands for apublisher or subscriber of the DDS app in a do-main acting like a container for other objects

Advantages

(i) It is highly reliable and offers QoS support(ii) It is highly scalable(iii) It is appropriate for real-time apps(iv) It is fault tolerant(v) It has high level of interoperability(vi) It does not need brokers that may act as

bottlenecks(vii) It offers automatic discovery(viii) It has a user-defined data structure for topic(ix) It supports publisher and subscriber decoupling

Disadvantages

(i) It is originally developed for only stand-aloneLANs

(ii) A lot of memory is needed(iii) It has no open-source libraries to support con-

strained devices(iv) It is hard to design and configure compared to

other protocols [49]

38 RESTful Services REpresentational State Transfer (REST)-is is an architectural style based on a set of principles thatdefine and address networked resources It offers web servicesthat enable transmission from the HTTP device to the devicewithin the IoT architecture and enables clients to reach theserver [16]-is protocol utilizes HTTP commands to backupthe requestresponse messaging model HTTP is a popularand secure protocol in the WWW using TLSSSL [16] ARESTful application has the following characteristics

(i) State and functionality are separated into distrib-uted resources

(ii) All resources are distinctively named by the use ofunique HTTP commands (GET POST PUT orDELETE over the Internet)

22 Journal of Computer Networks and Communications

(iii) It is a clientserver stateless layered protocol thatsupports caching REST uses the same methods asHTTP for all requests or responses with POST andGET for creating and retrieving resources PUT forupdating and changing the state of resources andDELETE for removing resourcesAdvantages of REST over arbitrary web services(SOAP) in the IoT context

(i) It reduces overhead(ii) It reduces parsing complexity(iii) It is stateless(iii) It integrates more tightly with HTTP [58](iv) It is easy to implement and learnis together

with its natural t over HTTP makes it apreferred method for Web 20 applications toshowcase their data and secure M2M sessionsin the IoT

(v) Applications that support RESTful web ser-vices exhibit better performance than the re-source-constrained WSN nodes [59]

(vi) REST is important for the IoT since it supportscommercial M2M cloud platforms

(vii) Its implementation in smart phones and tabletapps is also easy since it needs only an HTTPlibrary that is accessible to all operating systemplatforms [56]

Disadvantages

(i) Although it enjoys wide usage in commercialM2M platforms it is not likely to dominatebecause it is hard to implement

(ii) Its HTTP usage makes it incompatible withconstrained M2M devices leaving its usage toend applications

39 WebSockets It is a web-based protocol operating on asingle TCP channel and providing full-duplex transmissions

It permits secure duplex communication between a clientexecuting unreliable code in a restricted environment and adistant node communicating from that code while using theorigin-based security model employed by web browsers overTCP [60] e protocol is not a requestresponse model andis not a publishsubscribe model So unlike other protocolsa web-socket session is established by the client initiating ahandshake to the server to launch the session is onceestablished a full-duplex client-to-server connection beginsasynchronously and runs until terminated by either party[56] It gives a model for browser-based apps utilizing bi-directional communication with no need of many HTTPconnections to servers

WebSocket comprises two parts ie (i) the handshakehaving a clientrsquos message and the serverrsquos handshake responseand (ii) the data transfer HTTP headers should be repeated byapplications in all client requests and server responses arisingfrom the endless polling which might increase the commu-nication overhead subject to the application [61] WebSocketcan be used to come up with scalable real-time web appli-cations since it can enable full-duplex transmission via onesocket across the Web Its security can be ensured with thehelp of TLSSSL is protocol is suitable for resource-con-strained devices and IoTapplications will not have a good userexperience with it since it operates a clientserver architectureNevertheless it can outcompete all TCP-based protocolsbecause of the following advantages

(i) It is t for real-time transmission(ii) It has improved security(iii) It minimizes transmission overhead(iv) It can provide ecient messaging systems using

WebSocket ApplicationMessaging Protocol (WAMP)[17]

310 Streaming (Simple) Text-Oriented Messaging Protocol(STOMP) is is a TCP-based protocol designed for textmessage-oriented middleware (MOM) with the use of

Topic

Datareader

Datareader

Datareader

Datawriter

Datawriter

Datawriter

Subscriber PublisherPublisher

Data bus (DDS domain)

Subscriber

Topic Topic

Figure 7 DDS architecture

Journal of Computer Networks and Communications 23

HTTP-like commands for interoperability among platformslanguages and brokers Data communication takes placebetween the client and the broker in line with commands(like CONNECT DISCONNECT ACK NACK SUB-SCRIBE UNSUBSCRIBE SEND BEGIN COMMIT orABORT) then headers in the form ltkeygt ltvaluegt (one perline) then a blank line and then body content terminatingin a null character Server-client communication takes placeover a MESSAGE RECEIPT or ERROR frame with headerand body content format being analogous MOM productssupporting STOMP include Apache ActiveMQ or FuseMessage Broker HornetQ Open Message Queue(OpenMQ) RabbitMQ and syslog-ng through its STOMPdestination plugin [62] In addition to these other appli-cation layer protocols include simple media control protocol(SMCP) lightweight local automation protocol (LLAP)simple sensor interface (SSI) lightweight M2M (LWM2M)XMPP-IOT and simple object access protocol (SOAP)

Advantages

(i) It is lightweight for constrained networks(ii) It has flexibility to choose quality of services with

the given functionality(iii) It is standardized by OASIS Technical Committee(iv) It is easy and quick to implement

Challenges and opportunities

Here security threats are imminent because of the open-endedness of the application layer [63] -ese could includethe following

(i) -ere is poor or lack of a security design for anapplicationrsquos function Coming up with a gooddesign to address this could be a good ideaopportunity

(ii) Some of the programs might have backdoorsevading otherwise secure controls to allow un-authorized access to network resources We requiremechanisms to patch up such loopholes

(iii) Applications that lack or have weak authenticationmechanisms can be easily targeted by unlicensedusers to abuse the network Such mechanisms needto be strengthened further

(iv) Excessively complicated access control rules maybecause of their misunderstanding or poor writinggive unauthorized access allowing dubious activityrather than protecting the network they purport toprotect -ese should be simple and clear to un-derstand by all users

(v) In booting workstations without disks and managingnetwork devices we usually use the TFTP protocolalthough access does not require authenticating theuser by usernamepassword -is makes an intruderable to gain easy access to configaccess info as longas he can guess the filenames -is needs to be fixed

A few steps can be used to secure the application layerTo protect network data transmitted across the network

applications need to strengthen their encryption and au-thentication mechanisms Apps too need to enable strictcontrols of privileges for data access using good mechanismsto balance between usability and effectiveness [63] Appli-cations handling sensitive data need detailed logging andaudit capability as well as testing and reviewing

Table 3 summarizes the application layer protocols dis-cussed with their weaknesses and differences among them

4 Cross-Layer QoS Strategies

41 Service-Differentiated Real-Time Communication Scheme(SDRCS) -is is an event-driven routing protocol that routesreal-time traffic by featuring a cross-layer packet-forwardingdesign in which it integrates the real-time routing functionalitywith a new priority-based MAC scheme -e protocol ap-proximates distributed packet traversal speed for traffic clas-sification and admission control on the basis of this design Itfurther localizes decision-making by prioritizing packet for-warding to maximize the speed of packets [64]

Advantages

(i) It improves bandwidth utilization bandwidthdegradation due to unschedulable packets can beavoided

(ii) It offers a lightweight packet schedulability estima-tion mechanism Here it uses received signalstrength and admission control and then earlymissed-deadline packet-dropping policies are made

(iii) It delivers high end-to-end throughput via highersource data rates alongside strict end-to-end latencyneeds

(iv) Nodes which meet real-time requirements of lesstraffic load and better channel quality are givenhighest priority when it comes to packet forwarding

-e SDRCS operates via 5 mechanisms namely (1) RSS-(Rich Site SummaryReally Simple Syndication-) basedgrouping (2) admission control (3) prioritized queuing(4) real-time MAC and (5) dynamic forwarding -isprotocol has a queuing policy that employs per-hopdeadline-based queues and the nodes schedule theirpackets using FIFO priority-based queues [1] -e prin-cipal features of this protocol are mentioned as follows

(i) It is event driven and is easily adaptable to anynetwork changes

(ii) It does not need additional hardware for locali-zation or multichannel transmission

(iii) -e cross-layer packet-forwarding strategy can beused when transmitting multimedia traffic

(iv) It is able to efficiently circumvent any voids therein

Early deadline miss (EDM) policy drops all unscheduledpackets -e SDRCS circumvents wrong packet dropsdue to EDM since it adjusts to network changes easily

42 Sensor Fuzzy-Based Image Transport (SUIT) -is pro-tocol regulates congestion using a scheme based on fuzzy

24 Journal of Computer Networks and Communications

logic It transmits packets with reduced quality to allowablelevels if there is congestion enabling it to transmit lower-quality packetsframes without being dropped-e numberof packets delivered per second thus rises and while thisenhances video streaming apps it is more appropriatelymade for those that quickly transmit non-real-time videosuch as in video surveillance SUIT can stream JPEG pic-tures other than the traditional predictive video codingmethods that are not suitable for sensor devices because oftheir constraints of energy memory and computationalspeed Progressive JPEG-PJPEG performs better via packettransmission rate and delay though it degrades the imagequality because of congestion [65] -is protocol employscross-layer methods for interlayer data exchange though itstill employs some layering making it still possible topreserve all advantages of the layered design since it hasapplication and transport layers but has noMAC or routinglayer

For congestion detection and control SUIT bases on thepercentage number of in-bound and out-bound frames perwindow number of contenders and buffer occupancy of thenext-hop node Since buffer management has an effect oneffectiveness and QoS inWMSNs it is useful in the transportlayer so the protocol prioritizes packets for better QoS SUITreorders packets by tracking the sequence numbers of thedelivered packets since they all have a source ID framenumber and sequence number that it uses to detect anymissing packets -e protocol however does not providepacket reliability [65]

Advantages

(i) -e protocol prioritizes packets for better QoS(ii) It can transmit lower-quality packetsframes

without being dropped

Disadvantages

(i) It does not provide packet reliability

43 Network Layer QoS Support Enforced by a Cross-LayerController (NLQS) -is scheme permits packet-level servicedifferentiation as a function of throughput end-to-endpacket error rate and delay [66] (see Figure 8) -is en-hances network layer QoS It comprises a cross-layer controlunit (XLCU) for configuring and controlling networkingfunctionalities at physical MAC and network layers [18]-is depends upon the unified logic which affects choices forapplication layer needs and status of functional blocks whichdo implementation of networking functions [66]

Advantages

(i) Cross-layer interactions can be controlled withoutweakening the upgradability modularity andsimplicity of designing the system

44 Cross-Layer Signaling Shortcuts (CLASS) CLASS has ahigh level of effectiveness flexibility and comprehensive-ness According to [67] it has the following features

(i) Direct signaling among remote layers leading tofaster transmission

(ii) Lightweighted internal message format standard-ized protocols are not suitable for use in internalsignaling since they are not usually lightweight forinstance transmitting against faults in the networkCLASS requires only three fields namely destina-tion address event type and event contents

(iii) -e setup of its exterior messages is standardconsidering external signaling ICMP and TCPIPare suitable for general messages and short notifi-cations respectively

(iv) Message control protocol should ensure optimizedand organized exchange of dense simultaneousmessages across layers for high efficiency andavoidance of possible conflicts -e protocol appliesto many cross-layer signaling situations

Table 3 Differences and problems with application layer protocols

Protocol Architecture Transport QoS options Security Weaknesses

CoAP Requestresponse UDP DTLS No in-built security features DTLS does not supportmulticast

MQTT Publishsubscribe TCP TLSSSL Insufficient security at the protocol level sincepayload values are not encrypted

SMQTT Publishsubscribe TCP CPKP-ABE Key revocation and group pubsub for distributedSMQTT are still a challenge

MQTT-SN Publishsubscribe UDP TLSSSL SSLTLS suffers from attacks like BEAST CRIMERC4 and Heartbleed

XMPP RequestresponsePublishsubscribe TCP X TLSSSL Susceptible to spamming attacks and lacks worldwide

supportAMQP Publishsubscribe TCP TLSSSL Inappropriate for real-time applications

DDS Publishsubscribe TCPUDP TLSSSL Memory intensive and no open-source libraries forconstrained devices

REST Requestresponse HTTP X HTTPS Hard to implement Uses HTTP so incompatiblewith constrained apps

WebSockets PublishsubscribeClientserver TCP X TLSSSL Bad user experience for IoT apps since it runs the

clientserver architectureSTOMP Clientserver TCP HTTP -e broker can act as a bottleneck

Journal of Computer Networks and Communications 25

Advantages

(i) e protocol is scalable and has a very lowpropagation delay

(ii) It is very ecient and exible since it uses directsignaling between any two layers

(iii) It is very comprehensive and eective

45 Cross-Layer and Multipath-Based Video Transmission(CMVT) CMVT is a hybrid of application and networklayers At the application layer it carries out the encoding ofvideo streams to video data frames (I- P- and B-frames) bymeans of the MPEG-4 encoding format At the networklayer route discovery and data transfer take place Twoalgorithms greedy forwarding and rollback are used todiscover a number of paths from the source to the sink nodevia the route discovery method [20] A specied node icalculates the evaluation of its adjacent node j using thefollowing equation

fij (1 minus α)d2(j D) minus d2min(i)d2max(i) minus d

2min(i)

+ αeinit(j) minus eres(j)

einit(j)

(19)

where fij evaluation value for the node i to j d2(j D) isthe distance from the node j to the destination node D

d2min(i) and d2max(i) are the minimum and maximum dis-

tances of neighbors of the node i to D respectively einit(j) isthe initial energy of the node j eres(j) is the current residualenergy of the node j and α is the energy coecient given asfollows

α emax(i) minus emin(i)

emax(i) (20)

where emax(i) and emin(i) are respectively the maximumand minimum energy left of all node irsquos neighbors

e network layer further transmits video streams withCMVT doing status evaluation choosing an appropriatecommunication path for any packets and the QoS guaranteelevel for the path i is calculated as

fi (1 minus ω)hisum hi

+ ωnisum ni

(21)

where fi is the evaluation value for the path i hi are the hopsfor the path i ni is the summation of packets sent via isum ni isthe summation of packets sent by sources and ω is theenergy consumption factor [18]

Advantages

(i) e CMVT protocol is very superior in mediatransmission particularly large-scale WMSNs

To network peers

Flow requirements

-

Atimes

Admissioncontrol

Capacityassignment

Next-hopselection

Coding rate

Data rate

QoScontracts

Cross-layer controlunit (XLCU)

Localization subsystem

Synchronizationsubsystem

Networkcoordinationunit (NCU)

UWB modulator

Channel coder

QoS scheduler

Georouting

Source coder

Rate control

Figure 8 Cross-layer controller architecture

26 Journal of Computer Networks and Communications

46 Cross-Layer Cooperative MAC (CoopMAC)CoopMAC [68] is based on the IEEE 80211 DCF approach-ere is a reduction in interference of neighboring cellsleading to uniform coverage in densely deployed networks[68] A connection is established by two-way authenticationbetween the sender and the receiver When a Clear-to-Send(CTS) packet and short interframe space (SIFS) are obtainedby the sender packets are transmitted right to the receiverwith no valuable cooperation If available the two nodes findout whether they can exchange data -is is done by the useof a helper identification signal to establish how feasiblecooperative communication is It is established in thepresence of a signal and if it is not available direct com-munication is activated For distributed systems helper-initiated cooperation is preferable because of the presence ofthe RTSCTS packets [69]

Advantages

(i) CoopMAC has an advantage over others due to thespatial diversity amongst the three nodes [70]

(ii) It has the potential to achieve significantthroughput and delay with no complexity in thedesign of the system

47 Cooperative MAC Protocol for Multihop Networks (M-CMAC) Just as CoopMAC M-CMAC is made in such away that high data rate stations help low data rate stations toforward traffic for broadcasting [71] -ere are also helperschosen in such a way that 2 fast-hop transmissions replace aslow hop transmission -e helper via whom there existsleast delay from the sender to the receiver is considered bestbecause of theminimum two-hop transmission rate and so isthe chosen neighbor to the duo Here the working as-sumption is that all nodes have their position coordinatesknown and thus the Euclidean distances between all pairscan be calculated and converted to the data rate for that linkAll nodes have cooperative tables (CT) of potential helperswith destination and helper MAC address Euclidean dis-tance and total distance through the helper Any sender withdata to transmit checks out for the existence of any helper inthe CT for the destination and if available forwards RTS forchannel reservation for single-hop duration M-CMAC hasgot an RTS format having a shape with five fields similar tothe one given as follows

Framecontrol Duration Source

addressDestination

addressHelperaddress

-e destination nodersquos address is kept by the source in thepart for the helper address that is in turn saved as the desti-nation address Nodes inspect helper and destination addressfields on receipt of RTS whereby the node acts as the othernodersquos helper in case the helper address field is not the sameand the helper transmits CTS back to the source if at all it wantsto forward data On delivery of the CTS packet the sourcetransmits to the helper which in turn sends the packets to thedestination that also sends ACK to the helper on receipt of thepacket In M-CMAC there exists a higher level of channelreuse (parallel transmissions) due to the increased number of

nodes therein that subsequently raises availability of helpers fordata forwarding -is results in increased throughput incomparison with CoopMAC and IEEE 80211 DCF [71]

Advantages

(i) It promises a higher throughput compared toCoopMAC and IEEE 80211 DCF

48 Cluster-Based Cooperative Routing (CBCR) Protocol-e CBCR protocol has a multihop data-forwarding func-tion realized at the link layer with cooperative links that useM-CMAC -is protocol encompasses two stages namelythe routing relay selection phase and data forwarding phase

(i) Routing relay selection phase all nodes announcetheir presence to their neighbors by broadcastingperiodical beacon messages that carry the MAC ad-dress of the node Each of these constructs a relaytable that contains all neighboring nodes with whichit is able to communicate -e node further broad-casts its neighbor list in case of any change to itsentries since the previous broadcast -e MAC ad-dresses of nodes next to the node X are contained inits relay tablersquos column one and the neighbor nodersquosrow has MAC addresses for neighbors of the adjacentnode Based on its relay table each node in-dependently chooses routing relays Choosing a nodeas a relay node depends on the number of nodes itconnectsmdashit should be the highest number [71]

(ii) Data forwarding phase a node with packets totransmit needs to first verify if the receiver is in thesame cluster and if that receiver has a helper to whichthe packet is thence forwarded but in whose absencethe packet is delivered right to the receiver In casethe intended receiver is in a different cluster therelay table will be checked to see whether that re-ceiver is reachable via other routing relays If it isreachable packets will be sent to the routing relaydirectly or via the helper if at all the relay has one Incase of destination unreachable via relays packetsare multicast to all relays by the node [71]Advantages

(i) It enables multicasting

49 MAC-PHY Cross-Layer Protocol A cooperative cross-layer standard for cooperation at the physical layer in next-generation WMSNs is developed in [70] It provides acomplete MAC layer algorithm that gives a supportive shellfor the PHY-MAC layer Like CoopMAC its scheme de-pends on the middle node for communication betweennodes-eMAC layer protocol has been changed to regulateinformation exchanges at the physical layer -e receivernode receives duplicate packets from source and helpernodes to decode the data [70]

410 MAC-Centric Approach -is cross-layer protocoltargets multimedia applications by using the MPEG-4

Journal of Computer Networks and Communications 27

scheme [21] It is characterized by 4 access categories(AC3ndashAC0) according to their priority in transmission It ismeant for supporting different QoS needs in upcomingvideo applications and enabling MAC layer differentiationfor H264 partitioning [18] Because of low bandwidth delayand other QoS challenges that cause inefficiency in trans-mittingmultimedia data inWMSNs a number of algorithmsdepending on IEEE 80211e have been proposed to supporttransmission of quality videos [72] Selecting an AC dependson the measures for QoS such as the loss rate and delay-us AC3 having the highest priority is mapped to theparameter set concept since the stream is sensitive to bit lossespecially in video transmission [21]

411 Adaptive Cross-Layer Forward Error Correction(ACFEC) In the ACFEC [73] model data packets are ex-changed among nodes through the access point (AP) whichoperates the infrastructure mode in which it adds FEC to videodata [72] -ese data are dealt by encapsulation using astreaming server to the receiver through thewireless AP as RTPpackets An adaptive FEC controller senses the category ofpackets out of the RTP header and recovers the header ofpackets from UDP -e encoder generates some error-cor-recting packets whose number is determined by the sourcepacket number of the block Multimedia transmissions aremonitored by the controller using MAC failure data and itscounter is increased by one if there is a failed transmission-econtroller uses the failure counter to change the packet numberproduced after transmitting a block [18] In case there are lostpackets it changes the redundancy rates and produces extrapackets to replace the lost ones and satisfy the requirements ofthe receiving node [72] -e FEC packet number is enhancedor reduced to satisfy the receiverrsquos requirements and stoppacket losses-is is done through accurate detection of packetlosses and adjustment of redundancy rates If all video datapackets are well received there will not be generation of FECpackets [73]

Advantages

(i) It promises a good QoS through packet loss re-duction and redundancy rate adjustment

412 Balanced Cross-Layer Fuzzy Logic (BCFL) DesignRouting Algorithm in WSNs A new fuzzy logic-basedrouting algorithm (BCFL) [74] was designed using disper-sion of the cross-layer parameter as the fuzzy logic inferencesystem input Each cross-layer parameter has a dynamicweight depending on the value of dispersion -e designcomes with some innovations as follows (i) for fuzzy logicinference system input the absolute parameter value issubstituted by parameter dispersion thereby significantlyreducing algorithmic complexity (ii) dispersion does notchange with the order of magnitude according to the dis-persion formula and (iii) the weight of the parameter de-pends on the size of its dispersion and the two are inverselyproportional to each other -is enables BCFL to have somenobility unlike other algorithms as per the following dis-tinguishing properties making it advantageous

(i) It has got simple if-then rules which remain con-stant even when the constraints increase

(ii) It is capable of dealing with many constraints withno increment in complexity

(iii) It has capacity to yield a more balanced solutionthan other algorithms

(iv) It is easily adaptable to changes in network con-ditions and topology even when the changes arefrequent like in underwater WSNs -e algorithm isuseable in choosing the CH in cluster-based routingprotocols [74]

413 Minimum Hop Disjoint Multipath Routing Algorithmwith Time Slice Load-Balancing Congestion Control SchemeMHDMwTS is a two-phased routing protocol comprisingpath build-up and path acknowledgment phases It consists ofmultiple sources each of which has three build-up disjointpaths namely primary alternate and back-up paths Onactivation the source node requests to build up a route to thenearest hop neighbor which is the path build-up phase [19] Inthis phase step one has source activation in which therequested node for path building adds its number and time-stamp and sends to the least hop-count neighboring node-isgoes on until the least time latency sink having the informationneeded to construct the primary route is reached [75] In steptwo there is path extraction when the new package fromanother path arrives whereby the extracted path is comparedwith the primary path If the node is shared the package will berejected or an alternative path will be searched for to get abackup through comparison of the preceding two paths Phasetwo is path acknowledgment where we have the third step inwhich the sink returns the ACK packet to the sender with pathinformation having nodes and their time information after ithas computed it using the timestamp [1819]

Advantages

(i) -e protocol minimizes end-to-end latency(ii) It enhances congestion control

414 Cross-Layer Optimal Design (CLOD) Authors in [76]come up with a CLOD for scheduling at the data link layerrouting at the network layer and controlling congestion atthe transport layer with an assumption of fixed link capacity-rough congestion control energy efficiency is improvedTransport layer congestion at the nodes is minimized bycompressed sensing (CS) in which transmitted bits are re-duced whereas optimally allocating resources decreasescongestion on the links at the data link layer CLODpromises minimized computational complexity and betterperformance in light traffic scenarios [77] Generating andstoring CLOD is more efficient compared to Gaussianrandom matrices It prolongs network lifetime and savesenergy

Advantages

(i) It does congestion control subsequently increasingenergy efficiency

28 Journal of Computer Networks and Communications

(ii) It reduces computational complexity(iii) It improves performance in light traffic scenarios

415 Challenges and Opportunities of the Cross-Layer Design(CLD)

(i) Different layers in a real network assume differentfunctions andor services A layer only communi-cates with its neighboring layers -e layered modeldegrades system performance because of many

features in wireless transmission QoS parameters inthe layers of WMSNs are shared by routing pro-tocols to optimize the performance Neverthelessthere is a need to design cross-layer models forincreased efficiency in routing [1]

(ii) -e physical layer plays a highly vital role in CLDRate adaptation and channel allocation take place atthe physical layer via signal processing to enhanceQoS End-to-end performance in wireless media isaffected by the changes therein given the effect on

Table 4 A summary of some of the reviewed cross-layer models

Protocol Layers Aims Comments QoS parameters

SDRCS MACPHY Routing real-time traffic Transmits multimedia traffic viacross-layer packet forwarding

-roughputlatency

SUIT Applicationtransport Transmission of non-real-time video Uses fuzzy logic to regulate

congestionTransmission rate

delay

CLASS Any twoDesign serves as a framework fordifferent implementations ofdifferent application scenarios

Scalable very efficient flexible andhas very low propagation delay

Propagation delayjitter

NLQS PHYMACnetwork

Permits packet-level servicedifferentiation as a function of

throughput packet error rate anddelay

Network layer QoS is enhanced -roughput delaypacket error rate

CMVT Applicationnetwork

Encodes video streams to video dataframes via MPEG-4 encoding anddoes route discovery and data

transfer

Greedy forwarding and rollback areused to find source-to-sink paths Energy

CoopMAC PHYMAC Offers spatial diversity among thethree nodes

Helper ID signal finds out howfeasible cooperative communication

is

-roughput ratedelay

M-CMAC PHYMAC Increases end-to-end throughputand packet delivery ratio

Euclidean distances between nodesare calculated and converted to the

data rate for a link

-roughputpacket delivery

ratio

CBCR PHYMACMinimizes control overhead andtime consumed in establishing thecooperative paths than M-CMAC

Energy consumption is moreuniformly distributed in a network

enhancing network lifetime

-roughputpacket deliveryratio energy

MAC-PHY MACPHYMeant to enable PHY layer

cooperation and maximize gains ofcooperation at the MAC layer

Leverages both spatial diversity andcoding gain -roughput delay

MAC-centric MACAPPL

Meant to support QoS needs in newvideo apps and enable MAC layer

differentiation for H264partitioning

Targets multimedia applications byusing the MPEG-4 scheme

Delay packet lossrate

ACFEC MACnetwork(UDP)

Meant to enhance the quality ofvideo streaming over 80211WLANs

and overcome packet losses

Adjusts redundancy rates toovercome channel fluctuations and

detect and reduce packet lossPacket loss rate

MHDMwTS mdash

Meant to provide reliable datatransfer with the multipath routingand load-balancing congestioncontrol method in WMSNs

More reliable than basic routingschemes for transport multimedia

data

Latency packagetransmit rate

BCFL mdash

Introduces dispersion into fuzzylogic-based routing and sets every

cross-layer parameter with adynamic weight

Can be used to select a CH in cluster-based routing protocols andproposes a dispersion formula

Node utilitydispersion

CLODDatalinknetworktransport

Prolongs network lifetime andachieves congestion control anddesigned for lightly loaded WSNs

Assumes fixed link capacity andintegrates compressed sensing

technology

-roughputaverage energy CS

error ratio

Journal of Computer Networks and Communications 29

the purposefulness of the protocols at the networklayer CLD offers solutions to conserving powerminimizing energy and controlling both the flowand congestion in the network as well as fault tol-erance so it is an opportunity for designers con-sidering other layers We further desire to developCLDs that shall make an impact on network oper-ation and get close attention [18]

(iii) CLD interface standardization the architecturemust deliver module functionality although thereexist some queries amongst modules about possibleinterfaces predicted by the necessity to share in-formation among remote protocol layers Technicalchallenges include developing designing andstandardizing of cross-layer interfaces and algo-rithms that satisfy the cross-layer optimization re-quirements among protocol layers

Table 4 summarizes these cross-layer models with a fewcomments

5 Future Research Direction

51 At theMACLayer In WMSNs achieving duty cycling istricky since video traffic is volatile Listening to adjacentnodes wastes energy We need to switch states from awake tosleep -is requires further research

Self-organizing networks and green communicationalthough most CRWSN energy-saving algorithms aredesigned to reduce node transmission power energyconsumption is inevitable during the operational mode ofBSs because of internal processing Scholars are trying withpolicy-makers to encourage a shift to greenWSNs to reduceoperational costs and carbon footprints With self-orga-nizing CRWSNs for automatic switch-off when idle energyis saved which calls for cognitive switching algorithms Sointegrating cognitive switching with radio resource allo-cation in CRWSNs is a promising research area [78]

52 At the Application Layer Considering a WMSN ar-chitecture end users access the network via the BS to andfrom which they send their sensed data We desire to getguaranteed QoS and energy efficiency by designing a newarchitecture in the form of hardware and software whichneeds to be investigated further for a fix of the same [1]

QoS and energy efficiency are very crucial in WMSNsand mostly real-time apps that require guaranteed band-width and throughput in their network lifetime Mostprotocols ignore base station (BS) mobility and WMSNnodes Traffic management telemedicine and battlefieldsurveillance apps require mobile nodes or BSs so designingdynamic routing protocols to be adaptable in these cir-cumstances is necessary

53 Cross-Layer Security of data transmitted on WMSNseg in military surveillance and e-commerce is key withQoS and energy efficiency WMSNs are vulnerable to attacks

like worms and sinkhole Moreover their computationpower is not that high complicating implementation ofstrong secure protocols -is needs further attention

For fairness and priority issues in CRWSNs nodes areassigned diverse priorities depending on importance andurgency of their data For better network performance thereis a need to develop fair resource allocation mechanismswith priority-based fairness in sensors-ere is still potentialin this research area [78]

We must develop multiobjective adaptive protocols tooptimize QoS metrics involved in routing for best trans-mission results to get a trade-off between different opti-mization metrics We need to design standardized cross-layer algorithms which are able to meet cross-layer opti-mization standards

6 Conclusion

We have reviewed the different QoS strategies for WSNs inthe context of IoT from the MAC layer and application layeras well as the cross-layer paradigm For the MAC layer wehave reviewed protocols for WSNs such as the hybrid of thecontention-free and contention-based MAC protocols andthose for WMSNs We further reviewed a number of ap-plication layer protocols including many machine-to-ma-chine requestresponse and publishsubscribe protocols Forsystem optimization cross-layer QoS strategies are veryimportant in wireless communication We have reviewed anumber of cross-layer strategies For all categories reviewedchallenges and opportunities are discussed Finally somepossible future directions are discussed for research andapplication showing a promising potential for future re-search in this relatively new area especially as more WMSNapplications emerge recently

Conflicts of Interest

-e authors hereby declare no conflicts of interest regardingthe publication of this work

Acknowledgments

-is research was funded by the National Natural ScienceFoundation of China (Nos 61602398 61672447 and61711540306)

References

[1] V Bhandary A Malik and S Kumar ldquoRouting in wirelessmultimedia sensor networks a survey of existing protocolsand open research issuesrdquo Journal of Engineering vol 2016Article ID 9608757 27 pages 2016

[2] I F Akyildiz T Melodia and K R Chowdhury ldquoWirelessmultimedia sensor networks applications and testbedsrdquoProceedings of the IEEE vol 96 no 10 pp 1588ndash1605 2008

[3] M Iqbal M Naeem A Ahmed M Awais A Anpalagan andA Ahmad ldquoSwarm intelligence based resource managementfor cooperative cognitive radio network in smart hospitalsrdquoWireless Personal Communications vol 98 no 1 pp 571ndash5922018

30 Journal of Computer Networks and Communications

[4] S A Alvi B Afzal G A Shah L Atzori and W MahmoodldquoInternet of multimedia things vision and challengesrdquo AdHoc Networks vol 33 pp 87ndash111 2015

[5] Q F Hassan A U R Khan and S A Madani Internet ofings Challenges Advances and Applications CRC PressTaylor amp Francis Group A Chapman amp Hall Group BocaRaton FL USA 2018

[6] Q Wang Y Zhao W Wang et al ldquoMultimedia IoT systemsand applicationsrdquo in Proceedings of the 2017 Global Internet ofings Summit (GIoTS) vol 2 Geneva Switzerland June 2017

[7] A Aslam and E Curry ldquoTowards a generalized approach fordeep neural network based event processing for the internet ofmultimedia thingsrdquo IEEE Access vol 6 pp 25573ndash25587 2018

[8] M Jridi T Chapel V Dorez L Bougeant and A Le BotlanldquoSoC-based edge computing gateway in the context of theinternet of multimedia things experimental platformrdquoJournal of Low Power Electronics and Applications vol 8no 1 p 1 2018

[9] H Noura A Chehab L Sleem andM N Rapha ldquoOne roundcipher algorithm for multimedia IoT devicesrdquo MultimediaTools and Applications vol 77 no 14 pp 18383ndash18413 2018

[10] T Balan D Robu and F Sandu ldquoMultihoming for mobileinternet of multimedia thingsrdquo Mobile Information Systemsvol 2017 Article ID 6965028 16 pages 2017

[11] H Shen and G Bai ldquoRouting in wireless multimedia sensornetworks a survey and challenges aheadrdquo Journal of Networkand Computer Applications vol 71 pp 30ndash49 2016

[12] A Ahmad M H Rehmani H Tembine O A Mohammedand A Jamalipour ldquoIEEE access special section Editorial optimization for emerging wireless Networks IoT 5G andsmart grid communication networksrdquo IEEE Access vol 5pp 2096ndash2100 2017

[13] B A Muzakkari M A Mohamed M F A Kadir andZ Mohamad ldquoRecent advances in energy efficient-QoS awareMAC protocols for wireless sensor networksrdquo InternationalJournal of Advanced Computer Research vol 8 no 38pp 212ndash228 2018

[14] M A Yigitel O D Incel and C Ersoy ldquoQoS-aware MACprotocols for wireless sensor networks a surveyrdquo ComputerNetworks vol 55 no 8 pp 1982ndash2004 2011

[15] A M Abbas and O Kure ldquoQuality of Service in mobile ad hocnetworks a surveyrdquo International Journal of Ad Hoc andUbiquitous Computing vol 6 no 2 p 75 2010

[16] M Q Shatnawi ldquoApplication layer protocols for the internetof things a surveyrdquo in Proceedings of the InternationalConference on Engineering amp MIS (ICEMIS) Agadir Mo-rocco September 2016

[17] V Karagiannis P Chatzimisios F Vazquez-gallego andJ Alonso-zarate ldquoA survey on application layer protocols forthe internet of things research motivationrdquo InternationalJournal of Advanced Research in Computer Science vol 8no 3 pp 1ndash10 2015

[18] M S Bernard T Pei Z Li and K Li ldquoQoS strategies forwireless multimedia sensor networks in the context of IoTrdquoLecture Notes of the Institute for Computer Sciences SocialInformatics and Telecommunications Engineering vol 4pp 228ndash253 2019

[19] A AlAmri and M Abdullah ldquoCross-layer quality of serviceprotocols for wireless multimedia sensor networksrdquo in Pro-ceedings of the International Conference on CommunicationManagement and Information Technology pp 649ndash658Cosenza Italy April 2017

[20] J Guo L Sun and R Wang ldquoA cross-layer and multipathbased video transmission scheme for wireless multimedia

sensor networksrdquo Journal of Networks vol 7 no 9pp 1334ndash1340 2012

[21] A Ksentini A Naimi and M Gueroui ldquoToward an im-provement of H264 video transmission over IEEE 80211ethrough a cross-layer architecturerdquo IEEE CommunicationsMagazine vol 44 no 1 pp 107ndash114 2006

[22] W Ye J Heidemann and D Estrin ldquoAn energy-efficientMAC protocol for wireless sensor networksrdquo in Proceedings ofthe 21st International Annual Joint Conference of the IEEEComputer and Communications Societies (INFOCOM 2002)New York NY USA June 2002

[23] J Polastre J Hill and D Culler ldquoVersatile low power mediaaccess for wireless sensor networks categories and subjectdescriptorsrdquo in Proceedings of the Second InternationalConference on Embedded Networked Sensor Systems Sen-Sysrsquo04 pp 95ndash107 Baltimore MA USA November 2004

[24] T Danmanee K N Nakorn and K Rojviboonchai ldquoCU-MAC a duty-cycle MAC protocol for internet of things inwireless sensor networksrdquo ECTI Transactions on ElectricalEngineering Electronics and Communications vol 16 no 2pp 30ndash43 2018

[25] T Van Dam and K Langendoen ldquoAn adaptive energy-efficientMAC protocol for wireless sensor networksrdquo in Proceedings ofthe 1st International Conference on Embedded NetworkedSensor Systems Los Angeles CA USA November 2003

[26] P Xie and J Cui ldquoR-MAC an energy-efficient MAC protocolfor underwater sensor networksrdquo in International Conferenceon Wireless Algorithms Systems and Applications (WASA2007) Chicago IL USA August 2007

[27] A Rahim N Javaid M Aslam U Qasim and Z A KhanldquoAdaptive-reliable medium access control protocol forwireless body area networksrdquo in Proceedings of the AnnualIEEE Communications Society Conference on Sensor Meshand Ad Hoc Communications and Networks (SECON) SeoulKorea June 2012

[28] H S Kwak S Ullah D H Kwak and C H Lee ldquoA powerefficient MAC protocol for wireless body area networksrdquoJournal of the Korean Institute of Intelligent Systems vol 8no 6 pp 131ndash140 2009

[29] C Omeni A J Burdett and M Park ldquoEnergy efficientmedium access protocol for wireless medical body area sensornetworksrdquo in Proceedings of the 4th IEEEEMBS InternationalSummer School and Symposium on Medical Devices andBiosensors pp 29ndash32 Cambridge UK August 2007

[30] L Sitanayah C J Sreenan and K N Brown ldquoER-MAC ahybrid MAC protocol for emergency response wireless sensornetworksrdquo in Proceedings of the Fourth International Con-ference on Sensor Technologies and Applications pp 244ndash249Mestre Italy July 2010

[31] Y Sun and D B Johnson ldquoRI-MAC a receiver-initiatedasynchronous duty cycle MAC protocol for dynamic trafficloads in wireless sensor networksrdquo in Proceedings of the 6thACM conference on Embedded network sensorsystemsmdashSenSysrsquo08 pp 1ndash14 Raleigh NC USA November2008

[32] M Buettner G V Yee E Anderson and R Han ldquoX-MAC ashort preamble MAC protocol for duty-cycled wireless sensornetworksrdquo in Proceedings of the 4th International Conferenceon Embedded Networked Sensor Systems SenSys 2006pp 307ndash320 Boulder Colorado USA October 2006

[33] S Boulfekhar and M Benmohammed ldquoSynchronous receiverinitiated MAC protocol for long-lived sensor networksrdquoComputers amp Electrical Engineering vol 40 no 2 pp 504ndash516 2014

Journal of Computer Networks and Communications 31

[34] S Mehta and K S Kwak ldquoH-MAC a hybrid Mac protocol forwireless sensor networksrdquo International Journal of ComputerNetworks amp Communications vol 2 no 2 pp 108ndash117 2010

[35] I Rhee A Warrier M Aia J Min and M L Sichitiu ldquoZ-MAC a hybrid MAC for wireless sensor networksrdquo IEEEACM Transactions on Networking vol 16 no 3 pp 511ndash5242008

[36] S Arshad A Al-sadi and A Barnawi ldquoZ-MAC performanceevaluation and enhancementsrdquo Procedia Computer Sciencevol 21 pp 485ndash490 2013

[37] Y Liu I Elhanany and H Qi ldquoAn energy-efficient QoS-aware media access control protocol for wireless sensornetworksrdquo in Proceedings of the IEEE International Confer-ence on Mobile ad hoc and Sensor Systems pp 9ndash11 Wash-ington DC USA December 2005

[38] J Decotignie ldquoWiseMAC an ultra low power MAC protocolfor the downlink of infrastructure wireless sensor networksrdquoin Proceedings of the Ninth International Symposium onComputers And Communications pp 244ndash251 AlexandriaEgypt June 2004

[39] R A Rashid ldquoDevelopment of energy aware TDMA-basedMAC protocol for wireless sensor network systemrdquo EuropeanJournal of Scientific Research vol 30 no 4 pp 571ndash578 2009

[40] H P Van Hoesel ldquoA lightweight medium access protocol(LMAC) for wireless sensor networksrdquo in Proceedings of theInternational Workshop on Networked Sensing Systems Bal-timore MD USA November 2004

[41] C Li P Wang H Chen andM Guizani ldquoA cluster based on-demand multi-channel MAC protocol for wireless multi-media sensor networksrdquo in Proceedings of the IEEE In-ternational Conference on Communications pp 2371ndash2376Beijing China May 2008

[42] J L Sobrinho and A S Krishnakumar ldquoQuality-of-service inad hoc carrier sense multiple access wireless networksrdquo IEEEJournal on Selected Areas in Communications vol 17 no 8pp 1353ndash1368 1999

[43] J P John ldquoProviding MAC QoS for multimedia traffic in80211e based multi-hop ad hoc wireless networksrdquo ComputerNetworks vol 51 no 1 pp 153ndash176 2007

[44] S Lofty and N Padmavati ldquoA survey on mobility basedprotocols in WSNsrdquo in Proceedings of the InternationalConference Communication and Manufacturing pp 12ndash15Hangzhou China July 2014

[45] S Deng L Shen and J Li ldquoMobility-based clustering pro-tocol for wireless sensor networks with mobile nodesrdquo IETWireless Sensor Systems vol 1 no 1 pp 39ndash47 2011

[46] M Khalil A Khalid F U Khan and A Shabbir ldquoA review ofrouting protocol selection for wireless sensor networks insmart citiesrdquo in Proceedings of the 24th Asia-Pacific Confer-ence on Communications (APCC) Ningbo China November2014

[47] O Message Q Telemetry and T Mqtt Committee Specifi-cation Draft 02 April 2014

[48] M Collina M Bartolucci A Vanelli-coralli andG E Corazza ldquoInternet of things application layer protocolanalysis over error and delay prone linksrdquo in Proceedings ofthe 7th Advanced Satellite Multimedia Systems Conference andthe 13th Signal Processing for Space CommunicationsWorkshop Livorno Italy September 2014

[49] A Talaminos-barroso M A Estudillo-valderrama L M RoaJ Reina-tosina and F Ortega-ruiz ldquoA machine-to-machineprotocol benchmark for e health applicationsmdashuse case re-spiratory rehabilitationrdquo Computer Methods and Programs inBiomedicine vol 129 pp 1ndash11 2016

[50] V Goyal O Pandey A Sahai and B Waters ldquoAttribute-based encryption for fine-grained access control of encrypteddatardquo in Proceedings of the 13th ACM Conference on Com-puter and Communications Security pp 89ndash98 AlexandriaVA USA October 2006

[51] M Singh M A Rajan V L Shivraj and P BalamuralidharldquoSecure MQTT for internet of things (IoT)rdquo in Proceedings ofthe Fifth International Conference on Communication Systemsand Network Technologies Gwalior India April 2015

[52] A Stanford-Clark and H L Truong MQTT for Sensor Net-works (MQTT-SN) Protocol Specification httpmqttorgnewwp-contentuploads200906MQTT-SN_spec_v12pdf 2013

[53] RFC ldquoExtensible messaging and presence protocol (XMPP)rdquoin Proceedings of the IETF pp 1ndash9 San Diego CA USAAugust 2005

[54] S Vinoski and I Technologies ldquoQueuing protocolrdquo IEEEInternet Computing vol 10 no 6 pp 87ndash89 2006

[55] J L Fernandes I C Lopes J J P C Rodrigues and S UllahldquoPerformance evaluation of restful web services and AMQPprotocolrdquo in Proceedings of the Fifth International Conferenceon Ubiquitous and Future Networks (ICUFN) pp 2ndash7 DaNang Vietnam July 2013

[56] M Asim ldquoA survey on application layer protocols for internetof things (IoT)rdquo International Journal of Advanced Researchin Computer Science vol 8 no 3 pp 996ndash1000 2017

[57] K An A Gokhale D Schmidt S Tambe P Pazandak andG Pardo-castellote ldquoContent-based filtering discovery pro-tocol (CFDP) scalable and efficient OMG DDS discoveryprotocolrdquo in Proceedings of the 8th ACM InternationalConference on Distributed Event-Based Systems New YorkNY USA July 2014

[58] E Web What is a web service pp 52ndash57 2010[59] M Laine RESTful Web Services for the Internet of ings

pp 2ndash4 2012 httpmediaTkk[60] IETF -e WebSocket protocol pp 1ndash71 2011[61] V Pimentel ldquoCommunicating and displaying real-time data

with websocketrdquo IEEE Internet Computing vol 16 no 4pp 45ndash53 2012

[62] Wikipedia 2019 httpsenwikipediaorgwikiStreaming_Text_Oriented_Messaging_Protocol2019

[63] K Raghavendra and S Nireshwalya ldquoApplication layer se-curity issues and its solutionsrdquo IJCSET vol 2 no 6pp 1266ndash1269 2012

[64] Y Xue B Ramamurthy andM C Vuran ldquoSDRCS a service-differentiated real-time communication scheme for eventsensing in wireless sensor networksrdquo Computer Networksvol 55 no 15 pp 3287ndash3302 2011

[65] C Sonmez S Isik M Y Donmez O D Incel and C ErsoyldquoSUIT a cross layer image transport protocol with fuzzy logicbased congestion control for wireless multimedia sensornetworksrdquo in Proceedings of the 5th International Conferenceon New Technologies Mobility and Security (NTMS) pp 1ndash6IEEE Istanbul Turkey May 2012

[66] T Melodia and I Akyildiz ldquoCross-layer QoS-aware com-munication for ultra wide band wireless multimedia sensornetworksrdquo IEEE Journal on Selected Areas in Communica-tions vol 28 no 5 pp 653ndash663 2010

[67] Q Wang and M A Abu-Rgheff ldquoCross-layer signalling fornext-generation wireless systemsrdquo in Proceedings of the 2003IEEE Wireless Communications and Networking 2003WCNC 2003 vol 2 pp 1084ndash1089 IEEE New Orleans LAUSA March 2003

[68] P Liu Z Tao S Narayanan T Korakis and S PanwarldquoCoopMAC a cooperative MAC for wireless LANsrdquo IEEE

32 Journal of Computer Networks and Communications

Journal on Selected Areas in Communications vol 25 no 2pp 340ndash354 2007

[69] H Shan H T Cheng and W Zhuang ldquoCross-layer co-operative MAC protocol in distributed wireless networksrdquoIEEE Transactions onWireless Communications vol 10 no 8pp 2603ndash2615 2011

[70] F Liu T Korakis Z Tao and S Panwar ldquoAMAC-PHY cross-layer protocol for wireless Ad-Hoc networksrdquo in Proceedingsof the 2008 IEEE Wireless Communications and NetworkingConference pp 1792ndash1797 IEEE Las Vegas NV USA March2008

[71] L Jacob and H R Shamna ldquoEfficient cooperative MAC androuting in wireless networksrdquo Transactions on Networks andCommunications vol 3 no 5 2015

[72] S Rao and K Shama ldquoCross layer protocols for multimediatransmission in wireless networksrdquo International Journal ofComputer Science amp Engineering Survey vol 3 no 3pp 15ndash28 2012

[73] L Han S Park S Kang and H Peter ldquoAn adaptive cross-layer FEC mechanism for video transmission over 80211WLANsrdquo in Proceedings of the 2009 International Conferenceon Internet pp 209ndash215 Las Vegas NV USA December2009

[74] N Li J Martınez and V H Dıaz ldquo-e balanced cross-layerdesign routing algorithm in wireless sensor networks usingfuzzy logicrdquo Sensors vol 15 no 8 pp 19541ndash19559 2015

[75] G Sun J Qi Z Zang and Q Xu ldquoA reliable multipathrouting algorithm with related congestion control scheme inwireless multimedia sensor networksrdquo in Proceedings of the2011 3rd International Conference on Computer Research andDevelopment vol 4 pp 229ndash233 IEEE Shanghai ChinaMarch 2011

[76] M Li Y Jing and C Li ldquoA robust and efficient cross-layeroptimal design in wireless sensor networksrdquoWireless PersonalCommunications vol 72 no 4 pp 1889ndash1902 2013

[77] J Yan M Zhou and Z Ding ldquoRecent advances in energy-efficient routing protocols for wireless sensor networks areviewrdquo IEEE Access vol 4 pp 5673ndash5686 2016

[78] A Ahmad S Ahmad M H Rehmani and N U Hassan ldquoAsurvey on radio resource allocation in cognitive radio sensornetworksrdquo IEEE Communications Surveys amp Tutorials vol 17no 2 pp 888ndash917 2015

Journal of Computer Networks and Communications 33

International Journal of

AerospaceEngineeringHindawiwwwhindawicom Volume 2018

RoboticsJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Active and Passive Electronic Components

VLSI Design

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Shock and Vibration

Hindawiwwwhindawicom Volume 2018

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawiwwwhindawicom

Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Control Scienceand Engineering

Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Journal ofEngineeringVolume 2018

SensorsJournal of

Hindawiwwwhindawicom Volume 2018

International Journal of

RotatingMachinery

Hindawiwwwhindawicom Volume 2018

Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Navigation and Observation

International Journal of

Hindawi

wwwhindawicom Volume 2018

Advances in

Multimedia

Submit your manuscripts atwwwhindawicom

Page 11: ReviewArticle - downloads.hindawi.com · ReviewArticle QoSStrategiesforWirelessMultimediaSensorNetworksinthe ContextofIoTattheMACLayer,ApplicationLayer,and Cross-LayerAlgorithms MuwongeSsajjabbiBernard

2110 H-MAC -is protocol is built on the power savingmechanism (PSM) of IEEE 80211 and slotted Aloha pro-tocols to enhance performance using multiple slots Timehere comprises large frames each having an active and asleeping part [34] If any node has got data to send it will askfor slots from the receiver node during active time and latertransmit the packets in the prenegotiated slots Nodes gointo the sleep mode during sleep-time slots if they do nothave data to transmit

Latency in IEEE 80211 if N is the number of hops nthe value for the current hop tcsn the back-off time andttx the transmission delay the total latency is calculatedusing the following equation

D(N) 1113944N

n1tcsn + ttx1113872 1113873 (8)

and the average latency is given by the followingequation

E[D(N)] N tcsn + ttx1113872 1113873 (9)

-roughput in H-MAC is given by the followingequation

Th npnm

tact + Cts (10)

where np is the number of transmitted packets nm is themaximum number of nodes connected in a time frameC is the number of slots with the same length tact is thelistening time and ts is the sleep delayAdvantages

(i) It enhances performance using multiple slots

2111 Z-MAC (ZebraMAC) -is rides on the strengths andoffsets weaknesses of TDMA and CSMA It yields very goodchannel usage and low delay under low contention and lowerscollision among two-hop neighbors at a reduced cost Itsperformance is strong via error synchronization failure inassigning slots and channel conditions that change with timeIn the set-up phase the protocol operates through certainsteps including neighbor discovery slot assignment localframe exchange and global time synchronization whichunless the topology of the network changes significantly donot run again When the data are being transmitted there isimprovement in network throughput and energy efficiencywhich covers up for the initial upfront operational costs [35]

Disadvantages

(i) Initial slot assignment is quite a cumbersome taskhere since it is done at the beginning yet this maynot be at the same time in reality It further as-sumes static links until setup happens again whichtoo is impractical

(ii) It also suffers from the hidden-terminal problemsince slot owners have a small contention windowand priority for slots allocated to them due to theECN messages

(iii) Synchronizing time also poses a challenge in thisprotocol due to a large clock drift between nodes

(iv) Source and destination nodes too are not wellcoordinated in Z-MAC [36]

2112 Q-MAC (QoS-Aware Media Access Control) -isprotocol offers high QoS with improved energy efficiencyand accesses the wireless channel using MACAW as theunderlying protocol [37] Traffic from different nodes isprioritized for QoS satisfaction depending on how criticalthe data are It mainly has two scheduling algorithms inWSNs (i) intranode and (ii) internode scheduling -eintranode scheduling scheme adopts a first-in first-out-(FIFO-) based queuing algorithm based on application andMAC layer abstraction for data packet classification In-ternode scheduling reduces idle listening and collision on thechannel for improved energy consumption

Advantages

(i) It gives a high QoS with improved energy efficiency

2113 WiseMAC (Wireless Sensor MAC) WiseMAC isbased on the Aloha protocol and employs preamble sam-pling in achieving low-power transmissions in infrastructuresensor networks by occasionally sampling the channel tocheck for activity [3238] -e protocol employs the samemethod as B-MAC except that the sender shortens theextended preamblersquos length by scheduling its transmissionsaccordingly after learning the receiverrsquos awake schedules-is is done by the receiver node putting its subsequentawake time on the ACK packet such that when the senderneeds to send to it again it starts the preamble shortly beforeit wakes up so as to avoid energy wastage in transmitting thepreamble [32] -e data frame is repeated instead of theextended preamble for less traffic transmissions that have alengthier preamble than the data frame in WiseMAC Afterprocessing the frame a node that is not the target recipientwill go back into the sleep mode while the recipient staysawake till the session ends thereby sending an ACK packetFor an idle channel the power consumption is rather lowwith this method Nevertheless there is a problem of longwake-up preamble limiting throughput and causing anenormous power consumption overhead in reception [38]All nodes overhearing a transmission bear the overheadtogether with the recipient node -e protocol can addressissues concerning low-power transmission but has nomechanism of nodes adapting to varying traffic patterns[32] To compensate for drift between clock at the accesspoint and on the sensor node there is a need to compute thewake-up preamble duration -e drift varies directly withtime since previous resynchronization and is given by thefollowing equation

TP min 4θl TW( 1113857 (11)

where θ is the time-base quartz frequency tolerance and l isthe interval between two transmissions Increase in trafficvolume naturally reduces overhearing

Journal of Computer Networks and Communications 11

Radiomodel for protocols with lower power transitiondelays between transceiver states and their powerconsumption are modeled Some of the states includeDOZE where the transceiver neither transmits norreceives but is prepared to power on to the receivetransmit state very fast RXwhere the transceiver listensto the channel receives data or demodulates data fromthe noisyidle channel and TX which is the trans-mission state by the transceiver Let TS be the set-uptime needed to turn the transceiver from the DOZE tothe RX or TX state TT the turn-around time needed toswitch between RX and TX states and B the trans-ceiverrsquos bit rate PZ PR and PT are the values of powerconsumed in the DOZE RX and TX statesrespectively

1113954PR PR minus PZ

1113954PT PT minus PZ(12)

where 1113954PR and 1113954PT are the changes in power as a result ofbeing in RX and TX states respectivelyTraffic model let N be the number of sensor nodesconnected to an access point (AP) and λ be the global rateat which the downlink Poisson traffic arrives at the APfrom the fixed network Traffic toward each sensor node isthe same and is given by λN Average packet interarrivaltime Lwith which data packets arrive at a node is given byL Nλ If TC and TD are the control packets and datapackets respectively assuming a small traffic volume theglobal interarrival 1λ far exceeds the summation ofdurations for the data packet and control packet and theturn-around time Mathematically it is given as

1λ≫TD + TC + TT (13)

-e average power consumed by WiseMAC andtransmission delay incurred are given by equations (14)and (15) [38]

PW PZ +1113954PR TS +(1B)( 1113857

TW+

1113954PR X + TD + TT( 1113857

L

+ 1113954PR(N minus 1)Y

L

(14)

where

X 2θL 1 minus eminus TD4θL( )1113874 1113875

Y T2D + 12TDθL

2TW1 minus e

minus TW4θL( )1113874 1113875

DW TD +TW

21 minus e

minus TW4θL( )1113874 1113875

+ 2θL 2 minus eminus TD4θL( ) minus e

minus TW4θL( )1113874 1113875

(15)

where PW and DW are the power consumed byWiseMAC and the transmission delay for WiseMACrespectivelyAdvantages

(i) It achieves low-power transmissions in in-frastructure sensor networks through preamblesampling while checking for activity

(ii) It reduces overhearing due to increased trafficvolume

Disadvantages

(i) It has a problem of long wake-up preamble limitingthroughput and causing enormous power con-sumption overhead

2114 X-MAC It is a simple minimal transmission powerMAC protocol that uses a short preamble with decouplingfor the sending and receiving node sleep plans inWSNs-isleads to a significant reduction in energy usage at sendingand receiving nodes and decrease in per-hop latency andgives flexibility in adapting to bursty and periodic sensordata sources [32]

Protocol design X-MACrsquos design is aimed at increasingenergy efficiency reducing data latency and increasingthroughput -is enables application in different kindsof packetized and bit stream digitized radios as well assimplified low-overhead distributed application -edesign is as such in order to address the low-powerlistening problems including overhearing excessivepreamble and incompatibility with packetizing radiosas in [32]Asynchronous duty cycling a node with information totransmit will send the packets after sending a preamblewhereas the remaining nodes will preserve their sleepschedules that are not synchronized -e receiver willsample the link on waking up and stays awake if it detectsa preamble and if it is the target it will return to sleep if itis not the target after getting the whole preambleEmbedding the target ID in the preamble to avoidoverhearing the lengthy preamble is divided into manyshortened packets embedded with the receiver nodersquosID so as to eliminate overhearing One lengthy pre-amble is constituted by the shortened preamble packetstream Any node on receiving the short preamble afterwaking up will examine the target ID on the packet Itgoes back into the sleepmode right away resuming dutycycling if it is not the targeted recipient or stay awakefor successive packets if it is the one -is approachchecks energy wastage due to network density It canalso be used on several types of radiosReducing excessive preamble using strobing it ispossible to enable low-power transmission and alsosave energy by reducing time for preamble trans-mission through the use of enhanced preambles inaddition to preamble sampling X-MAC does not sendan endless preamble packet stream but puts slight

12 Journal of Computer Networks and Communications

pauses in between packets to allow the sending node topause and listen to the channel and allow the receivingnode to resend an early ACK packet during that pause-is makes the sending node begin sending datapackets and the receiving node cut excessive preamblesand thus decrease per-hop latency and energy wastageon unnecessary waiting and transmissionPacketizing radios low-power listening is limited incapability for supporting packetizing radios butX-MACrsquos short strobed preambles can support all typesof digital radiosAdaptation to traffic load while a number of WSNapplications are reliable in traffic production they stillmust adapt to changing traffic Nodes are going to haveseparate sleep schedules depending on traffic loadConsidering systems that have time-fluctuating trafficall predetermined static schedules are suboptimalOptimality Buettner et al [32] model the expectationof energy as follows

Es (preamble energy + energy perACK listen)

lowast(expected preamble minus listen iterations)

+(energy to send parket)

Es PTxSp + PRxSal1113872 1113873 Rl + Rs( 1113857

Rl minus Sp+ SdPTx

(16)

where Es expected energy to send a packet PTx andPRx are the transmission and receiver power re-spectively Sp Sal and Sd are the duration of thesenderrsquos preamble ACK listen and data transmissionperiods respectively and Rl and Rs are the receiverrsquoslisten and sleep periods respectively Expected energyto receive a packet is given as

Er (listen cycle energy + sleep cycle energy)

lowast(expected iterations for preamble arrival)

+(energy to sendACK) +(energy to receive packet)

Er PsRs + PRxRl( 1113857

1 minus 1 minus Pd(t)( 1113857 Rl + Rs( 1113857+ RaPTx + RdPRx

(17)

And the one-hop expected latency is given as

Lat (preamble duration + ACK listen)

lowast(expected number of iterations)

+ duration to send packet

Lat Sp + Sal1113872 1113873 Rl + Rs( 1113857

Rl minus Sp

(18)

Advantages

(i) It increases energy efficiency(ii) It reduces data latency(iii) It increases throughput

2115 A-MAC (Advanced MAC) A-MAC is a TDMA-based protocol whose design is aimed at enhancing thelifetime of the network for low rate and reliability of datatransmission [39] To assign time slots it does not depend ona central controller like most protocols but it gathers in-formation from its neighbors so as to allocate the slotsthrough a distributed algorithm Energy is controlled byscheduling power when there is not a single event Its designcomprises many frames with each having many time slotswhich at the beginning have got a beacon transmitted tosynchronize and exchange information with neighborsNodes to participate in the subsequent session are selectedby the controlled node so they do not go into the sleep modewith others [13] -is protocol operates under four time slotdivisions namely initial wait discover and active states Anode in the initial state begins by listening to the medium forthe beacon packet from other nodes for network synchro-nization which when received makes it change the timerthrough subtraction of beacon transmission time from re-ception time To avoid synchronization problems due todrift and allow for continuity the node has to get thestrongest beacon signal It then enters the wait state afterchoosing a number of waiting frames so as to reduce theprobability that nodes get into the discover state simulta-neously In case the synchronization beacon is lost while stillin the wait state it will return to the initial state for anotherone On expiry of the waiting counter it enters the discoverstate and collects information from neighbors by listening totheir beacon messages On successful selection of a time slotit enters the active state where it endlessly transmits beaconsat the initiation of the slot A node goes to sleep when abeacon has been transmitted and there are no more data fortransmission or in case the neighborrsquos beacon received in-dicates no data packets are coming in [39]

Advantages

(i) It can enable multicasting(ii) It enhances the lifetime of the network(iii) It offers reliable data transmission

2116 L-MAC (Lightweight MAC) Based on TDMA thisprotocol is aimed at minimizing the number of transceiverswitches adapting the nodal sleep interval to data trafficamount and making implementation less complex [40] Apacket is transmitted comprising a control and a data partControl packets have a fixed size and the time slot con-trollerrsquos ID -ey show the distance from the transmittingnode for simplified routing to the network gateway as well asaddressing target node and reporting data unit length andinternode synchronization Adjacent nodes are more in-terested in getting control packets from nearby nodes -enodersquos transceivers are switched off until the subsequenttime slot unless the node is also addressed or the message isomnicast If so it listens to the data unit which may not fillthe remaining time slot in totality-e transceivers on eitherside are turned off when the transfer of packets is successfulEnergy wastage through idle listening in uncontrolled timeslots is avoided via a short time-out interval Network nodes

Journal of Computer Networks and Communications 13

are able to communicate without colliding-is saves energyand increases the lifetime of the network A node in L-MACcan only transmit one message per frame [13] -e networkis set up in such a way that all nodes are not synchronized onbeing switched on-e gateway takes control of a timeslot soas to have them synchronized and its immediate neighborgets the control messages thereby synchronizing their timerswith the gatewayrsquos Neighboring nodes learn of multiplegatewaysrsquo time slots in their vicinity just after one frame-ejust synchronized nodes randomly choose time slots forcontrolling as long as they are not occupied -e nodes thuspreserve a table for the neighboring nodes which makes atimeslot reusable after a minimum of 3 hops and preventsmessage collision All nodes keep their slots before thebattery is down or learn of an impending collision All nodescan track their hop distances to a selected gateway node andthe information is communicated in control messages [40]

Advantages

(i) It minimizes the number of transceiver switches(ii) It adapts the nodal sleep interval to data traffic

amount(ii) It makes implementation less complex(iii) It offers improved network lifetime(iv) Energy wastage through idle listening in un-

controlled time slots is avoided via a short time-out interval

Disadvantages

(i) L-MAC is not very good for mobile SNs since slotsare calculated only once

(ii) It suffers idle listening because nodes keep listeningto slotsrsquo control parts so as to get data and enableother nodes to join the network all the time [13]

2117 C-MAC -is was designed for WSNs with 3 seg-ments namely (i) aggressive RTS having double channelcheck to assess the channel (ii) anycast to initialize the flowand (iii) convergent packet forwarding to stabilize the flowIt promises an improvement in energy efficiency lowerlatency and enhanced throughput while avoiding syn-chronization overhead though it operates only in low duty-cycling apps especially where there is less traffic But whenthere is more traffic the protocol employs anycast-basedpacket forwarding for waking the nodes up or finding aforwarder quickly and converges from route-suboptimal toroute-optimal unicast After this energy could be preservedby the nodes using a synchronized wake-up plan [13]

Advantages

(i) It improves energy efficiency(ii) It lowers latency(iii) It enhances throughput avoiding synchronization

overhead

Disadvantages

(i) It operates only in low duty-cycling applicationsespecially where there is less traffic

22 MAC Protocols for WMSNs

221 Diff-MAC -is is a CSMACA-based QoS-awarehybrid-priority-based MAC protocol for WMSNs It targetsenhancing the channel utilization while making use of ef-ficient service differentiation to do coordination for mediumaccess of each traffic class while giving fair quick datatransfer It offers QoS using the following features [14]

(i) It balances energy consumption and delay throughsensor node duty-cycling adaptation depending onthe dominant traffic class

(ii) It has a feature for fragmentation and messagepassing that breaks lengthy video frames intosmaller packets that are transmitted as a burst thusreducing retransmission costs if there are any MACfailures

(iii) -ere is fair data delivery amongst the nodes andtraffic classes respectively resulting from intranodeand intraqueue prioritization reducing unbearableperformance

(iv) -e number of collisions can be reduced and packetlatencies are kept within limits by adjusting the sizeof the congestion window in this protocol as per thetraffic needsAdvantages

(i) Diff-MAC is very suitable for WMSNs since itadapts quickly to changes in networkconditions

(ii) It is a fair protocol(iii) It is scalable

Disadvantages

(i) It is quite difficult to monitor the statistics ofthe network

(ii) Adapting dynamically to the network condi-tions is not an easy task

(iii) It exhibits some degree of latency in packetdelivery

-e implementation of Diff-MAC on Imote2 indicatesthat resource requirements can be met on the sensorhardware Results of extensive simulation runs showed thatDiff-MAC outperforms its competitors

222 Cluster-Based On-Demand Multichannel MAC Pro-tocol for WMSN (COM-MAC) -is is an energy-efficientprotocol in WMSNs with high throughput and reliable datatransmission It achieves energy efficiency by schedulingmultichannel media access to reduce channel contentionand remove collision idle listening and overhearing [41]

Design it has a three-phased operation with the (i)request phase (ii) scheduling phase and (iii) datatransmission phaseRequest phase here all nodes have two protocols eachfor sending requests to the cluster head (CH) ie thecontention-based and contention-free TDMAFDMA

14 Journal of Computer Networks and Communications

protocols -e contention-based protocol allocates achannel per node for sending the request message andfree channels are useable as control channels Forcontention-free protocols a control slot is allocatedwhen a network is deployed Here a slot per channel isallocated to a different node enabling requests to be sentwithout interfering with transmissions from othernodes -e contention-based protocol operates undertwo steps (i) Control channel assignment phase achannel is assigned per node for transmitting a request(REQ) message A channel is assigned to multiplenodes and nodes are evenly distributed to availablechannels for congestion control (ii) Request trans-mission phase nodes with data send a REQmessage viathe assigned control channel to notify the CH whichresponds with an ACK message -e contention-freeTDMAFDMA protocol too has two phases includingthe control slot assignment and request transmissionphases For unreliable channel conditions and lowtraffic load the contention-based protocol is fit for usebecause of less collisions and congestion whereas thecontention-free protocol is preferable in heavy trafficload and reliable channel applicationsScheduling phase from the above phase a schedule isgenerated by the CH for coordination and this enablesnodes to transmit and broadcast their data via controlchannels-e schedule for broadcasting only comprisesdata for nodes assigned to a channel -e schedule isrebroadcast for transmission reliability enhancementbut for energy efficiency the nodersquos transceiver isswitched off when all the schedule is completely de-livered till it is the time for transmitting again-e timeslot and data communication radio channel are allincluded in the schedule generated for each nodeData transmission phase on getting the schedule thesensor node sends its data as per the time slot and thechannel assigned -e time slot is divided into datatransmission andACK sections ACK supports link layererror control Using implicit selective repeat ARQ theCH sends an ACK message for all received packets Ifthere is no ACK the packet is assumed lost and hence aretransmission during the subsequent interval is done-is enhances reliability of transmissions though excessdelay is possible here which is also intolerable for real-time critical apps like video or audio -is gives rise to ahybrid MAC protocol to exploit the unused spectrumduring data transmission sessions [41]Advantages

(i) It achieves energy efficiency by scheduling mul-tichannel media access to reduce channel con-tention and remove collision idle listening andoverhearing

(ii) It maximizes network throughput by dynamicallyallocating time slots and channels for nodes byexecution of a traffic-adaptive QoS-aware sched-uling algorithm depending on needs for QoS andtraffic data information

(iii) It increases transmission reliability by in-corporating spectrum-aware ARQ to avoid spec-trum wastage

223 EQoSA -is protocol is developed for video andimage transfer across WMSNs in a hybrid mode withprovision of QoS Concerning the active sensor nodes andtheir traffic loads it employs dynamic session sizes andchanges bitmap-assisted (BMA) MACrsquos fixed session sizeNodes declare any data available for transmission duringcontention time -is forces the cluster head to make aschedule of how slots have been assigned It communicatesthis schedule to the nodes giving each node a given numberof slots per session and so enabling the busty traffic to beaccommodated [14] -e protocol does not perform wellwith DiffServe mechanisms but does well with schemessupporting busty traffic loads and strong cluster heads arenecessary in WMSNs that can assign and broadcast slots

224 QoS-Supported Energy-Efficient MAC (QE-MAC)Based on the IEEE 80211e standard this protocol ensuresfairness and has low latency and jitter and better efficiencyon energy usage than other QoS-aware MAC protocols forWMSNs It operates under a two-phased routine

-e first phase introduces the innovating prioritymechanism in IEEE 80211e QoS and several data types areassigned priorities Its coordination function is a combi-nation of distributed contention-based channel allocationand centralized polling-based channel access mechanisms

-e second phase preserves the nodesrsquo energy usingdynamic duty cycling -ere is an exchange of RTSCTSpackets to supplement CSMACA by muting nodes near thesenderreceiver during packet duration [42]

Advantages

(i) It ensures fairness(ii) It has low latency and jitter and better efficiency on

energy usage than other QoS-awareMAC protocolsfor WMSNs

225 Black-Burst (BB) Contention A contention-basedMAC scheme for enhancing instantaneous QoS access tocarrier-sense WSNs is proposed in [42] In this schemenodes contend for the channel until it is free Real-timepackets are given higher priority over non-real-time packetsand if a node is transmitting real-time packets it is prior-itized over other nodes [15] Voice and video are some of thereal-time apps in consideration here which periodicallyaccess the common radio channel when transmitting -eseneed constrained end-to-end delay of packets at the MAClayer -is protocol majorly does the following [42]

(i) Prioritization of real-time traffic(ii) Implementation of the round-robin mechanism in

real-time nodes(iii) Bounded access delays among real-time packets

Journal of Computer Networks and Communications 15

(iv) Implementation of real-time communication withvarying needs of bandwidth good for multimediatraffic

Amechanism for dynamically assigning packet prioritiesbased on their deadlines and the traversed hops is proposedin [43] -e ReAllocative Priority (ReAP) mechanism ismeant to provide QoS at the MAC layer to video traffic -isscheme introduces adaptive TXOP by modifying the TXOPdynamically based on the queue length [15] Other protocolsin this category include dynamic duty cycle and adaptivecontention window-based MAC protocol frame sharing(FRASH) MAC protocol and real-time independentchannels (RICH) MAC protocol

Advantages

(i) It can implement real-time communication withvarying needs of bandwidth

23 Challenges and Opportunities(a) One of the challenges in satellite transmission is

cross-talk leading to interference in different fre-quency bands It is necessary to increase the rise andfall times as well as protect sensitive nodes as wecircumvent floating ones if we are to regulate cross-talk as in satellite communication therefore Wefurther need to investigate the possibility of usingdifferential signaling -is constitutes an opportu-nity for further research

(b) Another challenge is that the hidden and exposedterminal problems are still noticeable despite the useof RTSCTS solutions in MAC since collisions arestill possible from transmissions of separate nodesAt the MAC layer ACKs might block efficientchannel usage by prohibiting exposed terminalsfrom reusing the channel and may be eliminatedleaving no proof of sender data packet receptionUsing a directional antenna is one of the proposedtechniques of high-speed WSNs like IEEE 80211-is constitutes an opportunity for further study intothese issues for better solutions though

(c) Another challenge is that since nodes in contention-based MAC protocols are mobile there are manycollisions by the packets and schedule in-consistencies in schedule-based protocols in the two-hop neighborhood information when nodes areentering or leaving [44] Deng et al [45] propose amobility-based clustering (MBC) protocol for mo-bile nodes to enhance the packet delivery rate -isprotocol in comparison with CBR has 25 chancesof decreasing collision-induced packet losses and50 chances in comparison with low-energyadaptive clustering hierarchy (LEACH)

(d) -e process of analyzing and evaluating routingperformance for WMSNs is quite complex becauseof the bursty multimedia traffic Because of the wayencoding scheme frames are structured as well as thechanges in between the scenes that occur naturally

the compressed video frames display some substantialburstiness on several time scales -ere will also befrequent route recalculation that is as well energy in-tensive for nodes frequently capturing and transmittingenergy -e solution here could as well possibly be theuse of the MBC protocol Route recalculation tooshould be dynamic to cater for the mobile nodes whichmight be in and out of the range at different instants oftime Recently the use of smart dumpsters for effectivewaste management in smart cities has also been pro-posed as a possible solution [46]

(e) For WSNs their routing protocols do not look atmultimedia applications which require enormousbandwidth and processing energy and should betransmitted in real time with utmost fidelity Extrachallenges in routing are imposed in WMSNs indeployment of heterogeneous sensors since therecould be audio video and still pictures in addition toscalar data during transmission Heterogeneous datahandled also have different QoS requirements andwith many different business needs different ser-vices with varying requirements give rise to a bigchallenge for the routing design -ere is a need todesign protocols that can support WMSNs such asEQoSA [14] for this purpose

Table 2 summarizes the reviewed MAC protocols forWSNs and the ones designed for suitability in WMSNs

3 QoS Strategies at the Application Layer

We discuss many protocols used to solve different needs ofcommunication between machines

31 ConstrainedApplication Protocol (CoAP) It is a requestresponse protocol designed by the Internet Engineering TaskForce (IETF) for constrained devices such as those havinginsufficient RAM or CPU and WPANs with low-powerconstraints resulting in poor transfer of packets andenormous overhead -e protocol was designed for ma-chine-to-machine (M2M) apps and system automation toreduce bandwidth requirements increase packet transferdecrease the overhead and make work simpler leading tolightweight implementation HTTP commands like GETPOST PUT and DELETE are used in a client-server ar-chitecture Using the publishsubscribe architecture CoAPcan support multiple users and hence yield better perfor-mance via multicasting enabling asynchronous messageexchange -is protocol can support both multicasting andunicasting via UDP and hence is a reliable and simpleprotocol -is is done via the messaging sublayer (for re-liability based on UDP) and the requestresponse sublayer(for communication and interaction) [16]

It uses the following messages

(i) Confirmable message this guarantees reliabilityusing ACK messages sent by the receiver node-ese could be synchronous or asynchronous ifthere is a need for more computational time

16 Journal of Computer Networks and Communications

(ii) Nonconfirmable message here no ACK messagesare sent back

(iii) Acknowledgment (ACK) message this confirmsreceipt of a confirmable message

(iv) Reset message this confirms receipt of an un-processed message in case any critical parts nec-essary for interpreting the message are missing

(v) Piggybacked response the receiver directly re-sponds while getting the ACK message

(vi) Separate response the reply by the receiver is dis-crete from the ACK message

CoAP has no in-built security features It is secured bythe datagram transport layer security (DTLS) that runs ontop of UDP though it was not designed for the IoT -ismakes its suitability debatable since DTLS offers no supportfor multicasting It also needs more packets for handshakingthereby increasing the traffic volume and computationalresources and so shortening the lifespan of mobile devicessince they use batteries [17]

Advantages

(i) It can support multiple users and hence yieldbetter performance via multicasting

Table 2 Comparison of MAC layer protocols used in QoS provisioning

Protocol Features QoS parameter Comments Access Priorityassignment Energy eff Designed for

S-MAC[22] Collision avoidance Energy latency Requires periodic

synchronization CSMA Hybrid Medium WSNs

B-MAC[23] Collision avoidance Energy delay

throughput Flexible interface CSMA Dynamic High WSNs

T-MAC[25] Collision avoidance Energy Complex nonscalable CSMA Dynamic High WSNs

C-MAC[13]

Dynamic channelallocation Energy Multichannel MAC CSMA Dynamic Medium WSNs

CU-MAC[24]

Multichannelmechanism

-roughputlatency High packet delivery ratio TDMA Dynamic mdash WSNs

A-MAC[39]

Per flow serviceguarantees Delay m-ary tree model

dynamic prioritiesTDMACSMA Dynamic Medium WSNs

ER-MAC[30]

Emergencyresponse

Energy delaydelivery ratio

Adapts to traffic andtopology changes

TDMACSMA Hybrid Higher than Z-

MAC WSNs

RI-MAC[31] Collision detection Delay packet

delivery ratioHigh packet delivery ratio

and throughput CSMA Dynamic Medium WSNs

SRI-MAC[33] Collision avoidance Energy delay Good data transfer and

energy consumption TDMA Dynamic High WSNs

H-MAC[34] Slotted sleep time -roughput

latencyData transmission only in

sleep timeCSMAAloha Dynamic Medium WSNs

Z-MAC[36] Collision avoidance Energy delay

throughputVery good channel usage

and error syncCSMATDMA Dynamic Medium WSNs

Q-MAC[37]

Intrainternodescheduling Energy High QoS improved

energy efficiency CSMA Hybrid High WSNs

X-MAC[32]

Overhearingavoidance

Energy delaythroughput

Reduces excessivepreamble by strobing CSMA Adaptive

static Medium WSNs

L-MAC[40] Collision avoidance Network

lifetime energyReduces the number oftransceiver switches TDMA Static Medium WSNs

WiseMAC[38]

No downlinkchannel collision Power delay Significantly low-power

consumptionAlohaCSMA Static Higher than

ZigbeeInfrastructure

WSNsAR-MAC[27] Collision avoidance Energy delay Uses data packets and

control packets TDMA Static Higher thanIEEE 802154 WBANs

BB-MAC[42]

Real-time trafficprioritization Delay Employs round robin in

real-time nodesCSMATDMA Dynamic High WMSNs

COM-MAC [41] Collision avoidance Packet delay

throughputBalances reliability and

retransmissionTDMAFDMA Dynamic High WMSNs

EQoSA[14]

Busty traffic loadaccommodation Energy Provides QoS support for

video and imagesTDMACDMA mdash High WMSNs

QE-MAC[42] Collision avoidance Energy jitter

latencyRuns central polling-like

channel access CSMA Dynamic High WMSNs

R-MAC[26] Collision avoidance Energy delay Reservation based mdash Static High Underwater

SNs

Journal of Computer Networks and Communications 17

(ii) It enables asynchronous message exchange(iii) It is a reliable and simple protocol

32 Message Queue Telemetry Transport (MQTT)Developed by IBM and later improved by OASIS with anintention of decreasing bandwidth MQTT is a simple open-source publishsubscribe protocol analogous to the client-server version It is appropriate for lightweight IoT appli-cations and M2M transmissions in constrained situationslike limited power computation capability and memory orbandwidth -e network bandwidth and the computationalresource usage are decreasing since clients are not updatedmaking publishsubscribe protocols outperform requestresponse protocols via IoT requirements [16] -is protocolcan run over TCPIP and so has the following advantages

(i) Delivering of packets with reliable guarantee(ii) Enabling multicasting(iii) Remote device session establishment(iv) Reduction of transport overhead and protocol ex-

change to minimize network traffic(v) Provision of the notification mechanism when

anomalies occur [47]

MQTTrsquos design targets economical use of bandwidthand battery so it has a major application in FacebookMessenger MQTT ensures QoS and reliability of messagedelivery by providing the following three QoS levels

(1) Fire and forget a message is sent once depending onthe networkrsquos best effort and there is no need forACK (lowest QoS level)

(2) Delivered at least once a message is sent at least onceand an ACK message is needed to avoid duplicates

(3) Delivered exactly once by a four-way handshake itensures a message is sent exactly one time (highestQoS level)-e architecture it has three key components (i)publishersmdashlightweight nodes sending data to thebroker before returning to sleep (ii) broker whichorganizes sensory data in topics and sends them tointerested subscribers and (iii) subscribersmdashIoTapps interested in data sent by sensors through thebroker as shown in Figure 3

MQTT is a simple lightweight protocol that is betterthan CoAP via low and high packet loss probability interms of throughput and exhibits lower latency in lowsampling rates [48] It supports a number of imple-mentations for embedded devices and mobile applicationsand offers web-socket support too Nevertheless it hassome shortcomings

(i) It is only employed in simple data types(ii) It exhibits high latency in high sampling rates(iii) It is unsuitable for real-time apps in such cases(iv) It does not provide automatic discovery andor

sufficient security at the protocol level [49]

33 Secure Message Queue Telemetry Transport (SMQTT)-is is an extension of MQTT that is based on lightweightattribute-based encryption using elliptic curve cryptogra-phy -e protocol has three entities as shown in Figure 4(i) publisher that publishes information under a certaintopic (ii) subscriber that obtains the same topic in-formation via a broker and (iii) public key generator(PKG) or broker that acts as the trusted third party -eprotocol has four phases namely (a) set-up phase (b)encrypt phase (c) publish phase and (d) decrypt phase

(a) Set-up phase publisher and subscriber provide aunique ID with attributes for registration with thePKG -is generates a secret master key set andpublic parameters that it publishes with set U A1A2 An For ciphertext policy-attribute-basedencryption (CP-ABE) the PKG gives a key set todevices with attributes [50]

(b) Encrypt phase based on the access tree the publishercreates an access policy An access tree is sent by thepublisher to the PKG for generating the key-gen-eration policy in the key policy- (KP-) based ABEalgorithm -e publishers know beforehand thetopics and subscribers to access as well as the rulesfor access Subscribers are issued with private keysfor all communication sessions Publishers in theCP-ABE algorithm come up with access trees andrules where the payload is encrypted by the pub-lisher who also issues more info needed for de-cryption and the rules With the help of public keycryptography data are encrypted giving rise to ci-phertext on the basis of KPCP-ABE and the receivergets SUBACK from the broker [51]

(c) Publish phase ciphertext is embedded in theSPublish command as payload and topic name isdefined in the header by the publisher -e brokerreceives the SPublish packet and replies withPUBACK prompting the publisher to respond withthe PUBREL packet -e broker then broadcasts totopic subscribers and deletes the data forwarding thePUBCOMP packet to the sender node

(d) Decrypt phase the encrypted information isdecrypted by the subscriber with the help of a secretkey Considering KP-ABE on verification that itsatisfies the access policy the subscriber invokes thePKG for a corresponding key which it sends afterverifying the request With CP-ABE subscribers usethe private key for decryption after authenticationthereby enabling offline interaction [51]Advantages

(i) -e information being encrypted makes it moresecure

34 MQTT for Sensor Networks (MQTT-SN) -is is a ver-sion of MQTT designed to operate under environments oflow bandwidth high link failures and short message lengthIt is a publishsubscribe (pubsub) protocol designed for

18 Journal of Computer Networks and Communications

UDP-based WSNs to visualize communication for low-power devices [5152] It comprises a publisher device actingas the MQTT-SN client to forward communications to agateway that also changes the MQTT-SN message to theMQTTmessage before delivering it to the broker that sendsit to MQTT-SN subscribers [51]

It has a gateway for changing protocols from MQTT toMQTT-SN (Figure 5) [52] -is protocol supports nodes in

the sleeping mode using an offline keep-alive procedure thatcan be used by battery-operated devices to get into thesleeping mode To support sleeping clients MQTT-SN has anew offline keep-alive procedure with which battery-oper-ated devices can go to the sleeping state in which theirmessages are buffered at the server until they are awake InMQTT-SN clients with no preconfigured gateway addresscan figure out the real address of their operational gateway

Publisher

Subscriber

Subscriber

Broker

Publisher

4G

Figure 3 Architecture of the MQTT protocol

Register by mentioning URI Register by mentioning URI

Access tree

Set-up

Publisher ReceiverPKG(broker)

Key management Key management

Encrypt data using public parametersand random key This is done using

CP-ABE Publish it as

Subscribe (topics)SUBACK

S-Publish (TopicEncrypt (data))S-Publish (TopicEncrypt (data))

Encryptand publish

PUBACK

PUBACK

PUBRELPUBRELPUBCOMP

PUBCOMP

Figure 4 SMQTT protocol

Journal of Computer Networks and Communications 19

within the network through a discovery procedure Simul-taneously we may have several gateways in a single wirelessnetwork that can be cooperative via load sharing

MQTT-SN architecture it is composed of MQTT-SN clients gateways and forwarders (Figure 5)rough the MQTT-SN gateway (GW) the MQTT-SN protocol is used by the MQTT-SN clients toconnect to the MQTT server Integration betweenthe MQTT-SN GW and the MQTT-SN server mightbe possible or not Translating between the MQTT-SN and the MQTT for a stand-alone GW is doneusing the MQTT-SN protocol If a GW is detachedfrom the network it can be accessed by the MQTT-SN clients through a forwarder is just encapsu-lates the received MQTT-SN frames that are on thewireless side before forwarding them to the GW un-altered It then decapsulates the received frames tooresending them to the client as sent by the GW [52]ere are 2 types of gateways (a) transparent gatewayand (b) aggregating gateway as shown in Figure 6

(i) Transparent gateway this launches and sustains adirect MQTT connection between the MQTT-SNclient and the MQTT server which is entirely re-served for end-to-end client-server messaging enumber of servers and that of MQTT-SN clientsconnected to the GW are the same is transparent

GW translates syntax between these protocols and iseasier to implement compared to the aggregatingGW But the MQTT server must support dierentsessions for all connected clients Some MQTTservers are congured to enable few parallel sessions

(ii) Aggregating gateway this maintains a single MQTTconnection to the server on behalf of all clients Allexchanged messages from the MQTT-SN client to theaggregating GW are directed to the GW which sortsthe information to forward to the server e aggre-gating GW decreases the number of MQTT sessionssimultaneously supported by the server in WSNs withmany battery-operated sensors and actuators (SAs)having limited storage and processing capabilities It ishowever more complex to implement compared to atransparent GWe general format of the MQTT-SNprotocol message has two or four message headeroctets and n octets of variable header that is optionaldepending on the message in question MQTT-SN has256 message types (MsgType) with those from0X1Eminus 0XFD 0X19 and 0XFF reserved for futureusage [51]

Advantages

(i) e protocol can operate under environmentsof low bandwidth high link failures and shortmessage length

MQTT-SN gateway

MQTT-SN

MQTT

MQTT-SN

MQTT-SN

MQTT-SN

MQTT-SN client

MQTT-SN clients

MQTT broker

MQTT-SN client

MQTT-SN client

MQTT-SN gateway

MQTT-SN forwarder

Figure 5 MQTT-SN architecture

20 Journal of Computer Networks and Communications

(ii) It is a pubsub protocol designed for UDP-basedWSNs to visualize communication for low-power devices

(iii) It can be optimal for low-cost battery-operatednodes having low processing power and lessstorage

(iv) It supports nodes in the sleeping mode using theobrvbarine keep-alive procedure used by battery-operated devices

35e Extensible Messaging and Presence Protocol (XMPP)e standardization of this protocol by the IETF made itpopular for usage over the Internet especially in instantmessaging and chatting platforms It is however susceptibleto spamming attacks and its failure to have worldwidesupport has made Google too withdraw its support of thestandard [17] e protocol is mainly useable in the client-server architecture but can also be applicable in client-clientXMPP communication [53] Among the application layerprotocols XMPP is the only one that supports both asyn-chronous (publishsubscribe) and synchronous (requestresponse) services while running over TCP but can still runover other protocols

Given the simplicity of its implementation and ease ofusability in XMPP instant messaging (IM) is suitable forXMPP and is implemented using the message stanzaltmessagegt ese stanzas are usable in one-to-one ormultiuser sessions or chatting or in error reporting andsending headlines Attribute type of the stanza denes themessage type e attribute type could be a chat group chatheadline normal or error An instant messaging network inXMPP can support many XMPP clients and servers with agateway congured to enable other IM network users toconnect to other XMPP users e XMPP server mainlycontrols sessions and XML streams to clients and servers inaddition to routing XML stanzas in the streams [53] eTCP ports 5269 and 5222 have been respectively registeredfor the XMPP server-to-server connections and XMPP clientconnections by the Internet Assigned Numbers Authority(IANA) XMPP enables specication of XMPP ExtensionProtocols (XEP) to enhance its usefulness [17]

XMPP security clients connect to the server usingtransport layer security (TLS) and simple authentication andsecurity layer (SASL) protocols which are at the same timeused by the servers for interdomain communications Fordata condentiality and integrity TLS will encrypt thestreams of XML for all sessions To conrm the identity of anyclient that needs to access the server the SASL protocol doesthe authentication of the XML stream For any communi-cating session with a server to start the client needs to resolvethe serverrsquos DNS hostname For protection of the server of theclient from third-party attacks the server maintains the au-thentication details secret ie IP address and access methodand only original connections of the server are required

In server-to-server communications the SASL protocolis employed for authentication and condentiality Server-to-server communications may as well be deactivated from aserver by an administrator depending on the organizationalpolicy To guard against domain-spoong attacks serverdialback can be a good solution since servers support itsecuring more the XML stanzas Nevertheless it cannot beused to authenticate secure or encrypt in-server streamsand the server identication resulting therein is not strongenough to be relied on It further cannot secure against DNSpoisoning attacks and IP session hijacking for remote do-mains [53] High-security domains can use TLS and SASL

Advantages

(i) It is popular for usage over the Internet especially ininstant messaging and chatting platforms

Disadvantages

(i) It is susceptible to spamming attacks(ii) It has failed to have worldwide support making

Google withdraw support of the standard(iii) It is not secure against DNS poisoning attacks and

IP session hijacking for remote domains

36 Advanced Message Queuing Protocol (AMQP) isprotocol can be utilized by a number of platforms formessage exchange since it is open source and written inmany languages It is a publishsubscribe model depending

MQTT-SNclients Broker

MQTT

(a)

MQTT-SNclients Broker

(b)

Figure 6 (a) Transparent gateway and (b) aggregating gateway

Journal of Computer Networks and Communications 21

on a reliable and efficient messaging queue with capacity forinstant messaging or e-mail for different applications es-pecially in IoT environments It differs from others in thesense that message types can be specified and have theirsource traceable and how to do trade-off between perfor-mance security and reliability For reliability of messagetransmission AMQPmakes use of some delivery guaranteeslike at most once at least once and exactly once It can alsohowever utilize the TCP transport layer for guaranteeingreliability -e AMQPrsquos publishsubscribe method has twocomponents namely (i) exchange queue meant for routingof messages to an appropriate order in the queue and (ii)message queue that acts as storage pending their delivery toreceivers [16] In this protocol model the broker can makerouting decisions contrary to othermessaging systems whereapps using the queue have in them embedded the decision-making logic for recipients or senders -e developer mayneed to check each of the affected apps so as to change thelogic for routing and delivery of the messages [54]

Advantages

(i) It is convenient(ii) It has capacity for interoperability between dif-

ferent vendors(iii) It can use open standards to enable connection

between business partners(iv) It can enable innovations based on AMQP foun-

dations [55](v) It can enable supporting of mission-critical ap-

plications in e-commerce(vi) It can guarantee reliable delivery of messages

Disadvantages

(i) It is not very appropriate for real-timeapplications

(ii) It may not offer automatic discovery(iii) Interoperability is not guaranteed(iv) It lacks open-source libraries for constrained

gadgets [49]

37 Data Distribution Service (DDS) A product of theObject Management Group (OMG) DDS is a pubsubprotocol meant for M2M communications with two sub-layers -ese are the data-centric pubsub sublayer re-sponsible for connecting anonymous data publishers tosubscribers and the data-local reconstruction sublayer that isoptional and integrates DDS into the application layer [56]-e DDS architecture connects system components asshown in Figure 7 Publishers and subscribers are discon-nected with respect to (i) time (ii) space (iii) flow andbehavior (iv) platforms and (v) programming languages[57] A DDS data publisher generates topics and publisher-subscriber coupling is represented as topic name data typeschema and publishers and subscribersrsquo QoS attributesDelivering of data is done by the publisher layer -e datawriter and publishers together make a decision about anynecessary alterations for subsequent forwarding to thesubscribers who need to send data to IoT applications Data

readers are responsible for delivering topics to the sub-scribers after reading through them [56]

Some of the major components of the DDS architectureinclude the following [57]

(i) Topic this is a logical path specifying publicationand subscriptionrsquos data type between DataWritersand DataReaders For a successful session to begintopic names types and DataWriters and Data-Readersrsquo QoS should correspond

(ii) Domain this is a logical setting for transmissionthat can be utilized for separating and optimizingnetwork transmission in a setup of similar appli-cations If DDS apps share a domain ID then theycan exchange data among themselves

(iii) DataWriter and DataReader these are terminalobjects for writing and reading messages re-spectively to and from a universal data space

(iv) Participant this is an object that stands for apublisher or subscriber of the DDS app in a do-main acting like a container for other objects

Advantages

(i) It is highly reliable and offers QoS support(ii) It is highly scalable(iii) It is appropriate for real-time apps(iv) It is fault tolerant(v) It has high level of interoperability(vi) It does not need brokers that may act as

bottlenecks(vii) It offers automatic discovery(viii) It has a user-defined data structure for topic(ix) It supports publisher and subscriber decoupling

Disadvantages

(i) It is originally developed for only stand-aloneLANs

(ii) A lot of memory is needed(iii) It has no open-source libraries to support con-

strained devices(iv) It is hard to design and configure compared to

other protocols [49]

38 RESTful Services REpresentational State Transfer (REST)-is is an architectural style based on a set of principles thatdefine and address networked resources It offers web servicesthat enable transmission from the HTTP device to the devicewithin the IoT architecture and enables clients to reach theserver [16]-is protocol utilizes HTTP commands to backupthe requestresponse messaging model HTTP is a popularand secure protocol in the WWW using TLSSSL [16] ARESTful application has the following characteristics

(i) State and functionality are separated into distrib-uted resources

(ii) All resources are distinctively named by the use ofunique HTTP commands (GET POST PUT orDELETE over the Internet)

22 Journal of Computer Networks and Communications

(iii) It is a clientserver stateless layered protocol thatsupports caching REST uses the same methods asHTTP for all requests or responses with POST andGET for creating and retrieving resources PUT forupdating and changing the state of resources andDELETE for removing resourcesAdvantages of REST over arbitrary web services(SOAP) in the IoT context

(i) It reduces overhead(ii) It reduces parsing complexity(iii) It is stateless(iii) It integrates more tightly with HTTP [58](iv) It is easy to implement and learnis together

with its natural t over HTTP makes it apreferred method for Web 20 applications toshowcase their data and secure M2M sessionsin the IoT

(v) Applications that support RESTful web ser-vices exhibit better performance than the re-source-constrained WSN nodes [59]

(vi) REST is important for the IoT since it supportscommercial M2M cloud platforms

(vii) Its implementation in smart phones and tabletapps is also easy since it needs only an HTTPlibrary that is accessible to all operating systemplatforms [56]

Disadvantages

(i) Although it enjoys wide usage in commercialM2M platforms it is not likely to dominatebecause it is hard to implement

(ii) Its HTTP usage makes it incompatible withconstrained M2M devices leaving its usage toend applications

39 WebSockets It is a web-based protocol operating on asingle TCP channel and providing full-duplex transmissions

It permits secure duplex communication between a clientexecuting unreliable code in a restricted environment and adistant node communicating from that code while using theorigin-based security model employed by web browsers overTCP [60] e protocol is not a requestresponse model andis not a publishsubscribe model So unlike other protocolsa web-socket session is established by the client initiating ahandshake to the server to launch the session is onceestablished a full-duplex client-to-server connection beginsasynchronously and runs until terminated by either party[56] It gives a model for browser-based apps utilizing bi-directional communication with no need of many HTTPconnections to servers

WebSocket comprises two parts ie (i) the handshakehaving a clientrsquos message and the serverrsquos handshake responseand (ii) the data transfer HTTP headers should be repeated byapplications in all client requests and server responses arisingfrom the endless polling which might increase the commu-nication overhead subject to the application [61] WebSocketcan be used to come up with scalable real-time web appli-cations since it can enable full-duplex transmission via onesocket across the Web Its security can be ensured with thehelp of TLSSSL is protocol is suitable for resource-con-strained devices and IoTapplications will not have a good userexperience with it since it operates a clientserver architectureNevertheless it can outcompete all TCP-based protocolsbecause of the following advantages

(i) It is t for real-time transmission(ii) It has improved security(iii) It minimizes transmission overhead(iv) It can provide ecient messaging systems using

WebSocket ApplicationMessaging Protocol (WAMP)[17]

310 Streaming (Simple) Text-Oriented Messaging Protocol(STOMP) is is a TCP-based protocol designed for textmessage-oriented middleware (MOM) with the use of

Topic

Datareader

Datareader

Datareader

Datawriter

Datawriter

Datawriter

Subscriber PublisherPublisher

Data bus (DDS domain)

Subscriber

Topic Topic

Figure 7 DDS architecture

Journal of Computer Networks and Communications 23

HTTP-like commands for interoperability among platformslanguages and brokers Data communication takes placebetween the client and the broker in line with commands(like CONNECT DISCONNECT ACK NACK SUB-SCRIBE UNSUBSCRIBE SEND BEGIN COMMIT orABORT) then headers in the form ltkeygt ltvaluegt (one perline) then a blank line and then body content terminatingin a null character Server-client communication takes placeover a MESSAGE RECEIPT or ERROR frame with headerand body content format being analogous MOM productssupporting STOMP include Apache ActiveMQ or FuseMessage Broker HornetQ Open Message Queue(OpenMQ) RabbitMQ and syslog-ng through its STOMPdestination plugin [62] In addition to these other appli-cation layer protocols include simple media control protocol(SMCP) lightweight local automation protocol (LLAP)simple sensor interface (SSI) lightweight M2M (LWM2M)XMPP-IOT and simple object access protocol (SOAP)

Advantages

(i) It is lightweight for constrained networks(ii) It has flexibility to choose quality of services with

the given functionality(iii) It is standardized by OASIS Technical Committee(iv) It is easy and quick to implement

Challenges and opportunities

Here security threats are imminent because of the open-endedness of the application layer [63] -ese could includethe following

(i) -ere is poor or lack of a security design for anapplicationrsquos function Coming up with a gooddesign to address this could be a good ideaopportunity

(ii) Some of the programs might have backdoorsevading otherwise secure controls to allow un-authorized access to network resources We requiremechanisms to patch up such loopholes

(iii) Applications that lack or have weak authenticationmechanisms can be easily targeted by unlicensedusers to abuse the network Such mechanisms needto be strengthened further

(iv) Excessively complicated access control rules maybecause of their misunderstanding or poor writinggive unauthorized access allowing dubious activityrather than protecting the network they purport toprotect -ese should be simple and clear to un-derstand by all users

(v) In booting workstations without disks and managingnetwork devices we usually use the TFTP protocolalthough access does not require authenticating theuser by usernamepassword -is makes an intruderable to gain easy access to configaccess info as longas he can guess the filenames -is needs to be fixed

A few steps can be used to secure the application layerTo protect network data transmitted across the network

applications need to strengthen their encryption and au-thentication mechanisms Apps too need to enable strictcontrols of privileges for data access using good mechanismsto balance between usability and effectiveness [63] Appli-cations handling sensitive data need detailed logging andaudit capability as well as testing and reviewing

Table 3 summarizes the application layer protocols dis-cussed with their weaknesses and differences among them

4 Cross-Layer QoS Strategies

41 Service-Differentiated Real-Time Communication Scheme(SDRCS) -is is an event-driven routing protocol that routesreal-time traffic by featuring a cross-layer packet-forwardingdesign in which it integrates the real-time routing functionalitywith a new priority-based MAC scheme -e protocol ap-proximates distributed packet traversal speed for traffic clas-sification and admission control on the basis of this design Itfurther localizes decision-making by prioritizing packet for-warding to maximize the speed of packets [64]

Advantages

(i) It improves bandwidth utilization bandwidthdegradation due to unschedulable packets can beavoided

(ii) It offers a lightweight packet schedulability estima-tion mechanism Here it uses received signalstrength and admission control and then earlymissed-deadline packet-dropping policies are made

(iii) It delivers high end-to-end throughput via highersource data rates alongside strict end-to-end latencyneeds

(iv) Nodes which meet real-time requirements of lesstraffic load and better channel quality are givenhighest priority when it comes to packet forwarding

-e SDRCS operates via 5 mechanisms namely (1) RSS-(Rich Site SummaryReally Simple Syndication-) basedgrouping (2) admission control (3) prioritized queuing(4) real-time MAC and (5) dynamic forwarding -isprotocol has a queuing policy that employs per-hopdeadline-based queues and the nodes schedule theirpackets using FIFO priority-based queues [1] -e prin-cipal features of this protocol are mentioned as follows

(i) It is event driven and is easily adaptable to anynetwork changes

(ii) It does not need additional hardware for locali-zation or multichannel transmission

(iii) -e cross-layer packet-forwarding strategy can beused when transmitting multimedia traffic

(iv) It is able to efficiently circumvent any voids therein

Early deadline miss (EDM) policy drops all unscheduledpackets -e SDRCS circumvents wrong packet dropsdue to EDM since it adjusts to network changes easily

42 Sensor Fuzzy-Based Image Transport (SUIT) -is pro-tocol regulates congestion using a scheme based on fuzzy

24 Journal of Computer Networks and Communications

logic It transmits packets with reduced quality to allowablelevels if there is congestion enabling it to transmit lower-quality packetsframes without being dropped-e numberof packets delivered per second thus rises and while thisenhances video streaming apps it is more appropriatelymade for those that quickly transmit non-real-time videosuch as in video surveillance SUIT can stream JPEG pic-tures other than the traditional predictive video codingmethods that are not suitable for sensor devices because oftheir constraints of energy memory and computationalspeed Progressive JPEG-PJPEG performs better via packettransmission rate and delay though it degrades the imagequality because of congestion [65] -is protocol employscross-layer methods for interlayer data exchange though itstill employs some layering making it still possible topreserve all advantages of the layered design since it hasapplication and transport layers but has noMAC or routinglayer

For congestion detection and control SUIT bases on thepercentage number of in-bound and out-bound frames perwindow number of contenders and buffer occupancy of thenext-hop node Since buffer management has an effect oneffectiveness and QoS inWMSNs it is useful in the transportlayer so the protocol prioritizes packets for better QoS SUITreorders packets by tracking the sequence numbers of thedelivered packets since they all have a source ID framenumber and sequence number that it uses to detect anymissing packets -e protocol however does not providepacket reliability [65]

Advantages

(i) -e protocol prioritizes packets for better QoS(ii) It can transmit lower-quality packetsframes

without being dropped

Disadvantages

(i) It does not provide packet reliability

43 Network Layer QoS Support Enforced by a Cross-LayerController (NLQS) -is scheme permits packet-level servicedifferentiation as a function of throughput end-to-endpacket error rate and delay [66] (see Figure 8) -is en-hances network layer QoS It comprises a cross-layer controlunit (XLCU) for configuring and controlling networkingfunctionalities at physical MAC and network layers [18]-is depends upon the unified logic which affects choices forapplication layer needs and status of functional blocks whichdo implementation of networking functions [66]

Advantages

(i) Cross-layer interactions can be controlled withoutweakening the upgradability modularity andsimplicity of designing the system

44 Cross-Layer Signaling Shortcuts (CLASS) CLASS has ahigh level of effectiveness flexibility and comprehensive-ness According to [67] it has the following features

(i) Direct signaling among remote layers leading tofaster transmission

(ii) Lightweighted internal message format standard-ized protocols are not suitable for use in internalsignaling since they are not usually lightweight forinstance transmitting against faults in the networkCLASS requires only three fields namely destina-tion address event type and event contents

(iii) -e setup of its exterior messages is standardconsidering external signaling ICMP and TCPIPare suitable for general messages and short notifi-cations respectively

(iv) Message control protocol should ensure optimizedand organized exchange of dense simultaneousmessages across layers for high efficiency andavoidance of possible conflicts -e protocol appliesto many cross-layer signaling situations

Table 3 Differences and problems with application layer protocols

Protocol Architecture Transport QoS options Security Weaknesses

CoAP Requestresponse UDP DTLS No in-built security features DTLS does not supportmulticast

MQTT Publishsubscribe TCP TLSSSL Insufficient security at the protocol level sincepayload values are not encrypted

SMQTT Publishsubscribe TCP CPKP-ABE Key revocation and group pubsub for distributedSMQTT are still a challenge

MQTT-SN Publishsubscribe UDP TLSSSL SSLTLS suffers from attacks like BEAST CRIMERC4 and Heartbleed

XMPP RequestresponsePublishsubscribe TCP X TLSSSL Susceptible to spamming attacks and lacks worldwide

supportAMQP Publishsubscribe TCP TLSSSL Inappropriate for real-time applications

DDS Publishsubscribe TCPUDP TLSSSL Memory intensive and no open-source libraries forconstrained devices

REST Requestresponse HTTP X HTTPS Hard to implement Uses HTTP so incompatiblewith constrained apps

WebSockets PublishsubscribeClientserver TCP X TLSSSL Bad user experience for IoT apps since it runs the

clientserver architectureSTOMP Clientserver TCP HTTP -e broker can act as a bottleneck

Journal of Computer Networks and Communications 25

Advantages

(i) e protocol is scalable and has a very lowpropagation delay

(ii) It is very ecient and exible since it uses directsignaling between any two layers

(iii) It is very comprehensive and eective

45 Cross-Layer and Multipath-Based Video Transmission(CMVT) CMVT is a hybrid of application and networklayers At the application layer it carries out the encoding ofvideo streams to video data frames (I- P- and B-frames) bymeans of the MPEG-4 encoding format At the networklayer route discovery and data transfer take place Twoalgorithms greedy forwarding and rollback are used todiscover a number of paths from the source to the sink nodevia the route discovery method [20] A specied node icalculates the evaluation of its adjacent node j using thefollowing equation

fij (1 minus α)d2(j D) minus d2min(i)d2max(i) minus d

2min(i)

+ αeinit(j) minus eres(j)

einit(j)

(19)

where fij evaluation value for the node i to j d2(j D) isthe distance from the node j to the destination node D

d2min(i) and d2max(i) are the minimum and maximum dis-

tances of neighbors of the node i to D respectively einit(j) isthe initial energy of the node j eres(j) is the current residualenergy of the node j and α is the energy coecient given asfollows

α emax(i) minus emin(i)

emax(i) (20)

where emax(i) and emin(i) are respectively the maximumand minimum energy left of all node irsquos neighbors

e network layer further transmits video streams withCMVT doing status evaluation choosing an appropriatecommunication path for any packets and the QoS guaranteelevel for the path i is calculated as

fi (1 minus ω)hisum hi

+ ωnisum ni

(21)

where fi is the evaluation value for the path i hi are the hopsfor the path i ni is the summation of packets sent via isum ni isthe summation of packets sent by sources and ω is theenergy consumption factor [18]

Advantages

(i) e CMVT protocol is very superior in mediatransmission particularly large-scale WMSNs

To network peers

Flow requirements

-

Atimes

Admissioncontrol

Capacityassignment

Next-hopselection

Coding rate

Data rate

QoScontracts

Cross-layer controlunit (XLCU)

Localization subsystem

Synchronizationsubsystem

Networkcoordinationunit (NCU)

UWB modulator

Channel coder

QoS scheduler

Georouting

Source coder

Rate control

Figure 8 Cross-layer controller architecture

26 Journal of Computer Networks and Communications

46 Cross-Layer Cooperative MAC (CoopMAC)CoopMAC [68] is based on the IEEE 80211 DCF approach-ere is a reduction in interference of neighboring cellsleading to uniform coverage in densely deployed networks[68] A connection is established by two-way authenticationbetween the sender and the receiver When a Clear-to-Send(CTS) packet and short interframe space (SIFS) are obtainedby the sender packets are transmitted right to the receiverwith no valuable cooperation If available the two nodes findout whether they can exchange data -is is done by the useof a helper identification signal to establish how feasiblecooperative communication is It is established in thepresence of a signal and if it is not available direct com-munication is activated For distributed systems helper-initiated cooperation is preferable because of the presence ofthe RTSCTS packets [69]

Advantages

(i) CoopMAC has an advantage over others due to thespatial diversity amongst the three nodes [70]

(ii) It has the potential to achieve significantthroughput and delay with no complexity in thedesign of the system

47 Cooperative MAC Protocol for Multihop Networks (M-CMAC) Just as CoopMAC M-CMAC is made in such away that high data rate stations help low data rate stations toforward traffic for broadcasting [71] -ere are also helperschosen in such a way that 2 fast-hop transmissions replace aslow hop transmission -e helper via whom there existsleast delay from the sender to the receiver is considered bestbecause of theminimum two-hop transmission rate and so isthe chosen neighbor to the duo Here the working as-sumption is that all nodes have their position coordinatesknown and thus the Euclidean distances between all pairscan be calculated and converted to the data rate for that linkAll nodes have cooperative tables (CT) of potential helperswith destination and helper MAC address Euclidean dis-tance and total distance through the helper Any sender withdata to transmit checks out for the existence of any helper inthe CT for the destination and if available forwards RTS forchannel reservation for single-hop duration M-CMAC hasgot an RTS format having a shape with five fields similar tothe one given as follows

Framecontrol Duration Source

addressDestination

addressHelperaddress

-e destination nodersquos address is kept by the source in thepart for the helper address that is in turn saved as the desti-nation address Nodes inspect helper and destination addressfields on receipt of RTS whereby the node acts as the othernodersquos helper in case the helper address field is not the sameand the helper transmits CTS back to the source if at all it wantsto forward data On delivery of the CTS packet the sourcetransmits to the helper which in turn sends the packets to thedestination that also sends ACK to the helper on receipt of thepacket In M-CMAC there exists a higher level of channelreuse (parallel transmissions) due to the increased number of

nodes therein that subsequently raises availability of helpers fordata forwarding -is results in increased throughput incomparison with CoopMAC and IEEE 80211 DCF [71]

Advantages

(i) It promises a higher throughput compared toCoopMAC and IEEE 80211 DCF

48 Cluster-Based Cooperative Routing (CBCR) Protocol-e CBCR protocol has a multihop data-forwarding func-tion realized at the link layer with cooperative links that useM-CMAC -is protocol encompasses two stages namelythe routing relay selection phase and data forwarding phase

(i) Routing relay selection phase all nodes announcetheir presence to their neighbors by broadcastingperiodical beacon messages that carry the MAC ad-dress of the node Each of these constructs a relaytable that contains all neighboring nodes with whichit is able to communicate -e node further broad-casts its neighbor list in case of any change to itsentries since the previous broadcast -e MAC ad-dresses of nodes next to the node X are contained inits relay tablersquos column one and the neighbor nodersquosrow has MAC addresses for neighbors of the adjacentnode Based on its relay table each node in-dependently chooses routing relays Choosing a nodeas a relay node depends on the number of nodes itconnectsmdashit should be the highest number [71]

(ii) Data forwarding phase a node with packets totransmit needs to first verify if the receiver is in thesame cluster and if that receiver has a helper to whichthe packet is thence forwarded but in whose absencethe packet is delivered right to the receiver In casethe intended receiver is in a different cluster therelay table will be checked to see whether that re-ceiver is reachable via other routing relays If it isreachable packets will be sent to the routing relaydirectly or via the helper if at all the relay has one Incase of destination unreachable via relays packetsare multicast to all relays by the node [71]Advantages

(i) It enables multicasting

49 MAC-PHY Cross-Layer Protocol A cooperative cross-layer standard for cooperation at the physical layer in next-generation WMSNs is developed in [70] It provides acomplete MAC layer algorithm that gives a supportive shellfor the PHY-MAC layer Like CoopMAC its scheme de-pends on the middle node for communication betweennodes-eMAC layer protocol has been changed to regulateinformation exchanges at the physical layer -e receivernode receives duplicate packets from source and helpernodes to decode the data [70]

410 MAC-Centric Approach -is cross-layer protocoltargets multimedia applications by using the MPEG-4

Journal of Computer Networks and Communications 27

scheme [21] It is characterized by 4 access categories(AC3ndashAC0) according to their priority in transmission It ismeant for supporting different QoS needs in upcomingvideo applications and enabling MAC layer differentiationfor H264 partitioning [18] Because of low bandwidth delayand other QoS challenges that cause inefficiency in trans-mittingmultimedia data inWMSNs a number of algorithmsdepending on IEEE 80211e have been proposed to supporttransmission of quality videos [72] Selecting an AC dependson the measures for QoS such as the loss rate and delay-us AC3 having the highest priority is mapped to theparameter set concept since the stream is sensitive to bit lossespecially in video transmission [21]

411 Adaptive Cross-Layer Forward Error Correction(ACFEC) In the ACFEC [73] model data packets are ex-changed among nodes through the access point (AP) whichoperates the infrastructure mode in which it adds FEC to videodata [72] -ese data are dealt by encapsulation using astreaming server to the receiver through thewireless AP as RTPpackets An adaptive FEC controller senses the category ofpackets out of the RTP header and recovers the header ofpackets from UDP -e encoder generates some error-cor-recting packets whose number is determined by the sourcepacket number of the block Multimedia transmissions aremonitored by the controller using MAC failure data and itscounter is increased by one if there is a failed transmission-econtroller uses the failure counter to change the packet numberproduced after transmitting a block [18] In case there are lostpackets it changes the redundancy rates and produces extrapackets to replace the lost ones and satisfy the requirements ofthe receiving node [72] -e FEC packet number is enhancedor reduced to satisfy the receiverrsquos requirements and stoppacket losses-is is done through accurate detection of packetlosses and adjustment of redundancy rates If all video datapackets are well received there will not be generation of FECpackets [73]

Advantages

(i) It promises a good QoS through packet loss re-duction and redundancy rate adjustment

412 Balanced Cross-Layer Fuzzy Logic (BCFL) DesignRouting Algorithm in WSNs A new fuzzy logic-basedrouting algorithm (BCFL) [74] was designed using disper-sion of the cross-layer parameter as the fuzzy logic inferencesystem input Each cross-layer parameter has a dynamicweight depending on the value of dispersion -e designcomes with some innovations as follows (i) for fuzzy logicinference system input the absolute parameter value issubstituted by parameter dispersion thereby significantlyreducing algorithmic complexity (ii) dispersion does notchange with the order of magnitude according to the dis-persion formula and (iii) the weight of the parameter de-pends on the size of its dispersion and the two are inverselyproportional to each other -is enables BCFL to have somenobility unlike other algorithms as per the following dis-tinguishing properties making it advantageous

(i) It has got simple if-then rules which remain con-stant even when the constraints increase

(ii) It is capable of dealing with many constraints withno increment in complexity

(iii) It has capacity to yield a more balanced solutionthan other algorithms

(iv) It is easily adaptable to changes in network con-ditions and topology even when the changes arefrequent like in underwater WSNs -e algorithm isuseable in choosing the CH in cluster-based routingprotocols [74]

413 Minimum Hop Disjoint Multipath Routing Algorithmwith Time Slice Load-Balancing Congestion Control SchemeMHDMwTS is a two-phased routing protocol comprisingpath build-up and path acknowledgment phases It consists ofmultiple sources each of which has three build-up disjointpaths namely primary alternate and back-up paths Onactivation the source node requests to build up a route to thenearest hop neighbor which is the path build-up phase [19] Inthis phase step one has source activation in which therequested node for path building adds its number and time-stamp and sends to the least hop-count neighboring node-isgoes on until the least time latency sink having the informationneeded to construct the primary route is reached [75] In steptwo there is path extraction when the new package fromanother path arrives whereby the extracted path is comparedwith the primary path If the node is shared the package will berejected or an alternative path will be searched for to get abackup through comparison of the preceding two paths Phasetwo is path acknowledgment where we have the third step inwhich the sink returns the ACK packet to the sender with pathinformation having nodes and their time information after ithas computed it using the timestamp [1819]

Advantages

(i) -e protocol minimizes end-to-end latency(ii) It enhances congestion control

414 Cross-Layer Optimal Design (CLOD) Authors in [76]come up with a CLOD for scheduling at the data link layerrouting at the network layer and controlling congestion atthe transport layer with an assumption of fixed link capacity-rough congestion control energy efficiency is improvedTransport layer congestion at the nodes is minimized bycompressed sensing (CS) in which transmitted bits are re-duced whereas optimally allocating resources decreasescongestion on the links at the data link layer CLODpromises minimized computational complexity and betterperformance in light traffic scenarios [77] Generating andstoring CLOD is more efficient compared to Gaussianrandom matrices It prolongs network lifetime and savesenergy

Advantages

(i) It does congestion control subsequently increasingenergy efficiency

28 Journal of Computer Networks and Communications

(ii) It reduces computational complexity(iii) It improves performance in light traffic scenarios

415 Challenges and Opportunities of the Cross-Layer Design(CLD)

(i) Different layers in a real network assume differentfunctions andor services A layer only communi-cates with its neighboring layers -e layered modeldegrades system performance because of many

features in wireless transmission QoS parameters inthe layers of WMSNs are shared by routing pro-tocols to optimize the performance Neverthelessthere is a need to design cross-layer models forincreased efficiency in routing [1]

(ii) -e physical layer plays a highly vital role in CLDRate adaptation and channel allocation take place atthe physical layer via signal processing to enhanceQoS End-to-end performance in wireless media isaffected by the changes therein given the effect on

Table 4 A summary of some of the reviewed cross-layer models

Protocol Layers Aims Comments QoS parameters

SDRCS MACPHY Routing real-time traffic Transmits multimedia traffic viacross-layer packet forwarding

-roughputlatency

SUIT Applicationtransport Transmission of non-real-time video Uses fuzzy logic to regulate

congestionTransmission rate

delay

CLASS Any twoDesign serves as a framework fordifferent implementations ofdifferent application scenarios

Scalable very efficient flexible andhas very low propagation delay

Propagation delayjitter

NLQS PHYMACnetwork

Permits packet-level servicedifferentiation as a function of

throughput packet error rate anddelay

Network layer QoS is enhanced -roughput delaypacket error rate

CMVT Applicationnetwork

Encodes video streams to video dataframes via MPEG-4 encoding anddoes route discovery and data

transfer

Greedy forwarding and rollback areused to find source-to-sink paths Energy

CoopMAC PHYMAC Offers spatial diversity among thethree nodes

Helper ID signal finds out howfeasible cooperative communication

is

-roughput ratedelay

M-CMAC PHYMAC Increases end-to-end throughputand packet delivery ratio

Euclidean distances between nodesare calculated and converted to the

data rate for a link

-roughputpacket delivery

ratio

CBCR PHYMACMinimizes control overhead andtime consumed in establishing thecooperative paths than M-CMAC

Energy consumption is moreuniformly distributed in a network

enhancing network lifetime

-roughputpacket deliveryratio energy

MAC-PHY MACPHYMeant to enable PHY layer

cooperation and maximize gains ofcooperation at the MAC layer

Leverages both spatial diversity andcoding gain -roughput delay

MAC-centric MACAPPL

Meant to support QoS needs in newvideo apps and enable MAC layer

differentiation for H264partitioning

Targets multimedia applications byusing the MPEG-4 scheme

Delay packet lossrate

ACFEC MACnetwork(UDP)

Meant to enhance the quality ofvideo streaming over 80211WLANs

and overcome packet losses

Adjusts redundancy rates toovercome channel fluctuations and

detect and reduce packet lossPacket loss rate

MHDMwTS mdash

Meant to provide reliable datatransfer with the multipath routingand load-balancing congestioncontrol method in WMSNs

More reliable than basic routingschemes for transport multimedia

data

Latency packagetransmit rate

BCFL mdash

Introduces dispersion into fuzzylogic-based routing and sets every

cross-layer parameter with adynamic weight

Can be used to select a CH in cluster-based routing protocols andproposes a dispersion formula

Node utilitydispersion

CLODDatalinknetworktransport

Prolongs network lifetime andachieves congestion control anddesigned for lightly loaded WSNs

Assumes fixed link capacity andintegrates compressed sensing

technology

-roughputaverage energy CS

error ratio

Journal of Computer Networks and Communications 29

the purposefulness of the protocols at the networklayer CLD offers solutions to conserving powerminimizing energy and controlling both the flowand congestion in the network as well as fault tol-erance so it is an opportunity for designers con-sidering other layers We further desire to developCLDs that shall make an impact on network oper-ation and get close attention [18]

(iii) CLD interface standardization the architecturemust deliver module functionality although thereexist some queries amongst modules about possibleinterfaces predicted by the necessity to share in-formation among remote protocol layers Technicalchallenges include developing designing andstandardizing of cross-layer interfaces and algo-rithms that satisfy the cross-layer optimization re-quirements among protocol layers

Table 4 summarizes these cross-layer models with a fewcomments

5 Future Research Direction

51 At theMACLayer In WMSNs achieving duty cycling istricky since video traffic is volatile Listening to adjacentnodes wastes energy We need to switch states from awake tosleep -is requires further research

Self-organizing networks and green communicationalthough most CRWSN energy-saving algorithms aredesigned to reduce node transmission power energyconsumption is inevitable during the operational mode ofBSs because of internal processing Scholars are trying withpolicy-makers to encourage a shift to greenWSNs to reduceoperational costs and carbon footprints With self-orga-nizing CRWSNs for automatic switch-off when idle energyis saved which calls for cognitive switching algorithms Sointegrating cognitive switching with radio resource allo-cation in CRWSNs is a promising research area [78]

52 At the Application Layer Considering a WMSN ar-chitecture end users access the network via the BS to andfrom which they send their sensed data We desire to getguaranteed QoS and energy efficiency by designing a newarchitecture in the form of hardware and software whichneeds to be investigated further for a fix of the same [1]

QoS and energy efficiency are very crucial in WMSNsand mostly real-time apps that require guaranteed band-width and throughput in their network lifetime Mostprotocols ignore base station (BS) mobility and WMSNnodes Traffic management telemedicine and battlefieldsurveillance apps require mobile nodes or BSs so designingdynamic routing protocols to be adaptable in these cir-cumstances is necessary

53 Cross-Layer Security of data transmitted on WMSNseg in military surveillance and e-commerce is key withQoS and energy efficiency WMSNs are vulnerable to attacks

like worms and sinkhole Moreover their computationpower is not that high complicating implementation ofstrong secure protocols -is needs further attention

For fairness and priority issues in CRWSNs nodes areassigned diverse priorities depending on importance andurgency of their data For better network performance thereis a need to develop fair resource allocation mechanismswith priority-based fairness in sensors-ere is still potentialin this research area [78]

We must develop multiobjective adaptive protocols tooptimize QoS metrics involved in routing for best trans-mission results to get a trade-off between different opti-mization metrics We need to design standardized cross-layer algorithms which are able to meet cross-layer opti-mization standards

6 Conclusion

We have reviewed the different QoS strategies for WSNs inthe context of IoT from the MAC layer and application layeras well as the cross-layer paradigm For the MAC layer wehave reviewed protocols for WSNs such as the hybrid of thecontention-free and contention-based MAC protocols andthose for WMSNs We further reviewed a number of ap-plication layer protocols including many machine-to-ma-chine requestresponse and publishsubscribe protocols Forsystem optimization cross-layer QoS strategies are veryimportant in wireless communication We have reviewed anumber of cross-layer strategies For all categories reviewedchallenges and opportunities are discussed Finally somepossible future directions are discussed for research andapplication showing a promising potential for future re-search in this relatively new area especially as more WMSNapplications emerge recently

Conflicts of Interest

-e authors hereby declare no conflicts of interest regardingthe publication of this work

Acknowledgments

-is research was funded by the National Natural ScienceFoundation of China (Nos 61602398 61672447 and61711540306)

References

[1] V Bhandary A Malik and S Kumar ldquoRouting in wirelessmultimedia sensor networks a survey of existing protocolsand open research issuesrdquo Journal of Engineering vol 2016Article ID 9608757 27 pages 2016

[2] I F Akyildiz T Melodia and K R Chowdhury ldquoWirelessmultimedia sensor networks applications and testbedsrdquoProceedings of the IEEE vol 96 no 10 pp 1588ndash1605 2008

[3] M Iqbal M Naeem A Ahmed M Awais A Anpalagan andA Ahmad ldquoSwarm intelligence based resource managementfor cooperative cognitive radio network in smart hospitalsrdquoWireless Personal Communications vol 98 no 1 pp 571ndash5922018

30 Journal of Computer Networks and Communications

[4] S A Alvi B Afzal G A Shah L Atzori and W MahmoodldquoInternet of multimedia things vision and challengesrdquo AdHoc Networks vol 33 pp 87ndash111 2015

[5] Q F Hassan A U R Khan and S A Madani Internet ofings Challenges Advances and Applications CRC PressTaylor amp Francis Group A Chapman amp Hall Group BocaRaton FL USA 2018

[6] Q Wang Y Zhao W Wang et al ldquoMultimedia IoT systemsand applicationsrdquo in Proceedings of the 2017 Global Internet ofings Summit (GIoTS) vol 2 Geneva Switzerland June 2017

[7] A Aslam and E Curry ldquoTowards a generalized approach fordeep neural network based event processing for the internet ofmultimedia thingsrdquo IEEE Access vol 6 pp 25573ndash25587 2018

[8] M Jridi T Chapel V Dorez L Bougeant and A Le BotlanldquoSoC-based edge computing gateway in the context of theinternet of multimedia things experimental platformrdquoJournal of Low Power Electronics and Applications vol 8no 1 p 1 2018

[9] H Noura A Chehab L Sleem andM N Rapha ldquoOne roundcipher algorithm for multimedia IoT devicesrdquo MultimediaTools and Applications vol 77 no 14 pp 18383ndash18413 2018

[10] T Balan D Robu and F Sandu ldquoMultihoming for mobileinternet of multimedia thingsrdquo Mobile Information Systemsvol 2017 Article ID 6965028 16 pages 2017

[11] H Shen and G Bai ldquoRouting in wireless multimedia sensornetworks a survey and challenges aheadrdquo Journal of Networkand Computer Applications vol 71 pp 30ndash49 2016

[12] A Ahmad M H Rehmani H Tembine O A Mohammedand A Jamalipour ldquoIEEE access special section Editorial optimization for emerging wireless Networks IoT 5G andsmart grid communication networksrdquo IEEE Access vol 5pp 2096ndash2100 2017

[13] B A Muzakkari M A Mohamed M F A Kadir andZ Mohamad ldquoRecent advances in energy efficient-QoS awareMAC protocols for wireless sensor networksrdquo InternationalJournal of Advanced Computer Research vol 8 no 38pp 212ndash228 2018

[14] M A Yigitel O D Incel and C Ersoy ldquoQoS-aware MACprotocols for wireless sensor networks a surveyrdquo ComputerNetworks vol 55 no 8 pp 1982ndash2004 2011

[15] A M Abbas and O Kure ldquoQuality of Service in mobile ad hocnetworks a surveyrdquo International Journal of Ad Hoc andUbiquitous Computing vol 6 no 2 p 75 2010

[16] M Q Shatnawi ldquoApplication layer protocols for the internetof things a surveyrdquo in Proceedings of the InternationalConference on Engineering amp MIS (ICEMIS) Agadir Mo-rocco September 2016

[17] V Karagiannis P Chatzimisios F Vazquez-gallego andJ Alonso-zarate ldquoA survey on application layer protocols forthe internet of things research motivationrdquo InternationalJournal of Advanced Research in Computer Science vol 8no 3 pp 1ndash10 2015

[18] M S Bernard T Pei Z Li and K Li ldquoQoS strategies forwireless multimedia sensor networks in the context of IoTrdquoLecture Notes of the Institute for Computer Sciences SocialInformatics and Telecommunications Engineering vol 4pp 228ndash253 2019

[19] A AlAmri and M Abdullah ldquoCross-layer quality of serviceprotocols for wireless multimedia sensor networksrdquo in Pro-ceedings of the International Conference on CommunicationManagement and Information Technology pp 649ndash658Cosenza Italy April 2017

[20] J Guo L Sun and R Wang ldquoA cross-layer and multipathbased video transmission scheme for wireless multimedia

sensor networksrdquo Journal of Networks vol 7 no 9pp 1334ndash1340 2012

[21] A Ksentini A Naimi and M Gueroui ldquoToward an im-provement of H264 video transmission over IEEE 80211ethrough a cross-layer architecturerdquo IEEE CommunicationsMagazine vol 44 no 1 pp 107ndash114 2006

[22] W Ye J Heidemann and D Estrin ldquoAn energy-efficientMAC protocol for wireless sensor networksrdquo in Proceedings ofthe 21st International Annual Joint Conference of the IEEEComputer and Communications Societies (INFOCOM 2002)New York NY USA June 2002

[23] J Polastre J Hill and D Culler ldquoVersatile low power mediaaccess for wireless sensor networks categories and subjectdescriptorsrdquo in Proceedings of the Second InternationalConference on Embedded Networked Sensor Systems Sen-Sysrsquo04 pp 95ndash107 Baltimore MA USA November 2004

[24] T Danmanee K N Nakorn and K Rojviboonchai ldquoCU-MAC a duty-cycle MAC protocol for internet of things inwireless sensor networksrdquo ECTI Transactions on ElectricalEngineering Electronics and Communications vol 16 no 2pp 30ndash43 2018

[25] T Van Dam and K Langendoen ldquoAn adaptive energy-efficientMAC protocol for wireless sensor networksrdquo in Proceedings ofthe 1st International Conference on Embedded NetworkedSensor Systems Los Angeles CA USA November 2003

[26] P Xie and J Cui ldquoR-MAC an energy-efficient MAC protocolfor underwater sensor networksrdquo in International Conferenceon Wireless Algorithms Systems and Applications (WASA2007) Chicago IL USA August 2007

[27] A Rahim N Javaid M Aslam U Qasim and Z A KhanldquoAdaptive-reliable medium access control protocol forwireless body area networksrdquo in Proceedings of the AnnualIEEE Communications Society Conference on Sensor Meshand Ad Hoc Communications and Networks (SECON) SeoulKorea June 2012

[28] H S Kwak S Ullah D H Kwak and C H Lee ldquoA powerefficient MAC protocol for wireless body area networksrdquoJournal of the Korean Institute of Intelligent Systems vol 8no 6 pp 131ndash140 2009

[29] C Omeni A J Burdett and M Park ldquoEnergy efficientmedium access protocol for wireless medical body area sensornetworksrdquo in Proceedings of the 4th IEEEEMBS InternationalSummer School and Symposium on Medical Devices andBiosensors pp 29ndash32 Cambridge UK August 2007

[30] L Sitanayah C J Sreenan and K N Brown ldquoER-MAC ahybrid MAC protocol for emergency response wireless sensornetworksrdquo in Proceedings of the Fourth International Con-ference on Sensor Technologies and Applications pp 244ndash249Mestre Italy July 2010

[31] Y Sun and D B Johnson ldquoRI-MAC a receiver-initiatedasynchronous duty cycle MAC protocol for dynamic trafficloads in wireless sensor networksrdquo in Proceedings of the 6thACM conference on Embedded network sensorsystemsmdashSenSysrsquo08 pp 1ndash14 Raleigh NC USA November2008

[32] M Buettner G V Yee E Anderson and R Han ldquoX-MAC ashort preamble MAC protocol for duty-cycled wireless sensornetworksrdquo in Proceedings of the 4th International Conferenceon Embedded Networked Sensor Systems SenSys 2006pp 307ndash320 Boulder Colorado USA October 2006

[33] S Boulfekhar and M Benmohammed ldquoSynchronous receiverinitiated MAC protocol for long-lived sensor networksrdquoComputers amp Electrical Engineering vol 40 no 2 pp 504ndash516 2014

Journal of Computer Networks and Communications 31

[34] S Mehta and K S Kwak ldquoH-MAC a hybrid Mac protocol forwireless sensor networksrdquo International Journal of ComputerNetworks amp Communications vol 2 no 2 pp 108ndash117 2010

[35] I Rhee A Warrier M Aia J Min and M L Sichitiu ldquoZ-MAC a hybrid MAC for wireless sensor networksrdquo IEEEACM Transactions on Networking vol 16 no 3 pp 511ndash5242008

[36] S Arshad A Al-sadi and A Barnawi ldquoZ-MAC performanceevaluation and enhancementsrdquo Procedia Computer Sciencevol 21 pp 485ndash490 2013

[37] Y Liu I Elhanany and H Qi ldquoAn energy-efficient QoS-aware media access control protocol for wireless sensornetworksrdquo in Proceedings of the IEEE International Confer-ence on Mobile ad hoc and Sensor Systems pp 9ndash11 Wash-ington DC USA December 2005

[38] J Decotignie ldquoWiseMAC an ultra low power MAC protocolfor the downlink of infrastructure wireless sensor networksrdquoin Proceedings of the Ninth International Symposium onComputers And Communications pp 244ndash251 AlexandriaEgypt June 2004

[39] R A Rashid ldquoDevelopment of energy aware TDMA-basedMAC protocol for wireless sensor network systemrdquo EuropeanJournal of Scientific Research vol 30 no 4 pp 571ndash578 2009

[40] H P Van Hoesel ldquoA lightweight medium access protocol(LMAC) for wireless sensor networksrdquo in Proceedings of theInternational Workshop on Networked Sensing Systems Bal-timore MD USA November 2004

[41] C Li P Wang H Chen andM Guizani ldquoA cluster based on-demand multi-channel MAC protocol for wireless multi-media sensor networksrdquo in Proceedings of the IEEE In-ternational Conference on Communications pp 2371ndash2376Beijing China May 2008

[42] J L Sobrinho and A S Krishnakumar ldquoQuality-of-service inad hoc carrier sense multiple access wireless networksrdquo IEEEJournal on Selected Areas in Communications vol 17 no 8pp 1353ndash1368 1999

[43] J P John ldquoProviding MAC QoS for multimedia traffic in80211e based multi-hop ad hoc wireless networksrdquo ComputerNetworks vol 51 no 1 pp 153ndash176 2007

[44] S Lofty and N Padmavati ldquoA survey on mobility basedprotocols in WSNsrdquo in Proceedings of the InternationalConference Communication and Manufacturing pp 12ndash15Hangzhou China July 2014

[45] S Deng L Shen and J Li ldquoMobility-based clustering pro-tocol for wireless sensor networks with mobile nodesrdquo IETWireless Sensor Systems vol 1 no 1 pp 39ndash47 2011

[46] M Khalil A Khalid F U Khan and A Shabbir ldquoA review ofrouting protocol selection for wireless sensor networks insmart citiesrdquo in Proceedings of the 24th Asia-Pacific Confer-ence on Communications (APCC) Ningbo China November2014

[47] O Message Q Telemetry and T Mqtt Committee Specifi-cation Draft 02 April 2014

[48] M Collina M Bartolucci A Vanelli-coralli andG E Corazza ldquoInternet of things application layer protocolanalysis over error and delay prone linksrdquo in Proceedings ofthe 7th Advanced Satellite Multimedia Systems Conference andthe 13th Signal Processing for Space CommunicationsWorkshop Livorno Italy September 2014

[49] A Talaminos-barroso M A Estudillo-valderrama L M RoaJ Reina-tosina and F Ortega-ruiz ldquoA machine-to-machineprotocol benchmark for e health applicationsmdashuse case re-spiratory rehabilitationrdquo Computer Methods and Programs inBiomedicine vol 129 pp 1ndash11 2016

[50] V Goyal O Pandey A Sahai and B Waters ldquoAttribute-based encryption for fine-grained access control of encrypteddatardquo in Proceedings of the 13th ACM Conference on Com-puter and Communications Security pp 89ndash98 AlexandriaVA USA October 2006

[51] M Singh M A Rajan V L Shivraj and P BalamuralidharldquoSecure MQTT for internet of things (IoT)rdquo in Proceedings ofthe Fifth International Conference on Communication Systemsand Network Technologies Gwalior India April 2015

[52] A Stanford-Clark and H L Truong MQTT for Sensor Net-works (MQTT-SN) Protocol Specification httpmqttorgnewwp-contentuploads200906MQTT-SN_spec_v12pdf 2013

[53] RFC ldquoExtensible messaging and presence protocol (XMPP)rdquoin Proceedings of the IETF pp 1ndash9 San Diego CA USAAugust 2005

[54] S Vinoski and I Technologies ldquoQueuing protocolrdquo IEEEInternet Computing vol 10 no 6 pp 87ndash89 2006

[55] J L Fernandes I C Lopes J J P C Rodrigues and S UllahldquoPerformance evaluation of restful web services and AMQPprotocolrdquo in Proceedings of the Fifth International Conferenceon Ubiquitous and Future Networks (ICUFN) pp 2ndash7 DaNang Vietnam July 2013

[56] M Asim ldquoA survey on application layer protocols for internetof things (IoT)rdquo International Journal of Advanced Researchin Computer Science vol 8 no 3 pp 996ndash1000 2017

[57] K An A Gokhale D Schmidt S Tambe P Pazandak andG Pardo-castellote ldquoContent-based filtering discovery pro-tocol (CFDP) scalable and efficient OMG DDS discoveryprotocolrdquo in Proceedings of the 8th ACM InternationalConference on Distributed Event-Based Systems New YorkNY USA July 2014

[58] E Web What is a web service pp 52ndash57 2010[59] M Laine RESTful Web Services for the Internet of ings

pp 2ndash4 2012 httpmediaTkk[60] IETF -e WebSocket protocol pp 1ndash71 2011[61] V Pimentel ldquoCommunicating and displaying real-time data

with websocketrdquo IEEE Internet Computing vol 16 no 4pp 45ndash53 2012

[62] Wikipedia 2019 httpsenwikipediaorgwikiStreaming_Text_Oriented_Messaging_Protocol2019

[63] K Raghavendra and S Nireshwalya ldquoApplication layer se-curity issues and its solutionsrdquo IJCSET vol 2 no 6pp 1266ndash1269 2012

[64] Y Xue B Ramamurthy andM C Vuran ldquoSDRCS a service-differentiated real-time communication scheme for eventsensing in wireless sensor networksrdquo Computer Networksvol 55 no 15 pp 3287ndash3302 2011

[65] C Sonmez S Isik M Y Donmez O D Incel and C ErsoyldquoSUIT a cross layer image transport protocol with fuzzy logicbased congestion control for wireless multimedia sensornetworksrdquo in Proceedings of the 5th International Conferenceon New Technologies Mobility and Security (NTMS) pp 1ndash6IEEE Istanbul Turkey May 2012

[66] T Melodia and I Akyildiz ldquoCross-layer QoS-aware com-munication for ultra wide band wireless multimedia sensornetworksrdquo IEEE Journal on Selected Areas in Communica-tions vol 28 no 5 pp 653ndash663 2010

[67] Q Wang and M A Abu-Rgheff ldquoCross-layer signalling fornext-generation wireless systemsrdquo in Proceedings of the 2003IEEE Wireless Communications and Networking 2003WCNC 2003 vol 2 pp 1084ndash1089 IEEE New Orleans LAUSA March 2003

[68] P Liu Z Tao S Narayanan T Korakis and S PanwarldquoCoopMAC a cooperative MAC for wireless LANsrdquo IEEE

32 Journal of Computer Networks and Communications

Journal on Selected Areas in Communications vol 25 no 2pp 340ndash354 2007

[69] H Shan H T Cheng and W Zhuang ldquoCross-layer co-operative MAC protocol in distributed wireless networksrdquoIEEE Transactions onWireless Communications vol 10 no 8pp 2603ndash2615 2011

[70] F Liu T Korakis Z Tao and S Panwar ldquoAMAC-PHY cross-layer protocol for wireless Ad-Hoc networksrdquo in Proceedingsof the 2008 IEEE Wireless Communications and NetworkingConference pp 1792ndash1797 IEEE Las Vegas NV USA March2008

[71] L Jacob and H R Shamna ldquoEfficient cooperative MAC androuting in wireless networksrdquo Transactions on Networks andCommunications vol 3 no 5 2015

[72] S Rao and K Shama ldquoCross layer protocols for multimediatransmission in wireless networksrdquo International Journal ofComputer Science amp Engineering Survey vol 3 no 3pp 15ndash28 2012

[73] L Han S Park S Kang and H Peter ldquoAn adaptive cross-layer FEC mechanism for video transmission over 80211WLANsrdquo in Proceedings of the 2009 International Conferenceon Internet pp 209ndash215 Las Vegas NV USA December2009

[74] N Li J Martınez and V H Dıaz ldquo-e balanced cross-layerdesign routing algorithm in wireless sensor networks usingfuzzy logicrdquo Sensors vol 15 no 8 pp 19541ndash19559 2015

[75] G Sun J Qi Z Zang and Q Xu ldquoA reliable multipathrouting algorithm with related congestion control scheme inwireless multimedia sensor networksrdquo in Proceedings of the2011 3rd International Conference on Computer Research andDevelopment vol 4 pp 229ndash233 IEEE Shanghai ChinaMarch 2011

[76] M Li Y Jing and C Li ldquoA robust and efficient cross-layeroptimal design in wireless sensor networksrdquoWireless PersonalCommunications vol 72 no 4 pp 1889ndash1902 2013

[77] J Yan M Zhou and Z Ding ldquoRecent advances in energy-efficient routing protocols for wireless sensor networks areviewrdquo IEEE Access vol 4 pp 5673ndash5686 2016

[78] A Ahmad S Ahmad M H Rehmani and N U Hassan ldquoAsurvey on radio resource allocation in cognitive radio sensornetworksrdquo IEEE Communications Surveys amp Tutorials vol 17no 2 pp 888ndash917 2015

Journal of Computer Networks and Communications 33

International Journal of

AerospaceEngineeringHindawiwwwhindawicom Volume 2018

RoboticsJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Active and Passive Electronic Components

VLSI Design

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Shock and Vibration

Hindawiwwwhindawicom Volume 2018

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawiwwwhindawicom

Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Control Scienceand Engineering

Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Journal ofEngineeringVolume 2018

SensorsJournal of

Hindawiwwwhindawicom Volume 2018

International Journal of

RotatingMachinery

Hindawiwwwhindawicom Volume 2018

Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Navigation and Observation

International Journal of

Hindawi

wwwhindawicom Volume 2018

Advances in

Multimedia

Submit your manuscripts atwwwhindawicom

Page 12: ReviewArticle - downloads.hindawi.com · ReviewArticle QoSStrategiesforWirelessMultimediaSensorNetworksinthe ContextofIoTattheMACLayer,ApplicationLayer,and Cross-LayerAlgorithms MuwongeSsajjabbiBernard

Radiomodel for protocols with lower power transitiondelays between transceiver states and their powerconsumption are modeled Some of the states includeDOZE where the transceiver neither transmits norreceives but is prepared to power on to the receivetransmit state very fast RXwhere the transceiver listensto the channel receives data or demodulates data fromthe noisyidle channel and TX which is the trans-mission state by the transceiver Let TS be the set-uptime needed to turn the transceiver from the DOZE tothe RX or TX state TT the turn-around time needed toswitch between RX and TX states and B the trans-ceiverrsquos bit rate PZ PR and PT are the values of powerconsumed in the DOZE RX and TX statesrespectively

1113954PR PR minus PZ

1113954PT PT minus PZ(12)

where 1113954PR and 1113954PT are the changes in power as a result ofbeing in RX and TX states respectivelyTraffic model let N be the number of sensor nodesconnected to an access point (AP) and λ be the global rateat which the downlink Poisson traffic arrives at the APfrom the fixed network Traffic toward each sensor node isthe same and is given by λN Average packet interarrivaltime Lwith which data packets arrive at a node is given byL Nλ If TC and TD are the control packets and datapackets respectively assuming a small traffic volume theglobal interarrival 1λ far exceeds the summation ofdurations for the data packet and control packet and theturn-around time Mathematically it is given as

1λ≫TD + TC + TT (13)

-e average power consumed by WiseMAC andtransmission delay incurred are given by equations (14)and (15) [38]

PW PZ +1113954PR TS +(1B)( 1113857

TW+

1113954PR X + TD + TT( 1113857

L

+ 1113954PR(N minus 1)Y

L

(14)

where

X 2θL 1 minus eminus TD4θL( )1113874 1113875

Y T2D + 12TDθL

2TW1 minus e

minus TW4θL( )1113874 1113875

DW TD +TW

21 minus e

minus TW4θL( )1113874 1113875

+ 2θL 2 minus eminus TD4θL( ) minus e

minus TW4θL( )1113874 1113875

(15)

where PW and DW are the power consumed byWiseMAC and the transmission delay for WiseMACrespectivelyAdvantages

(i) It achieves low-power transmissions in in-frastructure sensor networks through preamblesampling while checking for activity

(ii) It reduces overhearing due to increased trafficvolume

Disadvantages

(i) It has a problem of long wake-up preamble limitingthroughput and causing enormous power con-sumption overhead

2114 X-MAC It is a simple minimal transmission powerMAC protocol that uses a short preamble with decouplingfor the sending and receiving node sleep plans inWSNs-isleads to a significant reduction in energy usage at sendingand receiving nodes and decrease in per-hop latency andgives flexibility in adapting to bursty and periodic sensordata sources [32]

Protocol design X-MACrsquos design is aimed at increasingenergy efficiency reducing data latency and increasingthroughput -is enables application in different kindsof packetized and bit stream digitized radios as well assimplified low-overhead distributed application -edesign is as such in order to address the low-powerlistening problems including overhearing excessivepreamble and incompatibility with packetizing radiosas in [32]Asynchronous duty cycling a node with information totransmit will send the packets after sending a preamblewhereas the remaining nodes will preserve their sleepschedules that are not synchronized -e receiver willsample the link on waking up and stays awake if it detectsa preamble and if it is the target it will return to sleep if itis not the target after getting the whole preambleEmbedding the target ID in the preamble to avoidoverhearing the lengthy preamble is divided into manyshortened packets embedded with the receiver nodersquosID so as to eliminate overhearing One lengthy pre-amble is constituted by the shortened preamble packetstream Any node on receiving the short preamble afterwaking up will examine the target ID on the packet Itgoes back into the sleepmode right away resuming dutycycling if it is not the targeted recipient or stay awakefor successive packets if it is the one -is approachchecks energy wastage due to network density It canalso be used on several types of radiosReducing excessive preamble using strobing it ispossible to enable low-power transmission and alsosave energy by reducing time for preamble trans-mission through the use of enhanced preambles inaddition to preamble sampling X-MAC does not sendan endless preamble packet stream but puts slight

12 Journal of Computer Networks and Communications

pauses in between packets to allow the sending node topause and listen to the channel and allow the receivingnode to resend an early ACK packet during that pause-is makes the sending node begin sending datapackets and the receiving node cut excessive preamblesand thus decrease per-hop latency and energy wastageon unnecessary waiting and transmissionPacketizing radios low-power listening is limited incapability for supporting packetizing radios butX-MACrsquos short strobed preambles can support all typesof digital radiosAdaptation to traffic load while a number of WSNapplications are reliable in traffic production they stillmust adapt to changing traffic Nodes are going to haveseparate sleep schedules depending on traffic loadConsidering systems that have time-fluctuating trafficall predetermined static schedules are suboptimalOptimality Buettner et al [32] model the expectationof energy as follows

Es (preamble energy + energy perACK listen)

lowast(expected preamble minus listen iterations)

+(energy to send parket)

Es PTxSp + PRxSal1113872 1113873 Rl + Rs( 1113857

Rl minus Sp+ SdPTx

(16)

where Es expected energy to send a packet PTx andPRx are the transmission and receiver power re-spectively Sp Sal and Sd are the duration of thesenderrsquos preamble ACK listen and data transmissionperiods respectively and Rl and Rs are the receiverrsquoslisten and sleep periods respectively Expected energyto receive a packet is given as

Er (listen cycle energy + sleep cycle energy)

lowast(expected iterations for preamble arrival)

+(energy to sendACK) +(energy to receive packet)

Er PsRs + PRxRl( 1113857

1 minus 1 minus Pd(t)( 1113857 Rl + Rs( 1113857+ RaPTx + RdPRx

(17)

And the one-hop expected latency is given as

Lat (preamble duration + ACK listen)

lowast(expected number of iterations)

+ duration to send packet

Lat Sp + Sal1113872 1113873 Rl + Rs( 1113857

Rl minus Sp

(18)

Advantages

(i) It increases energy efficiency(ii) It reduces data latency(iii) It increases throughput

2115 A-MAC (Advanced MAC) A-MAC is a TDMA-based protocol whose design is aimed at enhancing thelifetime of the network for low rate and reliability of datatransmission [39] To assign time slots it does not depend ona central controller like most protocols but it gathers in-formation from its neighbors so as to allocate the slotsthrough a distributed algorithm Energy is controlled byscheduling power when there is not a single event Its designcomprises many frames with each having many time slotswhich at the beginning have got a beacon transmitted tosynchronize and exchange information with neighborsNodes to participate in the subsequent session are selectedby the controlled node so they do not go into the sleep modewith others [13] -is protocol operates under four time slotdivisions namely initial wait discover and active states Anode in the initial state begins by listening to the medium forthe beacon packet from other nodes for network synchro-nization which when received makes it change the timerthrough subtraction of beacon transmission time from re-ception time To avoid synchronization problems due todrift and allow for continuity the node has to get thestrongest beacon signal It then enters the wait state afterchoosing a number of waiting frames so as to reduce theprobability that nodes get into the discover state simulta-neously In case the synchronization beacon is lost while stillin the wait state it will return to the initial state for anotherone On expiry of the waiting counter it enters the discoverstate and collects information from neighbors by listening totheir beacon messages On successful selection of a time slotit enters the active state where it endlessly transmits beaconsat the initiation of the slot A node goes to sleep when abeacon has been transmitted and there are no more data fortransmission or in case the neighborrsquos beacon received in-dicates no data packets are coming in [39]

Advantages

(i) It can enable multicasting(ii) It enhances the lifetime of the network(iii) It offers reliable data transmission

2116 L-MAC (Lightweight MAC) Based on TDMA thisprotocol is aimed at minimizing the number of transceiverswitches adapting the nodal sleep interval to data trafficamount and making implementation less complex [40] Apacket is transmitted comprising a control and a data partControl packets have a fixed size and the time slot con-trollerrsquos ID -ey show the distance from the transmittingnode for simplified routing to the network gateway as well asaddressing target node and reporting data unit length andinternode synchronization Adjacent nodes are more in-terested in getting control packets from nearby nodes -enodersquos transceivers are switched off until the subsequenttime slot unless the node is also addressed or the message isomnicast If so it listens to the data unit which may not fillthe remaining time slot in totality-e transceivers on eitherside are turned off when the transfer of packets is successfulEnergy wastage through idle listening in uncontrolled timeslots is avoided via a short time-out interval Network nodes

Journal of Computer Networks and Communications 13

are able to communicate without colliding-is saves energyand increases the lifetime of the network A node in L-MACcan only transmit one message per frame [13] -e networkis set up in such a way that all nodes are not synchronized onbeing switched on-e gateway takes control of a timeslot soas to have them synchronized and its immediate neighborgets the control messages thereby synchronizing their timerswith the gatewayrsquos Neighboring nodes learn of multiplegatewaysrsquo time slots in their vicinity just after one frame-ejust synchronized nodes randomly choose time slots forcontrolling as long as they are not occupied -e nodes thuspreserve a table for the neighboring nodes which makes atimeslot reusable after a minimum of 3 hops and preventsmessage collision All nodes keep their slots before thebattery is down or learn of an impending collision All nodescan track their hop distances to a selected gateway node andthe information is communicated in control messages [40]

Advantages

(i) It minimizes the number of transceiver switches(ii) It adapts the nodal sleep interval to data traffic

amount(ii) It makes implementation less complex(iii) It offers improved network lifetime(iv) Energy wastage through idle listening in un-

controlled time slots is avoided via a short time-out interval

Disadvantages

(i) L-MAC is not very good for mobile SNs since slotsare calculated only once

(ii) It suffers idle listening because nodes keep listeningto slotsrsquo control parts so as to get data and enableother nodes to join the network all the time [13]

2117 C-MAC -is was designed for WSNs with 3 seg-ments namely (i) aggressive RTS having double channelcheck to assess the channel (ii) anycast to initialize the flowand (iii) convergent packet forwarding to stabilize the flowIt promises an improvement in energy efficiency lowerlatency and enhanced throughput while avoiding syn-chronization overhead though it operates only in low duty-cycling apps especially where there is less traffic But whenthere is more traffic the protocol employs anycast-basedpacket forwarding for waking the nodes up or finding aforwarder quickly and converges from route-suboptimal toroute-optimal unicast After this energy could be preservedby the nodes using a synchronized wake-up plan [13]

Advantages

(i) It improves energy efficiency(ii) It lowers latency(iii) It enhances throughput avoiding synchronization

overhead

Disadvantages

(i) It operates only in low duty-cycling applicationsespecially where there is less traffic

22 MAC Protocols for WMSNs

221 Diff-MAC -is is a CSMACA-based QoS-awarehybrid-priority-based MAC protocol for WMSNs It targetsenhancing the channel utilization while making use of ef-ficient service differentiation to do coordination for mediumaccess of each traffic class while giving fair quick datatransfer It offers QoS using the following features [14]

(i) It balances energy consumption and delay throughsensor node duty-cycling adaptation depending onthe dominant traffic class

(ii) It has a feature for fragmentation and messagepassing that breaks lengthy video frames intosmaller packets that are transmitted as a burst thusreducing retransmission costs if there are any MACfailures

(iii) -ere is fair data delivery amongst the nodes andtraffic classes respectively resulting from intranodeand intraqueue prioritization reducing unbearableperformance

(iv) -e number of collisions can be reduced and packetlatencies are kept within limits by adjusting the sizeof the congestion window in this protocol as per thetraffic needsAdvantages

(i) Diff-MAC is very suitable for WMSNs since itadapts quickly to changes in networkconditions

(ii) It is a fair protocol(iii) It is scalable

Disadvantages

(i) It is quite difficult to monitor the statistics ofthe network

(ii) Adapting dynamically to the network condi-tions is not an easy task

(iii) It exhibits some degree of latency in packetdelivery

-e implementation of Diff-MAC on Imote2 indicatesthat resource requirements can be met on the sensorhardware Results of extensive simulation runs showed thatDiff-MAC outperforms its competitors

222 Cluster-Based On-Demand Multichannel MAC Pro-tocol for WMSN (COM-MAC) -is is an energy-efficientprotocol in WMSNs with high throughput and reliable datatransmission It achieves energy efficiency by schedulingmultichannel media access to reduce channel contentionand remove collision idle listening and overhearing [41]

Design it has a three-phased operation with the (i)request phase (ii) scheduling phase and (iii) datatransmission phaseRequest phase here all nodes have two protocols eachfor sending requests to the cluster head (CH) ie thecontention-based and contention-free TDMAFDMA

14 Journal of Computer Networks and Communications

protocols -e contention-based protocol allocates achannel per node for sending the request message andfree channels are useable as control channels Forcontention-free protocols a control slot is allocatedwhen a network is deployed Here a slot per channel isallocated to a different node enabling requests to be sentwithout interfering with transmissions from othernodes -e contention-based protocol operates undertwo steps (i) Control channel assignment phase achannel is assigned per node for transmitting a request(REQ) message A channel is assigned to multiplenodes and nodes are evenly distributed to availablechannels for congestion control (ii) Request trans-mission phase nodes with data send a REQmessage viathe assigned control channel to notify the CH whichresponds with an ACK message -e contention-freeTDMAFDMA protocol too has two phases includingthe control slot assignment and request transmissionphases For unreliable channel conditions and lowtraffic load the contention-based protocol is fit for usebecause of less collisions and congestion whereas thecontention-free protocol is preferable in heavy trafficload and reliable channel applicationsScheduling phase from the above phase a schedule isgenerated by the CH for coordination and this enablesnodes to transmit and broadcast their data via controlchannels-e schedule for broadcasting only comprisesdata for nodes assigned to a channel -e schedule isrebroadcast for transmission reliability enhancementbut for energy efficiency the nodersquos transceiver isswitched off when all the schedule is completely de-livered till it is the time for transmitting again-e timeslot and data communication radio channel are allincluded in the schedule generated for each nodeData transmission phase on getting the schedule thesensor node sends its data as per the time slot and thechannel assigned -e time slot is divided into datatransmission andACK sections ACK supports link layererror control Using implicit selective repeat ARQ theCH sends an ACK message for all received packets Ifthere is no ACK the packet is assumed lost and hence aretransmission during the subsequent interval is done-is enhances reliability of transmissions though excessdelay is possible here which is also intolerable for real-time critical apps like video or audio -is gives rise to ahybrid MAC protocol to exploit the unused spectrumduring data transmission sessions [41]Advantages

(i) It achieves energy efficiency by scheduling mul-tichannel media access to reduce channel con-tention and remove collision idle listening andoverhearing

(ii) It maximizes network throughput by dynamicallyallocating time slots and channels for nodes byexecution of a traffic-adaptive QoS-aware sched-uling algorithm depending on needs for QoS andtraffic data information

(iii) It increases transmission reliability by in-corporating spectrum-aware ARQ to avoid spec-trum wastage

223 EQoSA -is protocol is developed for video andimage transfer across WMSNs in a hybrid mode withprovision of QoS Concerning the active sensor nodes andtheir traffic loads it employs dynamic session sizes andchanges bitmap-assisted (BMA) MACrsquos fixed session sizeNodes declare any data available for transmission duringcontention time -is forces the cluster head to make aschedule of how slots have been assigned It communicatesthis schedule to the nodes giving each node a given numberof slots per session and so enabling the busty traffic to beaccommodated [14] -e protocol does not perform wellwith DiffServe mechanisms but does well with schemessupporting busty traffic loads and strong cluster heads arenecessary in WMSNs that can assign and broadcast slots

224 QoS-Supported Energy-Efficient MAC (QE-MAC)Based on the IEEE 80211e standard this protocol ensuresfairness and has low latency and jitter and better efficiencyon energy usage than other QoS-aware MAC protocols forWMSNs It operates under a two-phased routine

-e first phase introduces the innovating prioritymechanism in IEEE 80211e QoS and several data types areassigned priorities Its coordination function is a combi-nation of distributed contention-based channel allocationand centralized polling-based channel access mechanisms

-e second phase preserves the nodesrsquo energy usingdynamic duty cycling -ere is an exchange of RTSCTSpackets to supplement CSMACA by muting nodes near thesenderreceiver during packet duration [42]

Advantages

(i) It ensures fairness(ii) It has low latency and jitter and better efficiency on

energy usage than other QoS-awareMAC protocolsfor WMSNs

225 Black-Burst (BB) Contention A contention-basedMAC scheme for enhancing instantaneous QoS access tocarrier-sense WSNs is proposed in [42] In this schemenodes contend for the channel until it is free Real-timepackets are given higher priority over non-real-time packetsand if a node is transmitting real-time packets it is prior-itized over other nodes [15] Voice and video are some of thereal-time apps in consideration here which periodicallyaccess the common radio channel when transmitting -eseneed constrained end-to-end delay of packets at the MAClayer -is protocol majorly does the following [42]

(i) Prioritization of real-time traffic(ii) Implementation of the round-robin mechanism in

real-time nodes(iii) Bounded access delays among real-time packets

Journal of Computer Networks and Communications 15

(iv) Implementation of real-time communication withvarying needs of bandwidth good for multimediatraffic

Amechanism for dynamically assigning packet prioritiesbased on their deadlines and the traversed hops is proposedin [43] -e ReAllocative Priority (ReAP) mechanism ismeant to provide QoS at the MAC layer to video traffic -isscheme introduces adaptive TXOP by modifying the TXOPdynamically based on the queue length [15] Other protocolsin this category include dynamic duty cycle and adaptivecontention window-based MAC protocol frame sharing(FRASH) MAC protocol and real-time independentchannels (RICH) MAC protocol

Advantages

(i) It can implement real-time communication withvarying needs of bandwidth

23 Challenges and Opportunities(a) One of the challenges in satellite transmission is

cross-talk leading to interference in different fre-quency bands It is necessary to increase the rise andfall times as well as protect sensitive nodes as wecircumvent floating ones if we are to regulate cross-talk as in satellite communication therefore Wefurther need to investigate the possibility of usingdifferential signaling -is constitutes an opportu-nity for further research

(b) Another challenge is that the hidden and exposedterminal problems are still noticeable despite the useof RTSCTS solutions in MAC since collisions arestill possible from transmissions of separate nodesAt the MAC layer ACKs might block efficientchannel usage by prohibiting exposed terminalsfrom reusing the channel and may be eliminatedleaving no proof of sender data packet receptionUsing a directional antenna is one of the proposedtechniques of high-speed WSNs like IEEE 80211-is constitutes an opportunity for further study intothese issues for better solutions though

(c) Another challenge is that since nodes in contention-based MAC protocols are mobile there are manycollisions by the packets and schedule in-consistencies in schedule-based protocols in the two-hop neighborhood information when nodes areentering or leaving [44] Deng et al [45] propose amobility-based clustering (MBC) protocol for mo-bile nodes to enhance the packet delivery rate -isprotocol in comparison with CBR has 25 chancesof decreasing collision-induced packet losses and50 chances in comparison with low-energyadaptive clustering hierarchy (LEACH)

(d) -e process of analyzing and evaluating routingperformance for WMSNs is quite complex becauseof the bursty multimedia traffic Because of the wayencoding scheme frames are structured as well as thechanges in between the scenes that occur naturally

the compressed video frames display some substantialburstiness on several time scales -ere will also befrequent route recalculation that is as well energy in-tensive for nodes frequently capturing and transmittingenergy -e solution here could as well possibly be theuse of the MBC protocol Route recalculation tooshould be dynamic to cater for the mobile nodes whichmight be in and out of the range at different instants oftime Recently the use of smart dumpsters for effectivewaste management in smart cities has also been pro-posed as a possible solution [46]

(e) For WSNs their routing protocols do not look atmultimedia applications which require enormousbandwidth and processing energy and should betransmitted in real time with utmost fidelity Extrachallenges in routing are imposed in WMSNs indeployment of heterogeneous sensors since therecould be audio video and still pictures in addition toscalar data during transmission Heterogeneous datahandled also have different QoS requirements andwith many different business needs different ser-vices with varying requirements give rise to a bigchallenge for the routing design -ere is a need todesign protocols that can support WMSNs such asEQoSA [14] for this purpose

Table 2 summarizes the reviewed MAC protocols forWSNs and the ones designed for suitability in WMSNs

3 QoS Strategies at the Application Layer

We discuss many protocols used to solve different needs ofcommunication between machines

31 ConstrainedApplication Protocol (CoAP) It is a requestresponse protocol designed by the Internet Engineering TaskForce (IETF) for constrained devices such as those havinginsufficient RAM or CPU and WPANs with low-powerconstraints resulting in poor transfer of packets andenormous overhead -e protocol was designed for ma-chine-to-machine (M2M) apps and system automation toreduce bandwidth requirements increase packet transferdecrease the overhead and make work simpler leading tolightweight implementation HTTP commands like GETPOST PUT and DELETE are used in a client-server ar-chitecture Using the publishsubscribe architecture CoAPcan support multiple users and hence yield better perfor-mance via multicasting enabling asynchronous messageexchange -is protocol can support both multicasting andunicasting via UDP and hence is a reliable and simpleprotocol -is is done via the messaging sublayer (for re-liability based on UDP) and the requestresponse sublayer(for communication and interaction) [16]

It uses the following messages

(i) Confirmable message this guarantees reliabilityusing ACK messages sent by the receiver node-ese could be synchronous or asynchronous ifthere is a need for more computational time

16 Journal of Computer Networks and Communications

(ii) Nonconfirmable message here no ACK messagesare sent back

(iii) Acknowledgment (ACK) message this confirmsreceipt of a confirmable message

(iv) Reset message this confirms receipt of an un-processed message in case any critical parts nec-essary for interpreting the message are missing

(v) Piggybacked response the receiver directly re-sponds while getting the ACK message

(vi) Separate response the reply by the receiver is dis-crete from the ACK message

CoAP has no in-built security features It is secured bythe datagram transport layer security (DTLS) that runs ontop of UDP though it was not designed for the IoT -ismakes its suitability debatable since DTLS offers no supportfor multicasting It also needs more packets for handshakingthereby increasing the traffic volume and computationalresources and so shortening the lifespan of mobile devicessince they use batteries [17]

Advantages

(i) It can support multiple users and hence yieldbetter performance via multicasting

Table 2 Comparison of MAC layer protocols used in QoS provisioning

Protocol Features QoS parameter Comments Access Priorityassignment Energy eff Designed for

S-MAC[22] Collision avoidance Energy latency Requires periodic

synchronization CSMA Hybrid Medium WSNs

B-MAC[23] Collision avoidance Energy delay

throughput Flexible interface CSMA Dynamic High WSNs

T-MAC[25] Collision avoidance Energy Complex nonscalable CSMA Dynamic High WSNs

C-MAC[13]

Dynamic channelallocation Energy Multichannel MAC CSMA Dynamic Medium WSNs

CU-MAC[24]

Multichannelmechanism

-roughputlatency High packet delivery ratio TDMA Dynamic mdash WSNs

A-MAC[39]

Per flow serviceguarantees Delay m-ary tree model

dynamic prioritiesTDMACSMA Dynamic Medium WSNs

ER-MAC[30]

Emergencyresponse

Energy delaydelivery ratio

Adapts to traffic andtopology changes

TDMACSMA Hybrid Higher than Z-

MAC WSNs

RI-MAC[31] Collision detection Delay packet

delivery ratioHigh packet delivery ratio

and throughput CSMA Dynamic Medium WSNs

SRI-MAC[33] Collision avoidance Energy delay Good data transfer and

energy consumption TDMA Dynamic High WSNs

H-MAC[34] Slotted sleep time -roughput

latencyData transmission only in

sleep timeCSMAAloha Dynamic Medium WSNs

Z-MAC[36] Collision avoidance Energy delay

throughputVery good channel usage

and error syncCSMATDMA Dynamic Medium WSNs

Q-MAC[37]

Intrainternodescheduling Energy High QoS improved

energy efficiency CSMA Hybrid High WSNs

X-MAC[32]

Overhearingavoidance

Energy delaythroughput

Reduces excessivepreamble by strobing CSMA Adaptive

static Medium WSNs

L-MAC[40] Collision avoidance Network

lifetime energyReduces the number oftransceiver switches TDMA Static Medium WSNs

WiseMAC[38]

No downlinkchannel collision Power delay Significantly low-power

consumptionAlohaCSMA Static Higher than

ZigbeeInfrastructure

WSNsAR-MAC[27] Collision avoidance Energy delay Uses data packets and

control packets TDMA Static Higher thanIEEE 802154 WBANs

BB-MAC[42]

Real-time trafficprioritization Delay Employs round robin in

real-time nodesCSMATDMA Dynamic High WMSNs

COM-MAC [41] Collision avoidance Packet delay

throughputBalances reliability and

retransmissionTDMAFDMA Dynamic High WMSNs

EQoSA[14]

Busty traffic loadaccommodation Energy Provides QoS support for

video and imagesTDMACDMA mdash High WMSNs

QE-MAC[42] Collision avoidance Energy jitter

latencyRuns central polling-like

channel access CSMA Dynamic High WMSNs

R-MAC[26] Collision avoidance Energy delay Reservation based mdash Static High Underwater

SNs

Journal of Computer Networks and Communications 17

(ii) It enables asynchronous message exchange(iii) It is a reliable and simple protocol

32 Message Queue Telemetry Transport (MQTT)Developed by IBM and later improved by OASIS with anintention of decreasing bandwidth MQTT is a simple open-source publishsubscribe protocol analogous to the client-server version It is appropriate for lightweight IoT appli-cations and M2M transmissions in constrained situationslike limited power computation capability and memory orbandwidth -e network bandwidth and the computationalresource usage are decreasing since clients are not updatedmaking publishsubscribe protocols outperform requestresponse protocols via IoT requirements [16] -is protocolcan run over TCPIP and so has the following advantages

(i) Delivering of packets with reliable guarantee(ii) Enabling multicasting(iii) Remote device session establishment(iv) Reduction of transport overhead and protocol ex-

change to minimize network traffic(v) Provision of the notification mechanism when

anomalies occur [47]

MQTTrsquos design targets economical use of bandwidthand battery so it has a major application in FacebookMessenger MQTT ensures QoS and reliability of messagedelivery by providing the following three QoS levels

(1) Fire and forget a message is sent once depending onthe networkrsquos best effort and there is no need forACK (lowest QoS level)

(2) Delivered at least once a message is sent at least onceand an ACK message is needed to avoid duplicates

(3) Delivered exactly once by a four-way handshake itensures a message is sent exactly one time (highestQoS level)-e architecture it has three key components (i)publishersmdashlightweight nodes sending data to thebroker before returning to sleep (ii) broker whichorganizes sensory data in topics and sends them tointerested subscribers and (iii) subscribersmdashIoTapps interested in data sent by sensors through thebroker as shown in Figure 3

MQTT is a simple lightweight protocol that is betterthan CoAP via low and high packet loss probability interms of throughput and exhibits lower latency in lowsampling rates [48] It supports a number of imple-mentations for embedded devices and mobile applicationsand offers web-socket support too Nevertheless it hassome shortcomings

(i) It is only employed in simple data types(ii) It exhibits high latency in high sampling rates(iii) It is unsuitable for real-time apps in such cases(iv) It does not provide automatic discovery andor

sufficient security at the protocol level [49]

33 Secure Message Queue Telemetry Transport (SMQTT)-is is an extension of MQTT that is based on lightweightattribute-based encryption using elliptic curve cryptogra-phy -e protocol has three entities as shown in Figure 4(i) publisher that publishes information under a certaintopic (ii) subscriber that obtains the same topic in-formation via a broker and (iii) public key generator(PKG) or broker that acts as the trusted third party -eprotocol has four phases namely (a) set-up phase (b)encrypt phase (c) publish phase and (d) decrypt phase

(a) Set-up phase publisher and subscriber provide aunique ID with attributes for registration with thePKG -is generates a secret master key set andpublic parameters that it publishes with set U A1A2 An For ciphertext policy-attribute-basedencryption (CP-ABE) the PKG gives a key set todevices with attributes [50]

(b) Encrypt phase based on the access tree the publishercreates an access policy An access tree is sent by thepublisher to the PKG for generating the key-gen-eration policy in the key policy- (KP-) based ABEalgorithm -e publishers know beforehand thetopics and subscribers to access as well as the rulesfor access Subscribers are issued with private keysfor all communication sessions Publishers in theCP-ABE algorithm come up with access trees andrules where the payload is encrypted by the pub-lisher who also issues more info needed for de-cryption and the rules With the help of public keycryptography data are encrypted giving rise to ci-phertext on the basis of KPCP-ABE and the receivergets SUBACK from the broker [51]

(c) Publish phase ciphertext is embedded in theSPublish command as payload and topic name isdefined in the header by the publisher -e brokerreceives the SPublish packet and replies withPUBACK prompting the publisher to respond withthe PUBREL packet -e broker then broadcasts totopic subscribers and deletes the data forwarding thePUBCOMP packet to the sender node

(d) Decrypt phase the encrypted information isdecrypted by the subscriber with the help of a secretkey Considering KP-ABE on verification that itsatisfies the access policy the subscriber invokes thePKG for a corresponding key which it sends afterverifying the request With CP-ABE subscribers usethe private key for decryption after authenticationthereby enabling offline interaction [51]Advantages

(i) -e information being encrypted makes it moresecure

34 MQTT for Sensor Networks (MQTT-SN) -is is a ver-sion of MQTT designed to operate under environments oflow bandwidth high link failures and short message lengthIt is a publishsubscribe (pubsub) protocol designed for

18 Journal of Computer Networks and Communications

UDP-based WSNs to visualize communication for low-power devices [5152] It comprises a publisher device actingas the MQTT-SN client to forward communications to agateway that also changes the MQTT-SN message to theMQTTmessage before delivering it to the broker that sendsit to MQTT-SN subscribers [51]

It has a gateway for changing protocols from MQTT toMQTT-SN (Figure 5) [52] -is protocol supports nodes in

the sleeping mode using an offline keep-alive procedure thatcan be used by battery-operated devices to get into thesleeping mode To support sleeping clients MQTT-SN has anew offline keep-alive procedure with which battery-oper-ated devices can go to the sleeping state in which theirmessages are buffered at the server until they are awake InMQTT-SN clients with no preconfigured gateway addresscan figure out the real address of their operational gateway

Publisher

Subscriber

Subscriber

Broker

Publisher

4G

Figure 3 Architecture of the MQTT protocol

Register by mentioning URI Register by mentioning URI

Access tree

Set-up

Publisher ReceiverPKG(broker)

Key management Key management

Encrypt data using public parametersand random key This is done using

CP-ABE Publish it as

Subscribe (topics)SUBACK

S-Publish (TopicEncrypt (data))S-Publish (TopicEncrypt (data))

Encryptand publish

PUBACK

PUBACK

PUBRELPUBRELPUBCOMP

PUBCOMP

Figure 4 SMQTT protocol

Journal of Computer Networks and Communications 19

within the network through a discovery procedure Simul-taneously we may have several gateways in a single wirelessnetwork that can be cooperative via load sharing

MQTT-SN architecture it is composed of MQTT-SN clients gateways and forwarders (Figure 5)rough the MQTT-SN gateway (GW) the MQTT-SN protocol is used by the MQTT-SN clients toconnect to the MQTT server Integration betweenthe MQTT-SN GW and the MQTT-SN server mightbe possible or not Translating between the MQTT-SN and the MQTT for a stand-alone GW is doneusing the MQTT-SN protocol If a GW is detachedfrom the network it can be accessed by the MQTT-SN clients through a forwarder is just encapsu-lates the received MQTT-SN frames that are on thewireless side before forwarding them to the GW un-altered It then decapsulates the received frames tooresending them to the client as sent by the GW [52]ere are 2 types of gateways (a) transparent gatewayand (b) aggregating gateway as shown in Figure 6

(i) Transparent gateway this launches and sustains adirect MQTT connection between the MQTT-SNclient and the MQTT server which is entirely re-served for end-to-end client-server messaging enumber of servers and that of MQTT-SN clientsconnected to the GW are the same is transparent

GW translates syntax between these protocols and iseasier to implement compared to the aggregatingGW But the MQTT server must support dierentsessions for all connected clients Some MQTTservers are congured to enable few parallel sessions

(ii) Aggregating gateway this maintains a single MQTTconnection to the server on behalf of all clients Allexchanged messages from the MQTT-SN client to theaggregating GW are directed to the GW which sortsthe information to forward to the server e aggre-gating GW decreases the number of MQTT sessionssimultaneously supported by the server in WSNs withmany battery-operated sensors and actuators (SAs)having limited storage and processing capabilities It ishowever more complex to implement compared to atransparent GWe general format of the MQTT-SNprotocol message has two or four message headeroctets and n octets of variable header that is optionaldepending on the message in question MQTT-SN has256 message types (MsgType) with those from0X1Eminus 0XFD 0X19 and 0XFF reserved for futureusage [51]

Advantages

(i) e protocol can operate under environmentsof low bandwidth high link failures and shortmessage length

MQTT-SN gateway

MQTT-SN

MQTT

MQTT-SN

MQTT-SN

MQTT-SN

MQTT-SN client

MQTT-SN clients

MQTT broker

MQTT-SN client

MQTT-SN client

MQTT-SN gateway

MQTT-SN forwarder

Figure 5 MQTT-SN architecture

20 Journal of Computer Networks and Communications

(ii) It is a pubsub protocol designed for UDP-basedWSNs to visualize communication for low-power devices

(iii) It can be optimal for low-cost battery-operatednodes having low processing power and lessstorage

(iv) It supports nodes in the sleeping mode using theobrvbarine keep-alive procedure used by battery-operated devices

35e Extensible Messaging and Presence Protocol (XMPP)e standardization of this protocol by the IETF made itpopular for usage over the Internet especially in instantmessaging and chatting platforms It is however susceptibleto spamming attacks and its failure to have worldwidesupport has made Google too withdraw its support of thestandard [17] e protocol is mainly useable in the client-server architecture but can also be applicable in client-clientXMPP communication [53] Among the application layerprotocols XMPP is the only one that supports both asyn-chronous (publishsubscribe) and synchronous (requestresponse) services while running over TCP but can still runover other protocols

Given the simplicity of its implementation and ease ofusability in XMPP instant messaging (IM) is suitable forXMPP and is implemented using the message stanzaltmessagegt ese stanzas are usable in one-to-one ormultiuser sessions or chatting or in error reporting andsending headlines Attribute type of the stanza denes themessage type e attribute type could be a chat group chatheadline normal or error An instant messaging network inXMPP can support many XMPP clients and servers with agateway congured to enable other IM network users toconnect to other XMPP users e XMPP server mainlycontrols sessions and XML streams to clients and servers inaddition to routing XML stanzas in the streams [53] eTCP ports 5269 and 5222 have been respectively registeredfor the XMPP server-to-server connections and XMPP clientconnections by the Internet Assigned Numbers Authority(IANA) XMPP enables specication of XMPP ExtensionProtocols (XEP) to enhance its usefulness [17]

XMPP security clients connect to the server usingtransport layer security (TLS) and simple authentication andsecurity layer (SASL) protocols which are at the same timeused by the servers for interdomain communications Fordata condentiality and integrity TLS will encrypt thestreams of XML for all sessions To conrm the identity of anyclient that needs to access the server the SASL protocol doesthe authentication of the XML stream For any communi-cating session with a server to start the client needs to resolvethe serverrsquos DNS hostname For protection of the server of theclient from third-party attacks the server maintains the au-thentication details secret ie IP address and access methodand only original connections of the server are required

In server-to-server communications the SASL protocolis employed for authentication and condentiality Server-to-server communications may as well be deactivated from aserver by an administrator depending on the organizationalpolicy To guard against domain-spoong attacks serverdialback can be a good solution since servers support itsecuring more the XML stanzas Nevertheless it cannot beused to authenticate secure or encrypt in-server streamsand the server identication resulting therein is not strongenough to be relied on It further cannot secure against DNSpoisoning attacks and IP session hijacking for remote do-mains [53] High-security domains can use TLS and SASL

Advantages

(i) It is popular for usage over the Internet especially ininstant messaging and chatting platforms

Disadvantages

(i) It is susceptible to spamming attacks(ii) It has failed to have worldwide support making

Google withdraw support of the standard(iii) It is not secure against DNS poisoning attacks and

IP session hijacking for remote domains

36 Advanced Message Queuing Protocol (AMQP) isprotocol can be utilized by a number of platforms formessage exchange since it is open source and written inmany languages It is a publishsubscribe model depending

MQTT-SNclients Broker

MQTT

(a)

MQTT-SNclients Broker

(b)

Figure 6 (a) Transparent gateway and (b) aggregating gateway

Journal of Computer Networks and Communications 21

on a reliable and efficient messaging queue with capacity forinstant messaging or e-mail for different applications es-pecially in IoT environments It differs from others in thesense that message types can be specified and have theirsource traceable and how to do trade-off between perfor-mance security and reliability For reliability of messagetransmission AMQPmakes use of some delivery guaranteeslike at most once at least once and exactly once It can alsohowever utilize the TCP transport layer for guaranteeingreliability -e AMQPrsquos publishsubscribe method has twocomponents namely (i) exchange queue meant for routingof messages to an appropriate order in the queue and (ii)message queue that acts as storage pending their delivery toreceivers [16] In this protocol model the broker can makerouting decisions contrary to othermessaging systems whereapps using the queue have in them embedded the decision-making logic for recipients or senders -e developer mayneed to check each of the affected apps so as to change thelogic for routing and delivery of the messages [54]

Advantages

(i) It is convenient(ii) It has capacity for interoperability between dif-

ferent vendors(iii) It can use open standards to enable connection

between business partners(iv) It can enable innovations based on AMQP foun-

dations [55](v) It can enable supporting of mission-critical ap-

plications in e-commerce(vi) It can guarantee reliable delivery of messages

Disadvantages

(i) It is not very appropriate for real-timeapplications

(ii) It may not offer automatic discovery(iii) Interoperability is not guaranteed(iv) It lacks open-source libraries for constrained

gadgets [49]

37 Data Distribution Service (DDS) A product of theObject Management Group (OMG) DDS is a pubsubprotocol meant for M2M communications with two sub-layers -ese are the data-centric pubsub sublayer re-sponsible for connecting anonymous data publishers tosubscribers and the data-local reconstruction sublayer that isoptional and integrates DDS into the application layer [56]-e DDS architecture connects system components asshown in Figure 7 Publishers and subscribers are discon-nected with respect to (i) time (ii) space (iii) flow andbehavior (iv) platforms and (v) programming languages[57] A DDS data publisher generates topics and publisher-subscriber coupling is represented as topic name data typeschema and publishers and subscribersrsquo QoS attributesDelivering of data is done by the publisher layer -e datawriter and publishers together make a decision about anynecessary alterations for subsequent forwarding to thesubscribers who need to send data to IoT applications Data

readers are responsible for delivering topics to the sub-scribers after reading through them [56]

Some of the major components of the DDS architectureinclude the following [57]

(i) Topic this is a logical path specifying publicationand subscriptionrsquos data type between DataWritersand DataReaders For a successful session to begintopic names types and DataWriters and Data-Readersrsquo QoS should correspond

(ii) Domain this is a logical setting for transmissionthat can be utilized for separating and optimizingnetwork transmission in a setup of similar appli-cations If DDS apps share a domain ID then theycan exchange data among themselves

(iii) DataWriter and DataReader these are terminalobjects for writing and reading messages re-spectively to and from a universal data space

(iv) Participant this is an object that stands for apublisher or subscriber of the DDS app in a do-main acting like a container for other objects

Advantages

(i) It is highly reliable and offers QoS support(ii) It is highly scalable(iii) It is appropriate for real-time apps(iv) It is fault tolerant(v) It has high level of interoperability(vi) It does not need brokers that may act as

bottlenecks(vii) It offers automatic discovery(viii) It has a user-defined data structure for topic(ix) It supports publisher and subscriber decoupling

Disadvantages

(i) It is originally developed for only stand-aloneLANs

(ii) A lot of memory is needed(iii) It has no open-source libraries to support con-

strained devices(iv) It is hard to design and configure compared to

other protocols [49]

38 RESTful Services REpresentational State Transfer (REST)-is is an architectural style based on a set of principles thatdefine and address networked resources It offers web servicesthat enable transmission from the HTTP device to the devicewithin the IoT architecture and enables clients to reach theserver [16]-is protocol utilizes HTTP commands to backupthe requestresponse messaging model HTTP is a popularand secure protocol in the WWW using TLSSSL [16] ARESTful application has the following characteristics

(i) State and functionality are separated into distrib-uted resources

(ii) All resources are distinctively named by the use ofunique HTTP commands (GET POST PUT orDELETE over the Internet)

22 Journal of Computer Networks and Communications

(iii) It is a clientserver stateless layered protocol thatsupports caching REST uses the same methods asHTTP for all requests or responses with POST andGET for creating and retrieving resources PUT forupdating and changing the state of resources andDELETE for removing resourcesAdvantages of REST over arbitrary web services(SOAP) in the IoT context

(i) It reduces overhead(ii) It reduces parsing complexity(iii) It is stateless(iii) It integrates more tightly with HTTP [58](iv) It is easy to implement and learnis together

with its natural t over HTTP makes it apreferred method for Web 20 applications toshowcase their data and secure M2M sessionsin the IoT

(v) Applications that support RESTful web ser-vices exhibit better performance than the re-source-constrained WSN nodes [59]

(vi) REST is important for the IoT since it supportscommercial M2M cloud platforms

(vii) Its implementation in smart phones and tabletapps is also easy since it needs only an HTTPlibrary that is accessible to all operating systemplatforms [56]

Disadvantages

(i) Although it enjoys wide usage in commercialM2M platforms it is not likely to dominatebecause it is hard to implement

(ii) Its HTTP usage makes it incompatible withconstrained M2M devices leaving its usage toend applications

39 WebSockets It is a web-based protocol operating on asingle TCP channel and providing full-duplex transmissions

It permits secure duplex communication between a clientexecuting unreliable code in a restricted environment and adistant node communicating from that code while using theorigin-based security model employed by web browsers overTCP [60] e protocol is not a requestresponse model andis not a publishsubscribe model So unlike other protocolsa web-socket session is established by the client initiating ahandshake to the server to launch the session is onceestablished a full-duplex client-to-server connection beginsasynchronously and runs until terminated by either party[56] It gives a model for browser-based apps utilizing bi-directional communication with no need of many HTTPconnections to servers

WebSocket comprises two parts ie (i) the handshakehaving a clientrsquos message and the serverrsquos handshake responseand (ii) the data transfer HTTP headers should be repeated byapplications in all client requests and server responses arisingfrom the endless polling which might increase the commu-nication overhead subject to the application [61] WebSocketcan be used to come up with scalable real-time web appli-cations since it can enable full-duplex transmission via onesocket across the Web Its security can be ensured with thehelp of TLSSSL is protocol is suitable for resource-con-strained devices and IoTapplications will not have a good userexperience with it since it operates a clientserver architectureNevertheless it can outcompete all TCP-based protocolsbecause of the following advantages

(i) It is t for real-time transmission(ii) It has improved security(iii) It minimizes transmission overhead(iv) It can provide ecient messaging systems using

WebSocket ApplicationMessaging Protocol (WAMP)[17]

310 Streaming (Simple) Text-Oriented Messaging Protocol(STOMP) is is a TCP-based protocol designed for textmessage-oriented middleware (MOM) with the use of

Topic

Datareader

Datareader

Datareader

Datawriter

Datawriter

Datawriter

Subscriber PublisherPublisher

Data bus (DDS domain)

Subscriber

Topic Topic

Figure 7 DDS architecture

Journal of Computer Networks and Communications 23

HTTP-like commands for interoperability among platformslanguages and brokers Data communication takes placebetween the client and the broker in line with commands(like CONNECT DISCONNECT ACK NACK SUB-SCRIBE UNSUBSCRIBE SEND BEGIN COMMIT orABORT) then headers in the form ltkeygt ltvaluegt (one perline) then a blank line and then body content terminatingin a null character Server-client communication takes placeover a MESSAGE RECEIPT or ERROR frame with headerand body content format being analogous MOM productssupporting STOMP include Apache ActiveMQ or FuseMessage Broker HornetQ Open Message Queue(OpenMQ) RabbitMQ and syslog-ng through its STOMPdestination plugin [62] In addition to these other appli-cation layer protocols include simple media control protocol(SMCP) lightweight local automation protocol (LLAP)simple sensor interface (SSI) lightweight M2M (LWM2M)XMPP-IOT and simple object access protocol (SOAP)

Advantages

(i) It is lightweight for constrained networks(ii) It has flexibility to choose quality of services with

the given functionality(iii) It is standardized by OASIS Technical Committee(iv) It is easy and quick to implement

Challenges and opportunities

Here security threats are imminent because of the open-endedness of the application layer [63] -ese could includethe following

(i) -ere is poor or lack of a security design for anapplicationrsquos function Coming up with a gooddesign to address this could be a good ideaopportunity

(ii) Some of the programs might have backdoorsevading otherwise secure controls to allow un-authorized access to network resources We requiremechanisms to patch up such loopholes

(iii) Applications that lack or have weak authenticationmechanisms can be easily targeted by unlicensedusers to abuse the network Such mechanisms needto be strengthened further

(iv) Excessively complicated access control rules maybecause of their misunderstanding or poor writinggive unauthorized access allowing dubious activityrather than protecting the network they purport toprotect -ese should be simple and clear to un-derstand by all users

(v) In booting workstations without disks and managingnetwork devices we usually use the TFTP protocolalthough access does not require authenticating theuser by usernamepassword -is makes an intruderable to gain easy access to configaccess info as longas he can guess the filenames -is needs to be fixed

A few steps can be used to secure the application layerTo protect network data transmitted across the network

applications need to strengthen their encryption and au-thentication mechanisms Apps too need to enable strictcontrols of privileges for data access using good mechanismsto balance between usability and effectiveness [63] Appli-cations handling sensitive data need detailed logging andaudit capability as well as testing and reviewing

Table 3 summarizes the application layer protocols dis-cussed with their weaknesses and differences among them

4 Cross-Layer QoS Strategies

41 Service-Differentiated Real-Time Communication Scheme(SDRCS) -is is an event-driven routing protocol that routesreal-time traffic by featuring a cross-layer packet-forwardingdesign in which it integrates the real-time routing functionalitywith a new priority-based MAC scheme -e protocol ap-proximates distributed packet traversal speed for traffic clas-sification and admission control on the basis of this design Itfurther localizes decision-making by prioritizing packet for-warding to maximize the speed of packets [64]

Advantages

(i) It improves bandwidth utilization bandwidthdegradation due to unschedulable packets can beavoided

(ii) It offers a lightweight packet schedulability estima-tion mechanism Here it uses received signalstrength and admission control and then earlymissed-deadline packet-dropping policies are made

(iii) It delivers high end-to-end throughput via highersource data rates alongside strict end-to-end latencyneeds

(iv) Nodes which meet real-time requirements of lesstraffic load and better channel quality are givenhighest priority when it comes to packet forwarding

-e SDRCS operates via 5 mechanisms namely (1) RSS-(Rich Site SummaryReally Simple Syndication-) basedgrouping (2) admission control (3) prioritized queuing(4) real-time MAC and (5) dynamic forwarding -isprotocol has a queuing policy that employs per-hopdeadline-based queues and the nodes schedule theirpackets using FIFO priority-based queues [1] -e prin-cipal features of this protocol are mentioned as follows

(i) It is event driven and is easily adaptable to anynetwork changes

(ii) It does not need additional hardware for locali-zation or multichannel transmission

(iii) -e cross-layer packet-forwarding strategy can beused when transmitting multimedia traffic

(iv) It is able to efficiently circumvent any voids therein

Early deadline miss (EDM) policy drops all unscheduledpackets -e SDRCS circumvents wrong packet dropsdue to EDM since it adjusts to network changes easily

42 Sensor Fuzzy-Based Image Transport (SUIT) -is pro-tocol regulates congestion using a scheme based on fuzzy

24 Journal of Computer Networks and Communications

logic It transmits packets with reduced quality to allowablelevels if there is congestion enabling it to transmit lower-quality packetsframes without being dropped-e numberof packets delivered per second thus rises and while thisenhances video streaming apps it is more appropriatelymade for those that quickly transmit non-real-time videosuch as in video surveillance SUIT can stream JPEG pic-tures other than the traditional predictive video codingmethods that are not suitable for sensor devices because oftheir constraints of energy memory and computationalspeed Progressive JPEG-PJPEG performs better via packettransmission rate and delay though it degrades the imagequality because of congestion [65] -is protocol employscross-layer methods for interlayer data exchange though itstill employs some layering making it still possible topreserve all advantages of the layered design since it hasapplication and transport layers but has noMAC or routinglayer

For congestion detection and control SUIT bases on thepercentage number of in-bound and out-bound frames perwindow number of contenders and buffer occupancy of thenext-hop node Since buffer management has an effect oneffectiveness and QoS inWMSNs it is useful in the transportlayer so the protocol prioritizes packets for better QoS SUITreorders packets by tracking the sequence numbers of thedelivered packets since they all have a source ID framenumber and sequence number that it uses to detect anymissing packets -e protocol however does not providepacket reliability [65]

Advantages

(i) -e protocol prioritizes packets for better QoS(ii) It can transmit lower-quality packetsframes

without being dropped

Disadvantages

(i) It does not provide packet reliability

43 Network Layer QoS Support Enforced by a Cross-LayerController (NLQS) -is scheme permits packet-level servicedifferentiation as a function of throughput end-to-endpacket error rate and delay [66] (see Figure 8) -is en-hances network layer QoS It comprises a cross-layer controlunit (XLCU) for configuring and controlling networkingfunctionalities at physical MAC and network layers [18]-is depends upon the unified logic which affects choices forapplication layer needs and status of functional blocks whichdo implementation of networking functions [66]

Advantages

(i) Cross-layer interactions can be controlled withoutweakening the upgradability modularity andsimplicity of designing the system

44 Cross-Layer Signaling Shortcuts (CLASS) CLASS has ahigh level of effectiveness flexibility and comprehensive-ness According to [67] it has the following features

(i) Direct signaling among remote layers leading tofaster transmission

(ii) Lightweighted internal message format standard-ized protocols are not suitable for use in internalsignaling since they are not usually lightweight forinstance transmitting against faults in the networkCLASS requires only three fields namely destina-tion address event type and event contents

(iii) -e setup of its exterior messages is standardconsidering external signaling ICMP and TCPIPare suitable for general messages and short notifi-cations respectively

(iv) Message control protocol should ensure optimizedand organized exchange of dense simultaneousmessages across layers for high efficiency andavoidance of possible conflicts -e protocol appliesto many cross-layer signaling situations

Table 3 Differences and problems with application layer protocols

Protocol Architecture Transport QoS options Security Weaknesses

CoAP Requestresponse UDP DTLS No in-built security features DTLS does not supportmulticast

MQTT Publishsubscribe TCP TLSSSL Insufficient security at the protocol level sincepayload values are not encrypted

SMQTT Publishsubscribe TCP CPKP-ABE Key revocation and group pubsub for distributedSMQTT are still a challenge

MQTT-SN Publishsubscribe UDP TLSSSL SSLTLS suffers from attacks like BEAST CRIMERC4 and Heartbleed

XMPP RequestresponsePublishsubscribe TCP X TLSSSL Susceptible to spamming attacks and lacks worldwide

supportAMQP Publishsubscribe TCP TLSSSL Inappropriate for real-time applications

DDS Publishsubscribe TCPUDP TLSSSL Memory intensive and no open-source libraries forconstrained devices

REST Requestresponse HTTP X HTTPS Hard to implement Uses HTTP so incompatiblewith constrained apps

WebSockets PublishsubscribeClientserver TCP X TLSSSL Bad user experience for IoT apps since it runs the

clientserver architectureSTOMP Clientserver TCP HTTP -e broker can act as a bottleneck

Journal of Computer Networks and Communications 25

Advantages

(i) e protocol is scalable and has a very lowpropagation delay

(ii) It is very ecient and exible since it uses directsignaling between any two layers

(iii) It is very comprehensive and eective

45 Cross-Layer and Multipath-Based Video Transmission(CMVT) CMVT is a hybrid of application and networklayers At the application layer it carries out the encoding ofvideo streams to video data frames (I- P- and B-frames) bymeans of the MPEG-4 encoding format At the networklayer route discovery and data transfer take place Twoalgorithms greedy forwarding and rollback are used todiscover a number of paths from the source to the sink nodevia the route discovery method [20] A specied node icalculates the evaluation of its adjacent node j using thefollowing equation

fij (1 minus α)d2(j D) minus d2min(i)d2max(i) minus d

2min(i)

+ αeinit(j) minus eres(j)

einit(j)

(19)

where fij evaluation value for the node i to j d2(j D) isthe distance from the node j to the destination node D

d2min(i) and d2max(i) are the minimum and maximum dis-

tances of neighbors of the node i to D respectively einit(j) isthe initial energy of the node j eres(j) is the current residualenergy of the node j and α is the energy coecient given asfollows

α emax(i) minus emin(i)

emax(i) (20)

where emax(i) and emin(i) are respectively the maximumand minimum energy left of all node irsquos neighbors

e network layer further transmits video streams withCMVT doing status evaluation choosing an appropriatecommunication path for any packets and the QoS guaranteelevel for the path i is calculated as

fi (1 minus ω)hisum hi

+ ωnisum ni

(21)

where fi is the evaluation value for the path i hi are the hopsfor the path i ni is the summation of packets sent via isum ni isthe summation of packets sent by sources and ω is theenergy consumption factor [18]

Advantages

(i) e CMVT protocol is very superior in mediatransmission particularly large-scale WMSNs

To network peers

Flow requirements

-

Atimes

Admissioncontrol

Capacityassignment

Next-hopselection

Coding rate

Data rate

QoScontracts

Cross-layer controlunit (XLCU)

Localization subsystem

Synchronizationsubsystem

Networkcoordinationunit (NCU)

UWB modulator

Channel coder

QoS scheduler

Georouting

Source coder

Rate control

Figure 8 Cross-layer controller architecture

26 Journal of Computer Networks and Communications

46 Cross-Layer Cooperative MAC (CoopMAC)CoopMAC [68] is based on the IEEE 80211 DCF approach-ere is a reduction in interference of neighboring cellsleading to uniform coverage in densely deployed networks[68] A connection is established by two-way authenticationbetween the sender and the receiver When a Clear-to-Send(CTS) packet and short interframe space (SIFS) are obtainedby the sender packets are transmitted right to the receiverwith no valuable cooperation If available the two nodes findout whether they can exchange data -is is done by the useof a helper identification signal to establish how feasiblecooperative communication is It is established in thepresence of a signal and if it is not available direct com-munication is activated For distributed systems helper-initiated cooperation is preferable because of the presence ofthe RTSCTS packets [69]

Advantages

(i) CoopMAC has an advantage over others due to thespatial diversity amongst the three nodes [70]

(ii) It has the potential to achieve significantthroughput and delay with no complexity in thedesign of the system

47 Cooperative MAC Protocol for Multihop Networks (M-CMAC) Just as CoopMAC M-CMAC is made in such away that high data rate stations help low data rate stations toforward traffic for broadcasting [71] -ere are also helperschosen in such a way that 2 fast-hop transmissions replace aslow hop transmission -e helper via whom there existsleast delay from the sender to the receiver is considered bestbecause of theminimum two-hop transmission rate and so isthe chosen neighbor to the duo Here the working as-sumption is that all nodes have their position coordinatesknown and thus the Euclidean distances between all pairscan be calculated and converted to the data rate for that linkAll nodes have cooperative tables (CT) of potential helperswith destination and helper MAC address Euclidean dis-tance and total distance through the helper Any sender withdata to transmit checks out for the existence of any helper inthe CT for the destination and if available forwards RTS forchannel reservation for single-hop duration M-CMAC hasgot an RTS format having a shape with five fields similar tothe one given as follows

Framecontrol Duration Source

addressDestination

addressHelperaddress

-e destination nodersquos address is kept by the source in thepart for the helper address that is in turn saved as the desti-nation address Nodes inspect helper and destination addressfields on receipt of RTS whereby the node acts as the othernodersquos helper in case the helper address field is not the sameand the helper transmits CTS back to the source if at all it wantsto forward data On delivery of the CTS packet the sourcetransmits to the helper which in turn sends the packets to thedestination that also sends ACK to the helper on receipt of thepacket In M-CMAC there exists a higher level of channelreuse (parallel transmissions) due to the increased number of

nodes therein that subsequently raises availability of helpers fordata forwarding -is results in increased throughput incomparison with CoopMAC and IEEE 80211 DCF [71]

Advantages

(i) It promises a higher throughput compared toCoopMAC and IEEE 80211 DCF

48 Cluster-Based Cooperative Routing (CBCR) Protocol-e CBCR protocol has a multihop data-forwarding func-tion realized at the link layer with cooperative links that useM-CMAC -is protocol encompasses two stages namelythe routing relay selection phase and data forwarding phase

(i) Routing relay selection phase all nodes announcetheir presence to their neighbors by broadcastingperiodical beacon messages that carry the MAC ad-dress of the node Each of these constructs a relaytable that contains all neighboring nodes with whichit is able to communicate -e node further broad-casts its neighbor list in case of any change to itsentries since the previous broadcast -e MAC ad-dresses of nodes next to the node X are contained inits relay tablersquos column one and the neighbor nodersquosrow has MAC addresses for neighbors of the adjacentnode Based on its relay table each node in-dependently chooses routing relays Choosing a nodeas a relay node depends on the number of nodes itconnectsmdashit should be the highest number [71]

(ii) Data forwarding phase a node with packets totransmit needs to first verify if the receiver is in thesame cluster and if that receiver has a helper to whichthe packet is thence forwarded but in whose absencethe packet is delivered right to the receiver In casethe intended receiver is in a different cluster therelay table will be checked to see whether that re-ceiver is reachable via other routing relays If it isreachable packets will be sent to the routing relaydirectly or via the helper if at all the relay has one Incase of destination unreachable via relays packetsare multicast to all relays by the node [71]Advantages

(i) It enables multicasting

49 MAC-PHY Cross-Layer Protocol A cooperative cross-layer standard for cooperation at the physical layer in next-generation WMSNs is developed in [70] It provides acomplete MAC layer algorithm that gives a supportive shellfor the PHY-MAC layer Like CoopMAC its scheme de-pends on the middle node for communication betweennodes-eMAC layer protocol has been changed to regulateinformation exchanges at the physical layer -e receivernode receives duplicate packets from source and helpernodes to decode the data [70]

410 MAC-Centric Approach -is cross-layer protocoltargets multimedia applications by using the MPEG-4

Journal of Computer Networks and Communications 27

scheme [21] It is characterized by 4 access categories(AC3ndashAC0) according to their priority in transmission It ismeant for supporting different QoS needs in upcomingvideo applications and enabling MAC layer differentiationfor H264 partitioning [18] Because of low bandwidth delayand other QoS challenges that cause inefficiency in trans-mittingmultimedia data inWMSNs a number of algorithmsdepending on IEEE 80211e have been proposed to supporttransmission of quality videos [72] Selecting an AC dependson the measures for QoS such as the loss rate and delay-us AC3 having the highest priority is mapped to theparameter set concept since the stream is sensitive to bit lossespecially in video transmission [21]

411 Adaptive Cross-Layer Forward Error Correction(ACFEC) In the ACFEC [73] model data packets are ex-changed among nodes through the access point (AP) whichoperates the infrastructure mode in which it adds FEC to videodata [72] -ese data are dealt by encapsulation using astreaming server to the receiver through thewireless AP as RTPpackets An adaptive FEC controller senses the category ofpackets out of the RTP header and recovers the header ofpackets from UDP -e encoder generates some error-cor-recting packets whose number is determined by the sourcepacket number of the block Multimedia transmissions aremonitored by the controller using MAC failure data and itscounter is increased by one if there is a failed transmission-econtroller uses the failure counter to change the packet numberproduced after transmitting a block [18] In case there are lostpackets it changes the redundancy rates and produces extrapackets to replace the lost ones and satisfy the requirements ofthe receiving node [72] -e FEC packet number is enhancedor reduced to satisfy the receiverrsquos requirements and stoppacket losses-is is done through accurate detection of packetlosses and adjustment of redundancy rates If all video datapackets are well received there will not be generation of FECpackets [73]

Advantages

(i) It promises a good QoS through packet loss re-duction and redundancy rate adjustment

412 Balanced Cross-Layer Fuzzy Logic (BCFL) DesignRouting Algorithm in WSNs A new fuzzy logic-basedrouting algorithm (BCFL) [74] was designed using disper-sion of the cross-layer parameter as the fuzzy logic inferencesystem input Each cross-layer parameter has a dynamicweight depending on the value of dispersion -e designcomes with some innovations as follows (i) for fuzzy logicinference system input the absolute parameter value issubstituted by parameter dispersion thereby significantlyreducing algorithmic complexity (ii) dispersion does notchange with the order of magnitude according to the dis-persion formula and (iii) the weight of the parameter de-pends on the size of its dispersion and the two are inverselyproportional to each other -is enables BCFL to have somenobility unlike other algorithms as per the following dis-tinguishing properties making it advantageous

(i) It has got simple if-then rules which remain con-stant even when the constraints increase

(ii) It is capable of dealing with many constraints withno increment in complexity

(iii) It has capacity to yield a more balanced solutionthan other algorithms

(iv) It is easily adaptable to changes in network con-ditions and topology even when the changes arefrequent like in underwater WSNs -e algorithm isuseable in choosing the CH in cluster-based routingprotocols [74]

413 Minimum Hop Disjoint Multipath Routing Algorithmwith Time Slice Load-Balancing Congestion Control SchemeMHDMwTS is a two-phased routing protocol comprisingpath build-up and path acknowledgment phases It consists ofmultiple sources each of which has three build-up disjointpaths namely primary alternate and back-up paths Onactivation the source node requests to build up a route to thenearest hop neighbor which is the path build-up phase [19] Inthis phase step one has source activation in which therequested node for path building adds its number and time-stamp and sends to the least hop-count neighboring node-isgoes on until the least time latency sink having the informationneeded to construct the primary route is reached [75] In steptwo there is path extraction when the new package fromanother path arrives whereby the extracted path is comparedwith the primary path If the node is shared the package will berejected or an alternative path will be searched for to get abackup through comparison of the preceding two paths Phasetwo is path acknowledgment where we have the third step inwhich the sink returns the ACK packet to the sender with pathinformation having nodes and their time information after ithas computed it using the timestamp [1819]

Advantages

(i) -e protocol minimizes end-to-end latency(ii) It enhances congestion control

414 Cross-Layer Optimal Design (CLOD) Authors in [76]come up with a CLOD for scheduling at the data link layerrouting at the network layer and controlling congestion atthe transport layer with an assumption of fixed link capacity-rough congestion control energy efficiency is improvedTransport layer congestion at the nodes is minimized bycompressed sensing (CS) in which transmitted bits are re-duced whereas optimally allocating resources decreasescongestion on the links at the data link layer CLODpromises minimized computational complexity and betterperformance in light traffic scenarios [77] Generating andstoring CLOD is more efficient compared to Gaussianrandom matrices It prolongs network lifetime and savesenergy

Advantages

(i) It does congestion control subsequently increasingenergy efficiency

28 Journal of Computer Networks and Communications

(ii) It reduces computational complexity(iii) It improves performance in light traffic scenarios

415 Challenges and Opportunities of the Cross-Layer Design(CLD)

(i) Different layers in a real network assume differentfunctions andor services A layer only communi-cates with its neighboring layers -e layered modeldegrades system performance because of many

features in wireless transmission QoS parameters inthe layers of WMSNs are shared by routing pro-tocols to optimize the performance Neverthelessthere is a need to design cross-layer models forincreased efficiency in routing [1]

(ii) -e physical layer plays a highly vital role in CLDRate adaptation and channel allocation take place atthe physical layer via signal processing to enhanceQoS End-to-end performance in wireless media isaffected by the changes therein given the effect on

Table 4 A summary of some of the reviewed cross-layer models

Protocol Layers Aims Comments QoS parameters

SDRCS MACPHY Routing real-time traffic Transmits multimedia traffic viacross-layer packet forwarding

-roughputlatency

SUIT Applicationtransport Transmission of non-real-time video Uses fuzzy logic to regulate

congestionTransmission rate

delay

CLASS Any twoDesign serves as a framework fordifferent implementations ofdifferent application scenarios

Scalable very efficient flexible andhas very low propagation delay

Propagation delayjitter

NLQS PHYMACnetwork

Permits packet-level servicedifferentiation as a function of

throughput packet error rate anddelay

Network layer QoS is enhanced -roughput delaypacket error rate

CMVT Applicationnetwork

Encodes video streams to video dataframes via MPEG-4 encoding anddoes route discovery and data

transfer

Greedy forwarding and rollback areused to find source-to-sink paths Energy

CoopMAC PHYMAC Offers spatial diversity among thethree nodes

Helper ID signal finds out howfeasible cooperative communication

is

-roughput ratedelay

M-CMAC PHYMAC Increases end-to-end throughputand packet delivery ratio

Euclidean distances between nodesare calculated and converted to the

data rate for a link

-roughputpacket delivery

ratio

CBCR PHYMACMinimizes control overhead andtime consumed in establishing thecooperative paths than M-CMAC

Energy consumption is moreuniformly distributed in a network

enhancing network lifetime

-roughputpacket deliveryratio energy

MAC-PHY MACPHYMeant to enable PHY layer

cooperation and maximize gains ofcooperation at the MAC layer

Leverages both spatial diversity andcoding gain -roughput delay

MAC-centric MACAPPL

Meant to support QoS needs in newvideo apps and enable MAC layer

differentiation for H264partitioning

Targets multimedia applications byusing the MPEG-4 scheme

Delay packet lossrate

ACFEC MACnetwork(UDP)

Meant to enhance the quality ofvideo streaming over 80211WLANs

and overcome packet losses

Adjusts redundancy rates toovercome channel fluctuations and

detect and reduce packet lossPacket loss rate

MHDMwTS mdash

Meant to provide reliable datatransfer with the multipath routingand load-balancing congestioncontrol method in WMSNs

More reliable than basic routingschemes for transport multimedia

data

Latency packagetransmit rate

BCFL mdash

Introduces dispersion into fuzzylogic-based routing and sets every

cross-layer parameter with adynamic weight

Can be used to select a CH in cluster-based routing protocols andproposes a dispersion formula

Node utilitydispersion

CLODDatalinknetworktransport

Prolongs network lifetime andachieves congestion control anddesigned for lightly loaded WSNs

Assumes fixed link capacity andintegrates compressed sensing

technology

-roughputaverage energy CS

error ratio

Journal of Computer Networks and Communications 29

the purposefulness of the protocols at the networklayer CLD offers solutions to conserving powerminimizing energy and controlling both the flowand congestion in the network as well as fault tol-erance so it is an opportunity for designers con-sidering other layers We further desire to developCLDs that shall make an impact on network oper-ation and get close attention [18]

(iii) CLD interface standardization the architecturemust deliver module functionality although thereexist some queries amongst modules about possibleinterfaces predicted by the necessity to share in-formation among remote protocol layers Technicalchallenges include developing designing andstandardizing of cross-layer interfaces and algo-rithms that satisfy the cross-layer optimization re-quirements among protocol layers

Table 4 summarizes these cross-layer models with a fewcomments

5 Future Research Direction

51 At theMACLayer In WMSNs achieving duty cycling istricky since video traffic is volatile Listening to adjacentnodes wastes energy We need to switch states from awake tosleep -is requires further research

Self-organizing networks and green communicationalthough most CRWSN energy-saving algorithms aredesigned to reduce node transmission power energyconsumption is inevitable during the operational mode ofBSs because of internal processing Scholars are trying withpolicy-makers to encourage a shift to greenWSNs to reduceoperational costs and carbon footprints With self-orga-nizing CRWSNs for automatic switch-off when idle energyis saved which calls for cognitive switching algorithms Sointegrating cognitive switching with radio resource allo-cation in CRWSNs is a promising research area [78]

52 At the Application Layer Considering a WMSN ar-chitecture end users access the network via the BS to andfrom which they send their sensed data We desire to getguaranteed QoS and energy efficiency by designing a newarchitecture in the form of hardware and software whichneeds to be investigated further for a fix of the same [1]

QoS and energy efficiency are very crucial in WMSNsand mostly real-time apps that require guaranteed band-width and throughput in their network lifetime Mostprotocols ignore base station (BS) mobility and WMSNnodes Traffic management telemedicine and battlefieldsurveillance apps require mobile nodes or BSs so designingdynamic routing protocols to be adaptable in these cir-cumstances is necessary

53 Cross-Layer Security of data transmitted on WMSNseg in military surveillance and e-commerce is key withQoS and energy efficiency WMSNs are vulnerable to attacks

like worms and sinkhole Moreover their computationpower is not that high complicating implementation ofstrong secure protocols -is needs further attention

For fairness and priority issues in CRWSNs nodes areassigned diverse priorities depending on importance andurgency of their data For better network performance thereis a need to develop fair resource allocation mechanismswith priority-based fairness in sensors-ere is still potentialin this research area [78]

We must develop multiobjective adaptive protocols tooptimize QoS metrics involved in routing for best trans-mission results to get a trade-off between different opti-mization metrics We need to design standardized cross-layer algorithms which are able to meet cross-layer opti-mization standards

6 Conclusion

We have reviewed the different QoS strategies for WSNs inthe context of IoT from the MAC layer and application layeras well as the cross-layer paradigm For the MAC layer wehave reviewed protocols for WSNs such as the hybrid of thecontention-free and contention-based MAC protocols andthose for WMSNs We further reviewed a number of ap-plication layer protocols including many machine-to-ma-chine requestresponse and publishsubscribe protocols Forsystem optimization cross-layer QoS strategies are veryimportant in wireless communication We have reviewed anumber of cross-layer strategies For all categories reviewedchallenges and opportunities are discussed Finally somepossible future directions are discussed for research andapplication showing a promising potential for future re-search in this relatively new area especially as more WMSNapplications emerge recently

Conflicts of Interest

-e authors hereby declare no conflicts of interest regardingthe publication of this work

Acknowledgments

-is research was funded by the National Natural ScienceFoundation of China (Nos 61602398 61672447 and61711540306)

References

[1] V Bhandary A Malik and S Kumar ldquoRouting in wirelessmultimedia sensor networks a survey of existing protocolsand open research issuesrdquo Journal of Engineering vol 2016Article ID 9608757 27 pages 2016

[2] I F Akyildiz T Melodia and K R Chowdhury ldquoWirelessmultimedia sensor networks applications and testbedsrdquoProceedings of the IEEE vol 96 no 10 pp 1588ndash1605 2008

[3] M Iqbal M Naeem A Ahmed M Awais A Anpalagan andA Ahmad ldquoSwarm intelligence based resource managementfor cooperative cognitive radio network in smart hospitalsrdquoWireless Personal Communications vol 98 no 1 pp 571ndash5922018

30 Journal of Computer Networks and Communications

[4] S A Alvi B Afzal G A Shah L Atzori and W MahmoodldquoInternet of multimedia things vision and challengesrdquo AdHoc Networks vol 33 pp 87ndash111 2015

[5] Q F Hassan A U R Khan and S A Madani Internet ofings Challenges Advances and Applications CRC PressTaylor amp Francis Group A Chapman amp Hall Group BocaRaton FL USA 2018

[6] Q Wang Y Zhao W Wang et al ldquoMultimedia IoT systemsand applicationsrdquo in Proceedings of the 2017 Global Internet ofings Summit (GIoTS) vol 2 Geneva Switzerland June 2017

[7] A Aslam and E Curry ldquoTowards a generalized approach fordeep neural network based event processing for the internet ofmultimedia thingsrdquo IEEE Access vol 6 pp 25573ndash25587 2018

[8] M Jridi T Chapel V Dorez L Bougeant and A Le BotlanldquoSoC-based edge computing gateway in the context of theinternet of multimedia things experimental platformrdquoJournal of Low Power Electronics and Applications vol 8no 1 p 1 2018

[9] H Noura A Chehab L Sleem andM N Rapha ldquoOne roundcipher algorithm for multimedia IoT devicesrdquo MultimediaTools and Applications vol 77 no 14 pp 18383ndash18413 2018

[10] T Balan D Robu and F Sandu ldquoMultihoming for mobileinternet of multimedia thingsrdquo Mobile Information Systemsvol 2017 Article ID 6965028 16 pages 2017

[11] H Shen and G Bai ldquoRouting in wireless multimedia sensornetworks a survey and challenges aheadrdquo Journal of Networkand Computer Applications vol 71 pp 30ndash49 2016

[12] A Ahmad M H Rehmani H Tembine O A Mohammedand A Jamalipour ldquoIEEE access special section Editorial optimization for emerging wireless Networks IoT 5G andsmart grid communication networksrdquo IEEE Access vol 5pp 2096ndash2100 2017

[13] B A Muzakkari M A Mohamed M F A Kadir andZ Mohamad ldquoRecent advances in energy efficient-QoS awareMAC protocols for wireless sensor networksrdquo InternationalJournal of Advanced Computer Research vol 8 no 38pp 212ndash228 2018

[14] M A Yigitel O D Incel and C Ersoy ldquoQoS-aware MACprotocols for wireless sensor networks a surveyrdquo ComputerNetworks vol 55 no 8 pp 1982ndash2004 2011

[15] A M Abbas and O Kure ldquoQuality of Service in mobile ad hocnetworks a surveyrdquo International Journal of Ad Hoc andUbiquitous Computing vol 6 no 2 p 75 2010

[16] M Q Shatnawi ldquoApplication layer protocols for the internetof things a surveyrdquo in Proceedings of the InternationalConference on Engineering amp MIS (ICEMIS) Agadir Mo-rocco September 2016

[17] V Karagiannis P Chatzimisios F Vazquez-gallego andJ Alonso-zarate ldquoA survey on application layer protocols forthe internet of things research motivationrdquo InternationalJournal of Advanced Research in Computer Science vol 8no 3 pp 1ndash10 2015

[18] M S Bernard T Pei Z Li and K Li ldquoQoS strategies forwireless multimedia sensor networks in the context of IoTrdquoLecture Notes of the Institute for Computer Sciences SocialInformatics and Telecommunications Engineering vol 4pp 228ndash253 2019

[19] A AlAmri and M Abdullah ldquoCross-layer quality of serviceprotocols for wireless multimedia sensor networksrdquo in Pro-ceedings of the International Conference on CommunicationManagement and Information Technology pp 649ndash658Cosenza Italy April 2017

[20] J Guo L Sun and R Wang ldquoA cross-layer and multipathbased video transmission scheme for wireless multimedia

sensor networksrdquo Journal of Networks vol 7 no 9pp 1334ndash1340 2012

[21] A Ksentini A Naimi and M Gueroui ldquoToward an im-provement of H264 video transmission over IEEE 80211ethrough a cross-layer architecturerdquo IEEE CommunicationsMagazine vol 44 no 1 pp 107ndash114 2006

[22] W Ye J Heidemann and D Estrin ldquoAn energy-efficientMAC protocol for wireless sensor networksrdquo in Proceedings ofthe 21st International Annual Joint Conference of the IEEEComputer and Communications Societies (INFOCOM 2002)New York NY USA June 2002

[23] J Polastre J Hill and D Culler ldquoVersatile low power mediaaccess for wireless sensor networks categories and subjectdescriptorsrdquo in Proceedings of the Second InternationalConference on Embedded Networked Sensor Systems Sen-Sysrsquo04 pp 95ndash107 Baltimore MA USA November 2004

[24] T Danmanee K N Nakorn and K Rojviboonchai ldquoCU-MAC a duty-cycle MAC protocol for internet of things inwireless sensor networksrdquo ECTI Transactions on ElectricalEngineering Electronics and Communications vol 16 no 2pp 30ndash43 2018

[25] T Van Dam and K Langendoen ldquoAn adaptive energy-efficientMAC protocol for wireless sensor networksrdquo in Proceedings ofthe 1st International Conference on Embedded NetworkedSensor Systems Los Angeles CA USA November 2003

[26] P Xie and J Cui ldquoR-MAC an energy-efficient MAC protocolfor underwater sensor networksrdquo in International Conferenceon Wireless Algorithms Systems and Applications (WASA2007) Chicago IL USA August 2007

[27] A Rahim N Javaid M Aslam U Qasim and Z A KhanldquoAdaptive-reliable medium access control protocol forwireless body area networksrdquo in Proceedings of the AnnualIEEE Communications Society Conference on Sensor Meshand Ad Hoc Communications and Networks (SECON) SeoulKorea June 2012

[28] H S Kwak S Ullah D H Kwak and C H Lee ldquoA powerefficient MAC protocol for wireless body area networksrdquoJournal of the Korean Institute of Intelligent Systems vol 8no 6 pp 131ndash140 2009

[29] C Omeni A J Burdett and M Park ldquoEnergy efficientmedium access protocol for wireless medical body area sensornetworksrdquo in Proceedings of the 4th IEEEEMBS InternationalSummer School and Symposium on Medical Devices andBiosensors pp 29ndash32 Cambridge UK August 2007

[30] L Sitanayah C J Sreenan and K N Brown ldquoER-MAC ahybrid MAC protocol for emergency response wireless sensornetworksrdquo in Proceedings of the Fourth International Con-ference on Sensor Technologies and Applications pp 244ndash249Mestre Italy July 2010

[31] Y Sun and D B Johnson ldquoRI-MAC a receiver-initiatedasynchronous duty cycle MAC protocol for dynamic trafficloads in wireless sensor networksrdquo in Proceedings of the 6thACM conference on Embedded network sensorsystemsmdashSenSysrsquo08 pp 1ndash14 Raleigh NC USA November2008

[32] M Buettner G V Yee E Anderson and R Han ldquoX-MAC ashort preamble MAC protocol for duty-cycled wireless sensornetworksrdquo in Proceedings of the 4th International Conferenceon Embedded Networked Sensor Systems SenSys 2006pp 307ndash320 Boulder Colorado USA October 2006

[33] S Boulfekhar and M Benmohammed ldquoSynchronous receiverinitiated MAC protocol for long-lived sensor networksrdquoComputers amp Electrical Engineering vol 40 no 2 pp 504ndash516 2014

Journal of Computer Networks and Communications 31

[34] S Mehta and K S Kwak ldquoH-MAC a hybrid Mac protocol forwireless sensor networksrdquo International Journal of ComputerNetworks amp Communications vol 2 no 2 pp 108ndash117 2010

[35] I Rhee A Warrier M Aia J Min and M L Sichitiu ldquoZ-MAC a hybrid MAC for wireless sensor networksrdquo IEEEACM Transactions on Networking vol 16 no 3 pp 511ndash5242008

[36] S Arshad A Al-sadi and A Barnawi ldquoZ-MAC performanceevaluation and enhancementsrdquo Procedia Computer Sciencevol 21 pp 485ndash490 2013

[37] Y Liu I Elhanany and H Qi ldquoAn energy-efficient QoS-aware media access control protocol for wireless sensornetworksrdquo in Proceedings of the IEEE International Confer-ence on Mobile ad hoc and Sensor Systems pp 9ndash11 Wash-ington DC USA December 2005

[38] J Decotignie ldquoWiseMAC an ultra low power MAC protocolfor the downlink of infrastructure wireless sensor networksrdquoin Proceedings of the Ninth International Symposium onComputers And Communications pp 244ndash251 AlexandriaEgypt June 2004

[39] R A Rashid ldquoDevelopment of energy aware TDMA-basedMAC protocol for wireless sensor network systemrdquo EuropeanJournal of Scientific Research vol 30 no 4 pp 571ndash578 2009

[40] H P Van Hoesel ldquoA lightweight medium access protocol(LMAC) for wireless sensor networksrdquo in Proceedings of theInternational Workshop on Networked Sensing Systems Bal-timore MD USA November 2004

[41] C Li P Wang H Chen andM Guizani ldquoA cluster based on-demand multi-channel MAC protocol for wireless multi-media sensor networksrdquo in Proceedings of the IEEE In-ternational Conference on Communications pp 2371ndash2376Beijing China May 2008

[42] J L Sobrinho and A S Krishnakumar ldquoQuality-of-service inad hoc carrier sense multiple access wireless networksrdquo IEEEJournal on Selected Areas in Communications vol 17 no 8pp 1353ndash1368 1999

[43] J P John ldquoProviding MAC QoS for multimedia traffic in80211e based multi-hop ad hoc wireless networksrdquo ComputerNetworks vol 51 no 1 pp 153ndash176 2007

[44] S Lofty and N Padmavati ldquoA survey on mobility basedprotocols in WSNsrdquo in Proceedings of the InternationalConference Communication and Manufacturing pp 12ndash15Hangzhou China July 2014

[45] S Deng L Shen and J Li ldquoMobility-based clustering pro-tocol for wireless sensor networks with mobile nodesrdquo IETWireless Sensor Systems vol 1 no 1 pp 39ndash47 2011

[46] M Khalil A Khalid F U Khan and A Shabbir ldquoA review ofrouting protocol selection for wireless sensor networks insmart citiesrdquo in Proceedings of the 24th Asia-Pacific Confer-ence on Communications (APCC) Ningbo China November2014

[47] O Message Q Telemetry and T Mqtt Committee Specifi-cation Draft 02 April 2014

[48] M Collina M Bartolucci A Vanelli-coralli andG E Corazza ldquoInternet of things application layer protocolanalysis over error and delay prone linksrdquo in Proceedings ofthe 7th Advanced Satellite Multimedia Systems Conference andthe 13th Signal Processing for Space CommunicationsWorkshop Livorno Italy September 2014

[49] A Talaminos-barroso M A Estudillo-valderrama L M RoaJ Reina-tosina and F Ortega-ruiz ldquoA machine-to-machineprotocol benchmark for e health applicationsmdashuse case re-spiratory rehabilitationrdquo Computer Methods and Programs inBiomedicine vol 129 pp 1ndash11 2016

[50] V Goyal O Pandey A Sahai and B Waters ldquoAttribute-based encryption for fine-grained access control of encrypteddatardquo in Proceedings of the 13th ACM Conference on Com-puter and Communications Security pp 89ndash98 AlexandriaVA USA October 2006

[51] M Singh M A Rajan V L Shivraj and P BalamuralidharldquoSecure MQTT for internet of things (IoT)rdquo in Proceedings ofthe Fifth International Conference on Communication Systemsand Network Technologies Gwalior India April 2015

[52] A Stanford-Clark and H L Truong MQTT for Sensor Net-works (MQTT-SN) Protocol Specification httpmqttorgnewwp-contentuploads200906MQTT-SN_spec_v12pdf 2013

[53] RFC ldquoExtensible messaging and presence protocol (XMPP)rdquoin Proceedings of the IETF pp 1ndash9 San Diego CA USAAugust 2005

[54] S Vinoski and I Technologies ldquoQueuing protocolrdquo IEEEInternet Computing vol 10 no 6 pp 87ndash89 2006

[55] J L Fernandes I C Lopes J J P C Rodrigues and S UllahldquoPerformance evaluation of restful web services and AMQPprotocolrdquo in Proceedings of the Fifth International Conferenceon Ubiquitous and Future Networks (ICUFN) pp 2ndash7 DaNang Vietnam July 2013

[56] M Asim ldquoA survey on application layer protocols for internetof things (IoT)rdquo International Journal of Advanced Researchin Computer Science vol 8 no 3 pp 996ndash1000 2017

[57] K An A Gokhale D Schmidt S Tambe P Pazandak andG Pardo-castellote ldquoContent-based filtering discovery pro-tocol (CFDP) scalable and efficient OMG DDS discoveryprotocolrdquo in Proceedings of the 8th ACM InternationalConference on Distributed Event-Based Systems New YorkNY USA July 2014

[58] E Web What is a web service pp 52ndash57 2010[59] M Laine RESTful Web Services for the Internet of ings

pp 2ndash4 2012 httpmediaTkk[60] IETF -e WebSocket protocol pp 1ndash71 2011[61] V Pimentel ldquoCommunicating and displaying real-time data

with websocketrdquo IEEE Internet Computing vol 16 no 4pp 45ndash53 2012

[62] Wikipedia 2019 httpsenwikipediaorgwikiStreaming_Text_Oriented_Messaging_Protocol2019

[63] K Raghavendra and S Nireshwalya ldquoApplication layer se-curity issues and its solutionsrdquo IJCSET vol 2 no 6pp 1266ndash1269 2012

[64] Y Xue B Ramamurthy andM C Vuran ldquoSDRCS a service-differentiated real-time communication scheme for eventsensing in wireless sensor networksrdquo Computer Networksvol 55 no 15 pp 3287ndash3302 2011

[65] C Sonmez S Isik M Y Donmez O D Incel and C ErsoyldquoSUIT a cross layer image transport protocol with fuzzy logicbased congestion control for wireless multimedia sensornetworksrdquo in Proceedings of the 5th International Conferenceon New Technologies Mobility and Security (NTMS) pp 1ndash6IEEE Istanbul Turkey May 2012

[66] T Melodia and I Akyildiz ldquoCross-layer QoS-aware com-munication for ultra wide band wireless multimedia sensornetworksrdquo IEEE Journal on Selected Areas in Communica-tions vol 28 no 5 pp 653ndash663 2010

[67] Q Wang and M A Abu-Rgheff ldquoCross-layer signalling fornext-generation wireless systemsrdquo in Proceedings of the 2003IEEE Wireless Communications and Networking 2003WCNC 2003 vol 2 pp 1084ndash1089 IEEE New Orleans LAUSA March 2003

[68] P Liu Z Tao S Narayanan T Korakis and S PanwarldquoCoopMAC a cooperative MAC for wireless LANsrdquo IEEE

32 Journal of Computer Networks and Communications

Journal on Selected Areas in Communications vol 25 no 2pp 340ndash354 2007

[69] H Shan H T Cheng and W Zhuang ldquoCross-layer co-operative MAC protocol in distributed wireless networksrdquoIEEE Transactions onWireless Communications vol 10 no 8pp 2603ndash2615 2011

[70] F Liu T Korakis Z Tao and S Panwar ldquoAMAC-PHY cross-layer protocol for wireless Ad-Hoc networksrdquo in Proceedingsof the 2008 IEEE Wireless Communications and NetworkingConference pp 1792ndash1797 IEEE Las Vegas NV USA March2008

[71] L Jacob and H R Shamna ldquoEfficient cooperative MAC androuting in wireless networksrdquo Transactions on Networks andCommunications vol 3 no 5 2015

[72] S Rao and K Shama ldquoCross layer protocols for multimediatransmission in wireless networksrdquo International Journal ofComputer Science amp Engineering Survey vol 3 no 3pp 15ndash28 2012

[73] L Han S Park S Kang and H Peter ldquoAn adaptive cross-layer FEC mechanism for video transmission over 80211WLANsrdquo in Proceedings of the 2009 International Conferenceon Internet pp 209ndash215 Las Vegas NV USA December2009

[74] N Li J Martınez and V H Dıaz ldquo-e balanced cross-layerdesign routing algorithm in wireless sensor networks usingfuzzy logicrdquo Sensors vol 15 no 8 pp 19541ndash19559 2015

[75] G Sun J Qi Z Zang and Q Xu ldquoA reliable multipathrouting algorithm with related congestion control scheme inwireless multimedia sensor networksrdquo in Proceedings of the2011 3rd International Conference on Computer Research andDevelopment vol 4 pp 229ndash233 IEEE Shanghai ChinaMarch 2011

[76] M Li Y Jing and C Li ldquoA robust and efficient cross-layeroptimal design in wireless sensor networksrdquoWireless PersonalCommunications vol 72 no 4 pp 1889ndash1902 2013

[77] J Yan M Zhou and Z Ding ldquoRecent advances in energy-efficient routing protocols for wireless sensor networks areviewrdquo IEEE Access vol 4 pp 5673ndash5686 2016

[78] A Ahmad S Ahmad M H Rehmani and N U Hassan ldquoAsurvey on radio resource allocation in cognitive radio sensornetworksrdquo IEEE Communications Surveys amp Tutorials vol 17no 2 pp 888ndash917 2015

Journal of Computer Networks and Communications 33

International Journal of

AerospaceEngineeringHindawiwwwhindawicom Volume 2018

RoboticsJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Active and Passive Electronic Components

VLSI Design

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Shock and Vibration

Hindawiwwwhindawicom Volume 2018

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawiwwwhindawicom

Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Control Scienceand Engineering

Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Journal ofEngineeringVolume 2018

SensorsJournal of

Hindawiwwwhindawicom Volume 2018

International Journal of

RotatingMachinery

Hindawiwwwhindawicom Volume 2018

Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Navigation and Observation

International Journal of

Hindawi

wwwhindawicom Volume 2018

Advances in

Multimedia

Submit your manuscripts atwwwhindawicom

Page 13: ReviewArticle - downloads.hindawi.com · ReviewArticle QoSStrategiesforWirelessMultimediaSensorNetworksinthe ContextofIoTattheMACLayer,ApplicationLayer,and Cross-LayerAlgorithms MuwongeSsajjabbiBernard

pauses in between packets to allow the sending node topause and listen to the channel and allow the receivingnode to resend an early ACK packet during that pause-is makes the sending node begin sending datapackets and the receiving node cut excessive preamblesand thus decrease per-hop latency and energy wastageon unnecessary waiting and transmissionPacketizing radios low-power listening is limited incapability for supporting packetizing radios butX-MACrsquos short strobed preambles can support all typesof digital radiosAdaptation to traffic load while a number of WSNapplications are reliable in traffic production they stillmust adapt to changing traffic Nodes are going to haveseparate sleep schedules depending on traffic loadConsidering systems that have time-fluctuating trafficall predetermined static schedules are suboptimalOptimality Buettner et al [32] model the expectationof energy as follows

Es (preamble energy + energy perACK listen)

lowast(expected preamble minus listen iterations)

+(energy to send parket)

Es PTxSp + PRxSal1113872 1113873 Rl + Rs( 1113857

Rl minus Sp+ SdPTx

(16)

where Es expected energy to send a packet PTx andPRx are the transmission and receiver power re-spectively Sp Sal and Sd are the duration of thesenderrsquos preamble ACK listen and data transmissionperiods respectively and Rl and Rs are the receiverrsquoslisten and sleep periods respectively Expected energyto receive a packet is given as

Er (listen cycle energy + sleep cycle energy)

lowast(expected iterations for preamble arrival)

+(energy to sendACK) +(energy to receive packet)

Er PsRs + PRxRl( 1113857

1 minus 1 minus Pd(t)( 1113857 Rl + Rs( 1113857+ RaPTx + RdPRx

(17)

And the one-hop expected latency is given as

Lat (preamble duration + ACK listen)

lowast(expected number of iterations)

+ duration to send packet

Lat Sp + Sal1113872 1113873 Rl + Rs( 1113857

Rl minus Sp

(18)

Advantages

(i) It increases energy efficiency(ii) It reduces data latency(iii) It increases throughput

2115 A-MAC (Advanced MAC) A-MAC is a TDMA-based protocol whose design is aimed at enhancing thelifetime of the network for low rate and reliability of datatransmission [39] To assign time slots it does not depend ona central controller like most protocols but it gathers in-formation from its neighbors so as to allocate the slotsthrough a distributed algorithm Energy is controlled byscheduling power when there is not a single event Its designcomprises many frames with each having many time slotswhich at the beginning have got a beacon transmitted tosynchronize and exchange information with neighborsNodes to participate in the subsequent session are selectedby the controlled node so they do not go into the sleep modewith others [13] -is protocol operates under four time slotdivisions namely initial wait discover and active states Anode in the initial state begins by listening to the medium forthe beacon packet from other nodes for network synchro-nization which when received makes it change the timerthrough subtraction of beacon transmission time from re-ception time To avoid synchronization problems due todrift and allow for continuity the node has to get thestrongest beacon signal It then enters the wait state afterchoosing a number of waiting frames so as to reduce theprobability that nodes get into the discover state simulta-neously In case the synchronization beacon is lost while stillin the wait state it will return to the initial state for anotherone On expiry of the waiting counter it enters the discoverstate and collects information from neighbors by listening totheir beacon messages On successful selection of a time slotit enters the active state where it endlessly transmits beaconsat the initiation of the slot A node goes to sleep when abeacon has been transmitted and there are no more data fortransmission or in case the neighborrsquos beacon received in-dicates no data packets are coming in [39]

Advantages

(i) It can enable multicasting(ii) It enhances the lifetime of the network(iii) It offers reliable data transmission

2116 L-MAC (Lightweight MAC) Based on TDMA thisprotocol is aimed at minimizing the number of transceiverswitches adapting the nodal sleep interval to data trafficamount and making implementation less complex [40] Apacket is transmitted comprising a control and a data partControl packets have a fixed size and the time slot con-trollerrsquos ID -ey show the distance from the transmittingnode for simplified routing to the network gateway as well asaddressing target node and reporting data unit length andinternode synchronization Adjacent nodes are more in-terested in getting control packets from nearby nodes -enodersquos transceivers are switched off until the subsequenttime slot unless the node is also addressed or the message isomnicast If so it listens to the data unit which may not fillthe remaining time slot in totality-e transceivers on eitherside are turned off when the transfer of packets is successfulEnergy wastage through idle listening in uncontrolled timeslots is avoided via a short time-out interval Network nodes

Journal of Computer Networks and Communications 13

are able to communicate without colliding-is saves energyand increases the lifetime of the network A node in L-MACcan only transmit one message per frame [13] -e networkis set up in such a way that all nodes are not synchronized onbeing switched on-e gateway takes control of a timeslot soas to have them synchronized and its immediate neighborgets the control messages thereby synchronizing their timerswith the gatewayrsquos Neighboring nodes learn of multiplegatewaysrsquo time slots in their vicinity just after one frame-ejust synchronized nodes randomly choose time slots forcontrolling as long as they are not occupied -e nodes thuspreserve a table for the neighboring nodes which makes atimeslot reusable after a minimum of 3 hops and preventsmessage collision All nodes keep their slots before thebattery is down or learn of an impending collision All nodescan track their hop distances to a selected gateway node andthe information is communicated in control messages [40]

Advantages

(i) It minimizes the number of transceiver switches(ii) It adapts the nodal sleep interval to data traffic

amount(ii) It makes implementation less complex(iii) It offers improved network lifetime(iv) Energy wastage through idle listening in un-

controlled time slots is avoided via a short time-out interval

Disadvantages

(i) L-MAC is not very good for mobile SNs since slotsare calculated only once

(ii) It suffers idle listening because nodes keep listeningto slotsrsquo control parts so as to get data and enableother nodes to join the network all the time [13]

2117 C-MAC -is was designed for WSNs with 3 seg-ments namely (i) aggressive RTS having double channelcheck to assess the channel (ii) anycast to initialize the flowand (iii) convergent packet forwarding to stabilize the flowIt promises an improvement in energy efficiency lowerlatency and enhanced throughput while avoiding syn-chronization overhead though it operates only in low duty-cycling apps especially where there is less traffic But whenthere is more traffic the protocol employs anycast-basedpacket forwarding for waking the nodes up or finding aforwarder quickly and converges from route-suboptimal toroute-optimal unicast After this energy could be preservedby the nodes using a synchronized wake-up plan [13]

Advantages

(i) It improves energy efficiency(ii) It lowers latency(iii) It enhances throughput avoiding synchronization

overhead

Disadvantages

(i) It operates only in low duty-cycling applicationsespecially where there is less traffic

22 MAC Protocols for WMSNs

221 Diff-MAC -is is a CSMACA-based QoS-awarehybrid-priority-based MAC protocol for WMSNs It targetsenhancing the channel utilization while making use of ef-ficient service differentiation to do coordination for mediumaccess of each traffic class while giving fair quick datatransfer It offers QoS using the following features [14]

(i) It balances energy consumption and delay throughsensor node duty-cycling adaptation depending onthe dominant traffic class

(ii) It has a feature for fragmentation and messagepassing that breaks lengthy video frames intosmaller packets that are transmitted as a burst thusreducing retransmission costs if there are any MACfailures

(iii) -ere is fair data delivery amongst the nodes andtraffic classes respectively resulting from intranodeand intraqueue prioritization reducing unbearableperformance

(iv) -e number of collisions can be reduced and packetlatencies are kept within limits by adjusting the sizeof the congestion window in this protocol as per thetraffic needsAdvantages

(i) Diff-MAC is very suitable for WMSNs since itadapts quickly to changes in networkconditions

(ii) It is a fair protocol(iii) It is scalable

Disadvantages

(i) It is quite difficult to monitor the statistics ofthe network

(ii) Adapting dynamically to the network condi-tions is not an easy task

(iii) It exhibits some degree of latency in packetdelivery

-e implementation of Diff-MAC on Imote2 indicatesthat resource requirements can be met on the sensorhardware Results of extensive simulation runs showed thatDiff-MAC outperforms its competitors

222 Cluster-Based On-Demand Multichannel MAC Pro-tocol for WMSN (COM-MAC) -is is an energy-efficientprotocol in WMSNs with high throughput and reliable datatransmission It achieves energy efficiency by schedulingmultichannel media access to reduce channel contentionand remove collision idle listening and overhearing [41]

Design it has a three-phased operation with the (i)request phase (ii) scheduling phase and (iii) datatransmission phaseRequest phase here all nodes have two protocols eachfor sending requests to the cluster head (CH) ie thecontention-based and contention-free TDMAFDMA

14 Journal of Computer Networks and Communications

protocols -e contention-based protocol allocates achannel per node for sending the request message andfree channels are useable as control channels Forcontention-free protocols a control slot is allocatedwhen a network is deployed Here a slot per channel isallocated to a different node enabling requests to be sentwithout interfering with transmissions from othernodes -e contention-based protocol operates undertwo steps (i) Control channel assignment phase achannel is assigned per node for transmitting a request(REQ) message A channel is assigned to multiplenodes and nodes are evenly distributed to availablechannels for congestion control (ii) Request trans-mission phase nodes with data send a REQmessage viathe assigned control channel to notify the CH whichresponds with an ACK message -e contention-freeTDMAFDMA protocol too has two phases includingthe control slot assignment and request transmissionphases For unreliable channel conditions and lowtraffic load the contention-based protocol is fit for usebecause of less collisions and congestion whereas thecontention-free protocol is preferable in heavy trafficload and reliable channel applicationsScheduling phase from the above phase a schedule isgenerated by the CH for coordination and this enablesnodes to transmit and broadcast their data via controlchannels-e schedule for broadcasting only comprisesdata for nodes assigned to a channel -e schedule isrebroadcast for transmission reliability enhancementbut for energy efficiency the nodersquos transceiver isswitched off when all the schedule is completely de-livered till it is the time for transmitting again-e timeslot and data communication radio channel are allincluded in the schedule generated for each nodeData transmission phase on getting the schedule thesensor node sends its data as per the time slot and thechannel assigned -e time slot is divided into datatransmission andACK sections ACK supports link layererror control Using implicit selective repeat ARQ theCH sends an ACK message for all received packets Ifthere is no ACK the packet is assumed lost and hence aretransmission during the subsequent interval is done-is enhances reliability of transmissions though excessdelay is possible here which is also intolerable for real-time critical apps like video or audio -is gives rise to ahybrid MAC protocol to exploit the unused spectrumduring data transmission sessions [41]Advantages

(i) It achieves energy efficiency by scheduling mul-tichannel media access to reduce channel con-tention and remove collision idle listening andoverhearing

(ii) It maximizes network throughput by dynamicallyallocating time slots and channels for nodes byexecution of a traffic-adaptive QoS-aware sched-uling algorithm depending on needs for QoS andtraffic data information

(iii) It increases transmission reliability by in-corporating spectrum-aware ARQ to avoid spec-trum wastage

223 EQoSA -is protocol is developed for video andimage transfer across WMSNs in a hybrid mode withprovision of QoS Concerning the active sensor nodes andtheir traffic loads it employs dynamic session sizes andchanges bitmap-assisted (BMA) MACrsquos fixed session sizeNodes declare any data available for transmission duringcontention time -is forces the cluster head to make aschedule of how slots have been assigned It communicatesthis schedule to the nodes giving each node a given numberof slots per session and so enabling the busty traffic to beaccommodated [14] -e protocol does not perform wellwith DiffServe mechanisms but does well with schemessupporting busty traffic loads and strong cluster heads arenecessary in WMSNs that can assign and broadcast slots

224 QoS-Supported Energy-Efficient MAC (QE-MAC)Based on the IEEE 80211e standard this protocol ensuresfairness and has low latency and jitter and better efficiencyon energy usage than other QoS-aware MAC protocols forWMSNs It operates under a two-phased routine

-e first phase introduces the innovating prioritymechanism in IEEE 80211e QoS and several data types areassigned priorities Its coordination function is a combi-nation of distributed contention-based channel allocationand centralized polling-based channel access mechanisms

-e second phase preserves the nodesrsquo energy usingdynamic duty cycling -ere is an exchange of RTSCTSpackets to supplement CSMACA by muting nodes near thesenderreceiver during packet duration [42]

Advantages

(i) It ensures fairness(ii) It has low latency and jitter and better efficiency on

energy usage than other QoS-awareMAC protocolsfor WMSNs

225 Black-Burst (BB) Contention A contention-basedMAC scheme for enhancing instantaneous QoS access tocarrier-sense WSNs is proposed in [42] In this schemenodes contend for the channel until it is free Real-timepackets are given higher priority over non-real-time packetsand if a node is transmitting real-time packets it is prior-itized over other nodes [15] Voice and video are some of thereal-time apps in consideration here which periodicallyaccess the common radio channel when transmitting -eseneed constrained end-to-end delay of packets at the MAClayer -is protocol majorly does the following [42]

(i) Prioritization of real-time traffic(ii) Implementation of the round-robin mechanism in

real-time nodes(iii) Bounded access delays among real-time packets

Journal of Computer Networks and Communications 15

(iv) Implementation of real-time communication withvarying needs of bandwidth good for multimediatraffic

Amechanism for dynamically assigning packet prioritiesbased on their deadlines and the traversed hops is proposedin [43] -e ReAllocative Priority (ReAP) mechanism ismeant to provide QoS at the MAC layer to video traffic -isscheme introduces adaptive TXOP by modifying the TXOPdynamically based on the queue length [15] Other protocolsin this category include dynamic duty cycle and adaptivecontention window-based MAC protocol frame sharing(FRASH) MAC protocol and real-time independentchannels (RICH) MAC protocol

Advantages

(i) It can implement real-time communication withvarying needs of bandwidth

23 Challenges and Opportunities(a) One of the challenges in satellite transmission is

cross-talk leading to interference in different fre-quency bands It is necessary to increase the rise andfall times as well as protect sensitive nodes as wecircumvent floating ones if we are to regulate cross-talk as in satellite communication therefore Wefurther need to investigate the possibility of usingdifferential signaling -is constitutes an opportu-nity for further research

(b) Another challenge is that the hidden and exposedterminal problems are still noticeable despite the useof RTSCTS solutions in MAC since collisions arestill possible from transmissions of separate nodesAt the MAC layer ACKs might block efficientchannel usage by prohibiting exposed terminalsfrom reusing the channel and may be eliminatedleaving no proof of sender data packet receptionUsing a directional antenna is one of the proposedtechniques of high-speed WSNs like IEEE 80211-is constitutes an opportunity for further study intothese issues for better solutions though

(c) Another challenge is that since nodes in contention-based MAC protocols are mobile there are manycollisions by the packets and schedule in-consistencies in schedule-based protocols in the two-hop neighborhood information when nodes areentering or leaving [44] Deng et al [45] propose amobility-based clustering (MBC) protocol for mo-bile nodes to enhance the packet delivery rate -isprotocol in comparison with CBR has 25 chancesof decreasing collision-induced packet losses and50 chances in comparison with low-energyadaptive clustering hierarchy (LEACH)

(d) -e process of analyzing and evaluating routingperformance for WMSNs is quite complex becauseof the bursty multimedia traffic Because of the wayencoding scheme frames are structured as well as thechanges in between the scenes that occur naturally

the compressed video frames display some substantialburstiness on several time scales -ere will also befrequent route recalculation that is as well energy in-tensive for nodes frequently capturing and transmittingenergy -e solution here could as well possibly be theuse of the MBC protocol Route recalculation tooshould be dynamic to cater for the mobile nodes whichmight be in and out of the range at different instants oftime Recently the use of smart dumpsters for effectivewaste management in smart cities has also been pro-posed as a possible solution [46]

(e) For WSNs their routing protocols do not look atmultimedia applications which require enormousbandwidth and processing energy and should betransmitted in real time with utmost fidelity Extrachallenges in routing are imposed in WMSNs indeployment of heterogeneous sensors since therecould be audio video and still pictures in addition toscalar data during transmission Heterogeneous datahandled also have different QoS requirements andwith many different business needs different ser-vices with varying requirements give rise to a bigchallenge for the routing design -ere is a need todesign protocols that can support WMSNs such asEQoSA [14] for this purpose

Table 2 summarizes the reviewed MAC protocols forWSNs and the ones designed for suitability in WMSNs

3 QoS Strategies at the Application Layer

We discuss many protocols used to solve different needs ofcommunication between machines

31 ConstrainedApplication Protocol (CoAP) It is a requestresponse protocol designed by the Internet Engineering TaskForce (IETF) for constrained devices such as those havinginsufficient RAM or CPU and WPANs with low-powerconstraints resulting in poor transfer of packets andenormous overhead -e protocol was designed for ma-chine-to-machine (M2M) apps and system automation toreduce bandwidth requirements increase packet transferdecrease the overhead and make work simpler leading tolightweight implementation HTTP commands like GETPOST PUT and DELETE are used in a client-server ar-chitecture Using the publishsubscribe architecture CoAPcan support multiple users and hence yield better perfor-mance via multicasting enabling asynchronous messageexchange -is protocol can support both multicasting andunicasting via UDP and hence is a reliable and simpleprotocol -is is done via the messaging sublayer (for re-liability based on UDP) and the requestresponse sublayer(for communication and interaction) [16]

It uses the following messages

(i) Confirmable message this guarantees reliabilityusing ACK messages sent by the receiver node-ese could be synchronous or asynchronous ifthere is a need for more computational time

16 Journal of Computer Networks and Communications

(ii) Nonconfirmable message here no ACK messagesare sent back

(iii) Acknowledgment (ACK) message this confirmsreceipt of a confirmable message

(iv) Reset message this confirms receipt of an un-processed message in case any critical parts nec-essary for interpreting the message are missing

(v) Piggybacked response the receiver directly re-sponds while getting the ACK message

(vi) Separate response the reply by the receiver is dis-crete from the ACK message

CoAP has no in-built security features It is secured bythe datagram transport layer security (DTLS) that runs ontop of UDP though it was not designed for the IoT -ismakes its suitability debatable since DTLS offers no supportfor multicasting It also needs more packets for handshakingthereby increasing the traffic volume and computationalresources and so shortening the lifespan of mobile devicessince they use batteries [17]

Advantages

(i) It can support multiple users and hence yieldbetter performance via multicasting

Table 2 Comparison of MAC layer protocols used in QoS provisioning

Protocol Features QoS parameter Comments Access Priorityassignment Energy eff Designed for

S-MAC[22] Collision avoidance Energy latency Requires periodic

synchronization CSMA Hybrid Medium WSNs

B-MAC[23] Collision avoidance Energy delay

throughput Flexible interface CSMA Dynamic High WSNs

T-MAC[25] Collision avoidance Energy Complex nonscalable CSMA Dynamic High WSNs

C-MAC[13]

Dynamic channelallocation Energy Multichannel MAC CSMA Dynamic Medium WSNs

CU-MAC[24]

Multichannelmechanism

-roughputlatency High packet delivery ratio TDMA Dynamic mdash WSNs

A-MAC[39]

Per flow serviceguarantees Delay m-ary tree model

dynamic prioritiesTDMACSMA Dynamic Medium WSNs

ER-MAC[30]

Emergencyresponse

Energy delaydelivery ratio

Adapts to traffic andtopology changes

TDMACSMA Hybrid Higher than Z-

MAC WSNs

RI-MAC[31] Collision detection Delay packet

delivery ratioHigh packet delivery ratio

and throughput CSMA Dynamic Medium WSNs

SRI-MAC[33] Collision avoidance Energy delay Good data transfer and

energy consumption TDMA Dynamic High WSNs

H-MAC[34] Slotted sleep time -roughput

latencyData transmission only in

sleep timeCSMAAloha Dynamic Medium WSNs

Z-MAC[36] Collision avoidance Energy delay

throughputVery good channel usage

and error syncCSMATDMA Dynamic Medium WSNs

Q-MAC[37]

Intrainternodescheduling Energy High QoS improved

energy efficiency CSMA Hybrid High WSNs

X-MAC[32]

Overhearingavoidance

Energy delaythroughput

Reduces excessivepreamble by strobing CSMA Adaptive

static Medium WSNs

L-MAC[40] Collision avoidance Network

lifetime energyReduces the number oftransceiver switches TDMA Static Medium WSNs

WiseMAC[38]

No downlinkchannel collision Power delay Significantly low-power

consumptionAlohaCSMA Static Higher than

ZigbeeInfrastructure

WSNsAR-MAC[27] Collision avoidance Energy delay Uses data packets and

control packets TDMA Static Higher thanIEEE 802154 WBANs

BB-MAC[42]

Real-time trafficprioritization Delay Employs round robin in

real-time nodesCSMATDMA Dynamic High WMSNs

COM-MAC [41] Collision avoidance Packet delay

throughputBalances reliability and

retransmissionTDMAFDMA Dynamic High WMSNs

EQoSA[14]

Busty traffic loadaccommodation Energy Provides QoS support for

video and imagesTDMACDMA mdash High WMSNs

QE-MAC[42] Collision avoidance Energy jitter

latencyRuns central polling-like

channel access CSMA Dynamic High WMSNs

R-MAC[26] Collision avoidance Energy delay Reservation based mdash Static High Underwater

SNs

Journal of Computer Networks and Communications 17

(ii) It enables asynchronous message exchange(iii) It is a reliable and simple protocol

32 Message Queue Telemetry Transport (MQTT)Developed by IBM and later improved by OASIS with anintention of decreasing bandwidth MQTT is a simple open-source publishsubscribe protocol analogous to the client-server version It is appropriate for lightweight IoT appli-cations and M2M transmissions in constrained situationslike limited power computation capability and memory orbandwidth -e network bandwidth and the computationalresource usage are decreasing since clients are not updatedmaking publishsubscribe protocols outperform requestresponse protocols via IoT requirements [16] -is protocolcan run over TCPIP and so has the following advantages

(i) Delivering of packets with reliable guarantee(ii) Enabling multicasting(iii) Remote device session establishment(iv) Reduction of transport overhead and protocol ex-

change to minimize network traffic(v) Provision of the notification mechanism when

anomalies occur [47]

MQTTrsquos design targets economical use of bandwidthand battery so it has a major application in FacebookMessenger MQTT ensures QoS and reliability of messagedelivery by providing the following three QoS levels

(1) Fire and forget a message is sent once depending onthe networkrsquos best effort and there is no need forACK (lowest QoS level)

(2) Delivered at least once a message is sent at least onceand an ACK message is needed to avoid duplicates

(3) Delivered exactly once by a four-way handshake itensures a message is sent exactly one time (highestQoS level)-e architecture it has three key components (i)publishersmdashlightweight nodes sending data to thebroker before returning to sleep (ii) broker whichorganizes sensory data in topics and sends them tointerested subscribers and (iii) subscribersmdashIoTapps interested in data sent by sensors through thebroker as shown in Figure 3

MQTT is a simple lightweight protocol that is betterthan CoAP via low and high packet loss probability interms of throughput and exhibits lower latency in lowsampling rates [48] It supports a number of imple-mentations for embedded devices and mobile applicationsand offers web-socket support too Nevertheless it hassome shortcomings

(i) It is only employed in simple data types(ii) It exhibits high latency in high sampling rates(iii) It is unsuitable for real-time apps in such cases(iv) It does not provide automatic discovery andor

sufficient security at the protocol level [49]

33 Secure Message Queue Telemetry Transport (SMQTT)-is is an extension of MQTT that is based on lightweightattribute-based encryption using elliptic curve cryptogra-phy -e protocol has three entities as shown in Figure 4(i) publisher that publishes information under a certaintopic (ii) subscriber that obtains the same topic in-formation via a broker and (iii) public key generator(PKG) or broker that acts as the trusted third party -eprotocol has four phases namely (a) set-up phase (b)encrypt phase (c) publish phase and (d) decrypt phase

(a) Set-up phase publisher and subscriber provide aunique ID with attributes for registration with thePKG -is generates a secret master key set andpublic parameters that it publishes with set U A1A2 An For ciphertext policy-attribute-basedencryption (CP-ABE) the PKG gives a key set todevices with attributes [50]

(b) Encrypt phase based on the access tree the publishercreates an access policy An access tree is sent by thepublisher to the PKG for generating the key-gen-eration policy in the key policy- (KP-) based ABEalgorithm -e publishers know beforehand thetopics and subscribers to access as well as the rulesfor access Subscribers are issued with private keysfor all communication sessions Publishers in theCP-ABE algorithm come up with access trees andrules where the payload is encrypted by the pub-lisher who also issues more info needed for de-cryption and the rules With the help of public keycryptography data are encrypted giving rise to ci-phertext on the basis of KPCP-ABE and the receivergets SUBACK from the broker [51]

(c) Publish phase ciphertext is embedded in theSPublish command as payload and topic name isdefined in the header by the publisher -e brokerreceives the SPublish packet and replies withPUBACK prompting the publisher to respond withthe PUBREL packet -e broker then broadcasts totopic subscribers and deletes the data forwarding thePUBCOMP packet to the sender node

(d) Decrypt phase the encrypted information isdecrypted by the subscriber with the help of a secretkey Considering KP-ABE on verification that itsatisfies the access policy the subscriber invokes thePKG for a corresponding key which it sends afterverifying the request With CP-ABE subscribers usethe private key for decryption after authenticationthereby enabling offline interaction [51]Advantages

(i) -e information being encrypted makes it moresecure

34 MQTT for Sensor Networks (MQTT-SN) -is is a ver-sion of MQTT designed to operate under environments oflow bandwidth high link failures and short message lengthIt is a publishsubscribe (pubsub) protocol designed for

18 Journal of Computer Networks and Communications

UDP-based WSNs to visualize communication for low-power devices [5152] It comprises a publisher device actingas the MQTT-SN client to forward communications to agateway that also changes the MQTT-SN message to theMQTTmessage before delivering it to the broker that sendsit to MQTT-SN subscribers [51]

It has a gateway for changing protocols from MQTT toMQTT-SN (Figure 5) [52] -is protocol supports nodes in

the sleeping mode using an offline keep-alive procedure thatcan be used by battery-operated devices to get into thesleeping mode To support sleeping clients MQTT-SN has anew offline keep-alive procedure with which battery-oper-ated devices can go to the sleeping state in which theirmessages are buffered at the server until they are awake InMQTT-SN clients with no preconfigured gateway addresscan figure out the real address of their operational gateway

Publisher

Subscriber

Subscriber

Broker

Publisher

4G

Figure 3 Architecture of the MQTT protocol

Register by mentioning URI Register by mentioning URI

Access tree

Set-up

Publisher ReceiverPKG(broker)

Key management Key management

Encrypt data using public parametersand random key This is done using

CP-ABE Publish it as

Subscribe (topics)SUBACK

S-Publish (TopicEncrypt (data))S-Publish (TopicEncrypt (data))

Encryptand publish

PUBACK

PUBACK

PUBRELPUBRELPUBCOMP

PUBCOMP

Figure 4 SMQTT protocol

Journal of Computer Networks and Communications 19

within the network through a discovery procedure Simul-taneously we may have several gateways in a single wirelessnetwork that can be cooperative via load sharing

MQTT-SN architecture it is composed of MQTT-SN clients gateways and forwarders (Figure 5)rough the MQTT-SN gateway (GW) the MQTT-SN protocol is used by the MQTT-SN clients toconnect to the MQTT server Integration betweenthe MQTT-SN GW and the MQTT-SN server mightbe possible or not Translating between the MQTT-SN and the MQTT for a stand-alone GW is doneusing the MQTT-SN protocol If a GW is detachedfrom the network it can be accessed by the MQTT-SN clients through a forwarder is just encapsu-lates the received MQTT-SN frames that are on thewireless side before forwarding them to the GW un-altered It then decapsulates the received frames tooresending them to the client as sent by the GW [52]ere are 2 types of gateways (a) transparent gatewayand (b) aggregating gateway as shown in Figure 6

(i) Transparent gateway this launches and sustains adirect MQTT connection between the MQTT-SNclient and the MQTT server which is entirely re-served for end-to-end client-server messaging enumber of servers and that of MQTT-SN clientsconnected to the GW are the same is transparent

GW translates syntax between these protocols and iseasier to implement compared to the aggregatingGW But the MQTT server must support dierentsessions for all connected clients Some MQTTservers are congured to enable few parallel sessions

(ii) Aggregating gateway this maintains a single MQTTconnection to the server on behalf of all clients Allexchanged messages from the MQTT-SN client to theaggregating GW are directed to the GW which sortsthe information to forward to the server e aggre-gating GW decreases the number of MQTT sessionssimultaneously supported by the server in WSNs withmany battery-operated sensors and actuators (SAs)having limited storage and processing capabilities It ishowever more complex to implement compared to atransparent GWe general format of the MQTT-SNprotocol message has two or four message headeroctets and n octets of variable header that is optionaldepending on the message in question MQTT-SN has256 message types (MsgType) with those from0X1Eminus 0XFD 0X19 and 0XFF reserved for futureusage [51]

Advantages

(i) e protocol can operate under environmentsof low bandwidth high link failures and shortmessage length

MQTT-SN gateway

MQTT-SN

MQTT

MQTT-SN

MQTT-SN

MQTT-SN

MQTT-SN client

MQTT-SN clients

MQTT broker

MQTT-SN client

MQTT-SN client

MQTT-SN gateway

MQTT-SN forwarder

Figure 5 MQTT-SN architecture

20 Journal of Computer Networks and Communications

(ii) It is a pubsub protocol designed for UDP-basedWSNs to visualize communication for low-power devices

(iii) It can be optimal for low-cost battery-operatednodes having low processing power and lessstorage

(iv) It supports nodes in the sleeping mode using theobrvbarine keep-alive procedure used by battery-operated devices

35e Extensible Messaging and Presence Protocol (XMPP)e standardization of this protocol by the IETF made itpopular for usage over the Internet especially in instantmessaging and chatting platforms It is however susceptibleto spamming attacks and its failure to have worldwidesupport has made Google too withdraw its support of thestandard [17] e protocol is mainly useable in the client-server architecture but can also be applicable in client-clientXMPP communication [53] Among the application layerprotocols XMPP is the only one that supports both asyn-chronous (publishsubscribe) and synchronous (requestresponse) services while running over TCP but can still runover other protocols

Given the simplicity of its implementation and ease ofusability in XMPP instant messaging (IM) is suitable forXMPP and is implemented using the message stanzaltmessagegt ese stanzas are usable in one-to-one ormultiuser sessions or chatting or in error reporting andsending headlines Attribute type of the stanza denes themessage type e attribute type could be a chat group chatheadline normal or error An instant messaging network inXMPP can support many XMPP clients and servers with agateway congured to enable other IM network users toconnect to other XMPP users e XMPP server mainlycontrols sessions and XML streams to clients and servers inaddition to routing XML stanzas in the streams [53] eTCP ports 5269 and 5222 have been respectively registeredfor the XMPP server-to-server connections and XMPP clientconnections by the Internet Assigned Numbers Authority(IANA) XMPP enables specication of XMPP ExtensionProtocols (XEP) to enhance its usefulness [17]

XMPP security clients connect to the server usingtransport layer security (TLS) and simple authentication andsecurity layer (SASL) protocols which are at the same timeused by the servers for interdomain communications Fordata condentiality and integrity TLS will encrypt thestreams of XML for all sessions To conrm the identity of anyclient that needs to access the server the SASL protocol doesthe authentication of the XML stream For any communi-cating session with a server to start the client needs to resolvethe serverrsquos DNS hostname For protection of the server of theclient from third-party attacks the server maintains the au-thentication details secret ie IP address and access methodand only original connections of the server are required

In server-to-server communications the SASL protocolis employed for authentication and condentiality Server-to-server communications may as well be deactivated from aserver by an administrator depending on the organizationalpolicy To guard against domain-spoong attacks serverdialback can be a good solution since servers support itsecuring more the XML stanzas Nevertheless it cannot beused to authenticate secure or encrypt in-server streamsand the server identication resulting therein is not strongenough to be relied on It further cannot secure against DNSpoisoning attacks and IP session hijacking for remote do-mains [53] High-security domains can use TLS and SASL

Advantages

(i) It is popular for usage over the Internet especially ininstant messaging and chatting platforms

Disadvantages

(i) It is susceptible to spamming attacks(ii) It has failed to have worldwide support making

Google withdraw support of the standard(iii) It is not secure against DNS poisoning attacks and

IP session hijacking for remote domains

36 Advanced Message Queuing Protocol (AMQP) isprotocol can be utilized by a number of platforms formessage exchange since it is open source and written inmany languages It is a publishsubscribe model depending

MQTT-SNclients Broker

MQTT

(a)

MQTT-SNclients Broker

(b)

Figure 6 (a) Transparent gateway and (b) aggregating gateway

Journal of Computer Networks and Communications 21

on a reliable and efficient messaging queue with capacity forinstant messaging or e-mail for different applications es-pecially in IoT environments It differs from others in thesense that message types can be specified and have theirsource traceable and how to do trade-off between perfor-mance security and reliability For reliability of messagetransmission AMQPmakes use of some delivery guaranteeslike at most once at least once and exactly once It can alsohowever utilize the TCP transport layer for guaranteeingreliability -e AMQPrsquos publishsubscribe method has twocomponents namely (i) exchange queue meant for routingof messages to an appropriate order in the queue and (ii)message queue that acts as storage pending their delivery toreceivers [16] In this protocol model the broker can makerouting decisions contrary to othermessaging systems whereapps using the queue have in them embedded the decision-making logic for recipients or senders -e developer mayneed to check each of the affected apps so as to change thelogic for routing and delivery of the messages [54]

Advantages

(i) It is convenient(ii) It has capacity for interoperability between dif-

ferent vendors(iii) It can use open standards to enable connection

between business partners(iv) It can enable innovations based on AMQP foun-

dations [55](v) It can enable supporting of mission-critical ap-

plications in e-commerce(vi) It can guarantee reliable delivery of messages

Disadvantages

(i) It is not very appropriate for real-timeapplications

(ii) It may not offer automatic discovery(iii) Interoperability is not guaranteed(iv) It lacks open-source libraries for constrained

gadgets [49]

37 Data Distribution Service (DDS) A product of theObject Management Group (OMG) DDS is a pubsubprotocol meant for M2M communications with two sub-layers -ese are the data-centric pubsub sublayer re-sponsible for connecting anonymous data publishers tosubscribers and the data-local reconstruction sublayer that isoptional and integrates DDS into the application layer [56]-e DDS architecture connects system components asshown in Figure 7 Publishers and subscribers are discon-nected with respect to (i) time (ii) space (iii) flow andbehavior (iv) platforms and (v) programming languages[57] A DDS data publisher generates topics and publisher-subscriber coupling is represented as topic name data typeschema and publishers and subscribersrsquo QoS attributesDelivering of data is done by the publisher layer -e datawriter and publishers together make a decision about anynecessary alterations for subsequent forwarding to thesubscribers who need to send data to IoT applications Data

readers are responsible for delivering topics to the sub-scribers after reading through them [56]

Some of the major components of the DDS architectureinclude the following [57]

(i) Topic this is a logical path specifying publicationand subscriptionrsquos data type between DataWritersand DataReaders For a successful session to begintopic names types and DataWriters and Data-Readersrsquo QoS should correspond

(ii) Domain this is a logical setting for transmissionthat can be utilized for separating and optimizingnetwork transmission in a setup of similar appli-cations If DDS apps share a domain ID then theycan exchange data among themselves

(iii) DataWriter and DataReader these are terminalobjects for writing and reading messages re-spectively to and from a universal data space

(iv) Participant this is an object that stands for apublisher or subscriber of the DDS app in a do-main acting like a container for other objects

Advantages

(i) It is highly reliable and offers QoS support(ii) It is highly scalable(iii) It is appropriate for real-time apps(iv) It is fault tolerant(v) It has high level of interoperability(vi) It does not need brokers that may act as

bottlenecks(vii) It offers automatic discovery(viii) It has a user-defined data structure for topic(ix) It supports publisher and subscriber decoupling

Disadvantages

(i) It is originally developed for only stand-aloneLANs

(ii) A lot of memory is needed(iii) It has no open-source libraries to support con-

strained devices(iv) It is hard to design and configure compared to

other protocols [49]

38 RESTful Services REpresentational State Transfer (REST)-is is an architectural style based on a set of principles thatdefine and address networked resources It offers web servicesthat enable transmission from the HTTP device to the devicewithin the IoT architecture and enables clients to reach theserver [16]-is protocol utilizes HTTP commands to backupthe requestresponse messaging model HTTP is a popularand secure protocol in the WWW using TLSSSL [16] ARESTful application has the following characteristics

(i) State and functionality are separated into distrib-uted resources

(ii) All resources are distinctively named by the use ofunique HTTP commands (GET POST PUT orDELETE over the Internet)

22 Journal of Computer Networks and Communications

(iii) It is a clientserver stateless layered protocol thatsupports caching REST uses the same methods asHTTP for all requests or responses with POST andGET for creating and retrieving resources PUT forupdating and changing the state of resources andDELETE for removing resourcesAdvantages of REST over arbitrary web services(SOAP) in the IoT context

(i) It reduces overhead(ii) It reduces parsing complexity(iii) It is stateless(iii) It integrates more tightly with HTTP [58](iv) It is easy to implement and learnis together

with its natural t over HTTP makes it apreferred method for Web 20 applications toshowcase their data and secure M2M sessionsin the IoT

(v) Applications that support RESTful web ser-vices exhibit better performance than the re-source-constrained WSN nodes [59]

(vi) REST is important for the IoT since it supportscommercial M2M cloud platforms

(vii) Its implementation in smart phones and tabletapps is also easy since it needs only an HTTPlibrary that is accessible to all operating systemplatforms [56]

Disadvantages

(i) Although it enjoys wide usage in commercialM2M platforms it is not likely to dominatebecause it is hard to implement

(ii) Its HTTP usage makes it incompatible withconstrained M2M devices leaving its usage toend applications

39 WebSockets It is a web-based protocol operating on asingle TCP channel and providing full-duplex transmissions

It permits secure duplex communication between a clientexecuting unreliable code in a restricted environment and adistant node communicating from that code while using theorigin-based security model employed by web browsers overTCP [60] e protocol is not a requestresponse model andis not a publishsubscribe model So unlike other protocolsa web-socket session is established by the client initiating ahandshake to the server to launch the session is onceestablished a full-duplex client-to-server connection beginsasynchronously and runs until terminated by either party[56] It gives a model for browser-based apps utilizing bi-directional communication with no need of many HTTPconnections to servers

WebSocket comprises two parts ie (i) the handshakehaving a clientrsquos message and the serverrsquos handshake responseand (ii) the data transfer HTTP headers should be repeated byapplications in all client requests and server responses arisingfrom the endless polling which might increase the commu-nication overhead subject to the application [61] WebSocketcan be used to come up with scalable real-time web appli-cations since it can enable full-duplex transmission via onesocket across the Web Its security can be ensured with thehelp of TLSSSL is protocol is suitable for resource-con-strained devices and IoTapplications will not have a good userexperience with it since it operates a clientserver architectureNevertheless it can outcompete all TCP-based protocolsbecause of the following advantages

(i) It is t for real-time transmission(ii) It has improved security(iii) It minimizes transmission overhead(iv) It can provide ecient messaging systems using

WebSocket ApplicationMessaging Protocol (WAMP)[17]

310 Streaming (Simple) Text-Oriented Messaging Protocol(STOMP) is is a TCP-based protocol designed for textmessage-oriented middleware (MOM) with the use of

Topic

Datareader

Datareader

Datareader

Datawriter

Datawriter

Datawriter

Subscriber PublisherPublisher

Data bus (DDS domain)

Subscriber

Topic Topic

Figure 7 DDS architecture

Journal of Computer Networks and Communications 23

HTTP-like commands for interoperability among platformslanguages and brokers Data communication takes placebetween the client and the broker in line with commands(like CONNECT DISCONNECT ACK NACK SUB-SCRIBE UNSUBSCRIBE SEND BEGIN COMMIT orABORT) then headers in the form ltkeygt ltvaluegt (one perline) then a blank line and then body content terminatingin a null character Server-client communication takes placeover a MESSAGE RECEIPT or ERROR frame with headerand body content format being analogous MOM productssupporting STOMP include Apache ActiveMQ or FuseMessage Broker HornetQ Open Message Queue(OpenMQ) RabbitMQ and syslog-ng through its STOMPdestination plugin [62] In addition to these other appli-cation layer protocols include simple media control protocol(SMCP) lightweight local automation protocol (LLAP)simple sensor interface (SSI) lightweight M2M (LWM2M)XMPP-IOT and simple object access protocol (SOAP)

Advantages

(i) It is lightweight for constrained networks(ii) It has flexibility to choose quality of services with

the given functionality(iii) It is standardized by OASIS Technical Committee(iv) It is easy and quick to implement

Challenges and opportunities

Here security threats are imminent because of the open-endedness of the application layer [63] -ese could includethe following

(i) -ere is poor or lack of a security design for anapplicationrsquos function Coming up with a gooddesign to address this could be a good ideaopportunity

(ii) Some of the programs might have backdoorsevading otherwise secure controls to allow un-authorized access to network resources We requiremechanisms to patch up such loopholes

(iii) Applications that lack or have weak authenticationmechanisms can be easily targeted by unlicensedusers to abuse the network Such mechanisms needto be strengthened further

(iv) Excessively complicated access control rules maybecause of their misunderstanding or poor writinggive unauthorized access allowing dubious activityrather than protecting the network they purport toprotect -ese should be simple and clear to un-derstand by all users

(v) In booting workstations without disks and managingnetwork devices we usually use the TFTP protocolalthough access does not require authenticating theuser by usernamepassword -is makes an intruderable to gain easy access to configaccess info as longas he can guess the filenames -is needs to be fixed

A few steps can be used to secure the application layerTo protect network data transmitted across the network

applications need to strengthen their encryption and au-thentication mechanisms Apps too need to enable strictcontrols of privileges for data access using good mechanismsto balance between usability and effectiveness [63] Appli-cations handling sensitive data need detailed logging andaudit capability as well as testing and reviewing

Table 3 summarizes the application layer protocols dis-cussed with their weaknesses and differences among them

4 Cross-Layer QoS Strategies

41 Service-Differentiated Real-Time Communication Scheme(SDRCS) -is is an event-driven routing protocol that routesreal-time traffic by featuring a cross-layer packet-forwardingdesign in which it integrates the real-time routing functionalitywith a new priority-based MAC scheme -e protocol ap-proximates distributed packet traversal speed for traffic clas-sification and admission control on the basis of this design Itfurther localizes decision-making by prioritizing packet for-warding to maximize the speed of packets [64]

Advantages

(i) It improves bandwidth utilization bandwidthdegradation due to unschedulable packets can beavoided

(ii) It offers a lightweight packet schedulability estima-tion mechanism Here it uses received signalstrength and admission control and then earlymissed-deadline packet-dropping policies are made

(iii) It delivers high end-to-end throughput via highersource data rates alongside strict end-to-end latencyneeds

(iv) Nodes which meet real-time requirements of lesstraffic load and better channel quality are givenhighest priority when it comes to packet forwarding

-e SDRCS operates via 5 mechanisms namely (1) RSS-(Rich Site SummaryReally Simple Syndication-) basedgrouping (2) admission control (3) prioritized queuing(4) real-time MAC and (5) dynamic forwarding -isprotocol has a queuing policy that employs per-hopdeadline-based queues and the nodes schedule theirpackets using FIFO priority-based queues [1] -e prin-cipal features of this protocol are mentioned as follows

(i) It is event driven and is easily adaptable to anynetwork changes

(ii) It does not need additional hardware for locali-zation or multichannel transmission

(iii) -e cross-layer packet-forwarding strategy can beused when transmitting multimedia traffic

(iv) It is able to efficiently circumvent any voids therein

Early deadline miss (EDM) policy drops all unscheduledpackets -e SDRCS circumvents wrong packet dropsdue to EDM since it adjusts to network changes easily

42 Sensor Fuzzy-Based Image Transport (SUIT) -is pro-tocol regulates congestion using a scheme based on fuzzy

24 Journal of Computer Networks and Communications

logic It transmits packets with reduced quality to allowablelevels if there is congestion enabling it to transmit lower-quality packetsframes without being dropped-e numberof packets delivered per second thus rises and while thisenhances video streaming apps it is more appropriatelymade for those that quickly transmit non-real-time videosuch as in video surveillance SUIT can stream JPEG pic-tures other than the traditional predictive video codingmethods that are not suitable for sensor devices because oftheir constraints of energy memory and computationalspeed Progressive JPEG-PJPEG performs better via packettransmission rate and delay though it degrades the imagequality because of congestion [65] -is protocol employscross-layer methods for interlayer data exchange though itstill employs some layering making it still possible topreserve all advantages of the layered design since it hasapplication and transport layers but has noMAC or routinglayer

For congestion detection and control SUIT bases on thepercentage number of in-bound and out-bound frames perwindow number of contenders and buffer occupancy of thenext-hop node Since buffer management has an effect oneffectiveness and QoS inWMSNs it is useful in the transportlayer so the protocol prioritizes packets for better QoS SUITreorders packets by tracking the sequence numbers of thedelivered packets since they all have a source ID framenumber and sequence number that it uses to detect anymissing packets -e protocol however does not providepacket reliability [65]

Advantages

(i) -e protocol prioritizes packets for better QoS(ii) It can transmit lower-quality packetsframes

without being dropped

Disadvantages

(i) It does not provide packet reliability

43 Network Layer QoS Support Enforced by a Cross-LayerController (NLQS) -is scheme permits packet-level servicedifferentiation as a function of throughput end-to-endpacket error rate and delay [66] (see Figure 8) -is en-hances network layer QoS It comprises a cross-layer controlunit (XLCU) for configuring and controlling networkingfunctionalities at physical MAC and network layers [18]-is depends upon the unified logic which affects choices forapplication layer needs and status of functional blocks whichdo implementation of networking functions [66]

Advantages

(i) Cross-layer interactions can be controlled withoutweakening the upgradability modularity andsimplicity of designing the system

44 Cross-Layer Signaling Shortcuts (CLASS) CLASS has ahigh level of effectiveness flexibility and comprehensive-ness According to [67] it has the following features

(i) Direct signaling among remote layers leading tofaster transmission

(ii) Lightweighted internal message format standard-ized protocols are not suitable for use in internalsignaling since they are not usually lightweight forinstance transmitting against faults in the networkCLASS requires only three fields namely destina-tion address event type and event contents

(iii) -e setup of its exterior messages is standardconsidering external signaling ICMP and TCPIPare suitable for general messages and short notifi-cations respectively

(iv) Message control protocol should ensure optimizedand organized exchange of dense simultaneousmessages across layers for high efficiency andavoidance of possible conflicts -e protocol appliesto many cross-layer signaling situations

Table 3 Differences and problems with application layer protocols

Protocol Architecture Transport QoS options Security Weaknesses

CoAP Requestresponse UDP DTLS No in-built security features DTLS does not supportmulticast

MQTT Publishsubscribe TCP TLSSSL Insufficient security at the protocol level sincepayload values are not encrypted

SMQTT Publishsubscribe TCP CPKP-ABE Key revocation and group pubsub for distributedSMQTT are still a challenge

MQTT-SN Publishsubscribe UDP TLSSSL SSLTLS suffers from attacks like BEAST CRIMERC4 and Heartbleed

XMPP RequestresponsePublishsubscribe TCP X TLSSSL Susceptible to spamming attacks and lacks worldwide

supportAMQP Publishsubscribe TCP TLSSSL Inappropriate for real-time applications

DDS Publishsubscribe TCPUDP TLSSSL Memory intensive and no open-source libraries forconstrained devices

REST Requestresponse HTTP X HTTPS Hard to implement Uses HTTP so incompatiblewith constrained apps

WebSockets PublishsubscribeClientserver TCP X TLSSSL Bad user experience for IoT apps since it runs the

clientserver architectureSTOMP Clientserver TCP HTTP -e broker can act as a bottleneck

Journal of Computer Networks and Communications 25

Advantages

(i) e protocol is scalable and has a very lowpropagation delay

(ii) It is very ecient and exible since it uses directsignaling between any two layers

(iii) It is very comprehensive and eective

45 Cross-Layer and Multipath-Based Video Transmission(CMVT) CMVT is a hybrid of application and networklayers At the application layer it carries out the encoding ofvideo streams to video data frames (I- P- and B-frames) bymeans of the MPEG-4 encoding format At the networklayer route discovery and data transfer take place Twoalgorithms greedy forwarding and rollback are used todiscover a number of paths from the source to the sink nodevia the route discovery method [20] A specied node icalculates the evaluation of its adjacent node j using thefollowing equation

fij (1 minus α)d2(j D) minus d2min(i)d2max(i) minus d

2min(i)

+ αeinit(j) minus eres(j)

einit(j)

(19)

where fij evaluation value for the node i to j d2(j D) isthe distance from the node j to the destination node D

d2min(i) and d2max(i) are the minimum and maximum dis-

tances of neighbors of the node i to D respectively einit(j) isthe initial energy of the node j eres(j) is the current residualenergy of the node j and α is the energy coecient given asfollows

α emax(i) minus emin(i)

emax(i) (20)

where emax(i) and emin(i) are respectively the maximumand minimum energy left of all node irsquos neighbors

e network layer further transmits video streams withCMVT doing status evaluation choosing an appropriatecommunication path for any packets and the QoS guaranteelevel for the path i is calculated as

fi (1 minus ω)hisum hi

+ ωnisum ni

(21)

where fi is the evaluation value for the path i hi are the hopsfor the path i ni is the summation of packets sent via isum ni isthe summation of packets sent by sources and ω is theenergy consumption factor [18]

Advantages

(i) e CMVT protocol is very superior in mediatransmission particularly large-scale WMSNs

To network peers

Flow requirements

-

Atimes

Admissioncontrol

Capacityassignment

Next-hopselection

Coding rate

Data rate

QoScontracts

Cross-layer controlunit (XLCU)

Localization subsystem

Synchronizationsubsystem

Networkcoordinationunit (NCU)

UWB modulator

Channel coder

QoS scheduler

Georouting

Source coder

Rate control

Figure 8 Cross-layer controller architecture

26 Journal of Computer Networks and Communications

46 Cross-Layer Cooperative MAC (CoopMAC)CoopMAC [68] is based on the IEEE 80211 DCF approach-ere is a reduction in interference of neighboring cellsleading to uniform coverage in densely deployed networks[68] A connection is established by two-way authenticationbetween the sender and the receiver When a Clear-to-Send(CTS) packet and short interframe space (SIFS) are obtainedby the sender packets are transmitted right to the receiverwith no valuable cooperation If available the two nodes findout whether they can exchange data -is is done by the useof a helper identification signal to establish how feasiblecooperative communication is It is established in thepresence of a signal and if it is not available direct com-munication is activated For distributed systems helper-initiated cooperation is preferable because of the presence ofthe RTSCTS packets [69]

Advantages

(i) CoopMAC has an advantage over others due to thespatial diversity amongst the three nodes [70]

(ii) It has the potential to achieve significantthroughput and delay with no complexity in thedesign of the system

47 Cooperative MAC Protocol for Multihop Networks (M-CMAC) Just as CoopMAC M-CMAC is made in such away that high data rate stations help low data rate stations toforward traffic for broadcasting [71] -ere are also helperschosen in such a way that 2 fast-hop transmissions replace aslow hop transmission -e helper via whom there existsleast delay from the sender to the receiver is considered bestbecause of theminimum two-hop transmission rate and so isthe chosen neighbor to the duo Here the working as-sumption is that all nodes have their position coordinatesknown and thus the Euclidean distances between all pairscan be calculated and converted to the data rate for that linkAll nodes have cooperative tables (CT) of potential helperswith destination and helper MAC address Euclidean dis-tance and total distance through the helper Any sender withdata to transmit checks out for the existence of any helper inthe CT for the destination and if available forwards RTS forchannel reservation for single-hop duration M-CMAC hasgot an RTS format having a shape with five fields similar tothe one given as follows

Framecontrol Duration Source

addressDestination

addressHelperaddress

-e destination nodersquos address is kept by the source in thepart for the helper address that is in turn saved as the desti-nation address Nodes inspect helper and destination addressfields on receipt of RTS whereby the node acts as the othernodersquos helper in case the helper address field is not the sameand the helper transmits CTS back to the source if at all it wantsto forward data On delivery of the CTS packet the sourcetransmits to the helper which in turn sends the packets to thedestination that also sends ACK to the helper on receipt of thepacket In M-CMAC there exists a higher level of channelreuse (parallel transmissions) due to the increased number of

nodes therein that subsequently raises availability of helpers fordata forwarding -is results in increased throughput incomparison with CoopMAC and IEEE 80211 DCF [71]

Advantages

(i) It promises a higher throughput compared toCoopMAC and IEEE 80211 DCF

48 Cluster-Based Cooperative Routing (CBCR) Protocol-e CBCR protocol has a multihop data-forwarding func-tion realized at the link layer with cooperative links that useM-CMAC -is protocol encompasses two stages namelythe routing relay selection phase and data forwarding phase

(i) Routing relay selection phase all nodes announcetheir presence to their neighbors by broadcastingperiodical beacon messages that carry the MAC ad-dress of the node Each of these constructs a relaytable that contains all neighboring nodes with whichit is able to communicate -e node further broad-casts its neighbor list in case of any change to itsentries since the previous broadcast -e MAC ad-dresses of nodes next to the node X are contained inits relay tablersquos column one and the neighbor nodersquosrow has MAC addresses for neighbors of the adjacentnode Based on its relay table each node in-dependently chooses routing relays Choosing a nodeas a relay node depends on the number of nodes itconnectsmdashit should be the highest number [71]

(ii) Data forwarding phase a node with packets totransmit needs to first verify if the receiver is in thesame cluster and if that receiver has a helper to whichthe packet is thence forwarded but in whose absencethe packet is delivered right to the receiver In casethe intended receiver is in a different cluster therelay table will be checked to see whether that re-ceiver is reachable via other routing relays If it isreachable packets will be sent to the routing relaydirectly or via the helper if at all the relay has one Incase of destination unreachable via relays packetsare multicast to all relays by the node [71]Advantages

(i) It enables multicasting

49 MAC-PHY Cross-Layer Protocol A cooperative cross-layer standard for cooperation at the physical layer in next-generation WMSNs is developed in [70] It provides acomplete MAC layer algorithm that gives a supportive shellfor the PHY-MAC layer Like CoopMAC its scheme de-pends on the middle node for communication betweennodes-eMAC layer protocol has been changed to regulateinformation exchanges at the physical layer -e receivernode receives duplicate packets from source and helpernodes to decode the data [70]

410 MAC-Centric Approach -is cross-layer protocoltargets multimedia applications by using the MPEG-4

Journal of Computer Networks and Communications 27

scheme [21] It is characterized by 4 access categories(AC3ndashAC0) according to their priority in transmission It ismeant for supporting different QoS needs in upcomingvideo applications and enabling MAC layer differentiationfor H264 partitioning [18] Because of low bandwidth delayand other QoS challenges that cause inefficiency in trans-mittingmultimedia data inWMSNs a number of algorithmsdepending on IEEE 80211e have been proposed to supporttransmission of quality videos [72] Selecting an AC dependson the measures for QoS such as the loss rate and delay-us AC3 having the highest priority is mapped to theparameter set concept since the stream is sensitive to bit lossespecially in video transmission [21]

411 Adaptive Cross-Layer Forward Error Correction(ACFEC) In the ACFEC [73] model data packets are ex-changed among nodes through the access point (AP) whichoperates the infrastructure mode in which it adds FEC to videodata [72] -ese data are dealt by encapsulation using astreaming server to the receiver through thewireless AP as RTPpackets An adaptive FEC controller senses the category ofpackets out of the RTP header and recovers the header ofpackets from UDP -e encoder generates some error-cor-recting packets whose number is determined by the sourcepacket number of the block Multimedia transmissions aremonitored by the controller using MAC failure data and itscounter is increased by one if there is a failed transmission-econtroller uses the failure counter to change the packet numberproduced after transmitting a block [18] In case there are lostpackets it changes the redundancy rates and produces extrapackets to replace the lost ones and satisfy the requirements ofthe receiving node [72] -e FEC packet number is enhancedor reduced to satisfy the receiverrsquos requirements and stoppacket losses-is is done through accurate detection of packetlosses and adjustment of redundancy rates If all video datapackets are well received there will not be generation of FECpackets [73]

Advantages

(i) It promises a good QoS through packet loss re-duction and redundancy rate adjustment

412 Balanced Cross-Layer Fuzzy Logic (BCFL) DesignRouting Algorithm in WSNs A new fuzzy logic-basedrouting algorithm (BCFL) [74] was designed using disper-sion of the cross-layer parameter as the fuzzy logic inferencesystem input Each cross-layer parameter has a dynamicweight depending on the value of dispersion -e designcomes with some innovations as follows (i) for fuzzy logicinference system input the absolute parameter value issubstituted by parameter dispersion thereby significantlyreducing algorithmic complexity (ii) dispersion does notchange with the order of magnitude according to the dis-persion formula and (iii) the weight of the parameter de-pends on the size of its dispersion and the two are inverselyproportional to each other -is enables BCFL to have somenobility unlike other algorithms as per the following dis-tinguishing properties making it advantageous

(i) It has got simple if-then rules which remain con-stant even when the constraints increase

(ii) It is capable of dealing with many constraints withno increment in complexity

(iii) It has capacity to yield a more balanced solutionthan other algorithms

(iv) It is easily adaptable to changes in network con-ditions and topology even when the changes arefrequent like in underwater WSNs -e algorithm isuseable in choosing the CH in cluster-based routingprotocols [74]

413 Minimum Hop Disjoint Multipath Routing Algorithmwith Time Slice Load-Balancing Congestion Control SchemeMHDMwTS is a two-phased routing protocol comprisingpath build-up and path acknowledgment phases It consists ofmultiple sources each of which has three build-up disjointpaths namely primary alternate and back-up paths Onactivation the source node requests to build up a route to thenearest hop neighbor which is the path build-up phase [19] Inthis phase step one has source activation in which therequested node for path building adds its number and time-stamp and sends to the least hop-count neighboring node-isgoes on until the least time latency sink having the informationneeded to construct the primary route is reached [75] In steptwo there is path extraction when the new package fromanother path arrives whereby the extracted path is comparedwith the primary path If the node is shared the package will berejected or an alternative path will be searched for to get abackup through comparison of the preceding two paths Phasetwo is path acknowledgment where we have the third step inwhich the sink returns the ACK packet to the sender with pathinformation having nodes and their time information after ithas computed it using the timestamp [1819]

Advantages

(i) -e protocol minimizes end-to-end latency(ii) It enhances congestion control

414 Cross-Layer Optimal Design (CLOD) Authors in [76]come up with a CLOD for scheduling at the data link layerrouting at the network layer and controlling congestion atthe transport layer with an assumption of fixed link capacity-rough congestion control energy efficiency is improvedTransport layer congestion at the nodes is minimized bycompressed sensing (CS) in which transmitted bits are re-duced whereas optimally allocating resources decreasescongestion on the links at the data link layer CLODpromises minimized computational complexity and betterperformance in light traffic scenarios [77] Generating andstoring CLOD is more efficient compared to Gaussianrandom matrices It prolongs network lifetime and savesenergy

Advantages

(i) It does congestion control subsequently increasingenergy efficiency

28 Journal of Computer Networks and Communications

(ii) It reduces computational complexity(iii) It improves performance in light traffic scenarios

415 Challenges and Opportunities of the Cross-Layer Design(CLD)

(i) Different layers in a real network assume differentfunctions andor services A layer only communi-cates with its neighboring layers -e layered modeldegrades system performance because of many

features in wireless transmission QoS parameters inthe layers of WMSNs are shared by routing pro-tocols to optimize the performance Neverthelessthere is a need to design cross-layer models forincreased efficiency in routing [1]

(ii) -e physical layer plays a highly vital role in CLDRate adaptation and channel allocation take place atthe physical layer via signal processing to enhanceQoS End-to-end performance in wireless media isaffected by the changes therein given the effect on

Table 4 A summary of some of the reviewed cross-layer models

Protocol Layers Aims Comments QoS parameters

SDRCS MACPHY Routing real-time traffic Transmits multimedia traffic viacross-layer packet forwarding

-roughputlatency

SUIT Applicationtransport Transmission of non-real-time video Uses fuzzy logic to regulate

congestionTransmission rate

delay

CLASS Any twoDesign serves as a framework fordifferent implementations ofdifferent application scenarios

Scalable very efficient flexible andhas very low propagation delay

Propagation delayjitter

NLQS PHYMACnetwork

Permits packet-level servicedifferentiation as a function of

throughput packet error rate anddelay

Network layer QoS is enhanced -roughput delaypacket error rate

CMVT Applicationnetwork

Encodes video streams to video dataframes via MPEG-4 encoding anddoes route discovery and data

transfer

Greedy forwarding and rollback areused to find source-to-sink paths Energy

CoopMAC PHYMAC Offers spatial diversity among thethree nodes

Helper ID signal finds out howfeasible cooperative communication

is

-roughput ratedelay

M-CMAC PHYMAC Increases end-to-end throughputand packet delivery ratio

Euclidean distances between nodesare calculated and converted to the

data rate for a link

-roughputpacket delivery

ratio

CBCR PHYMACMinimizes control overhead andtime consumed in establishing thecooperative paths than M-CMAC

Energy consumption is moreuniformly distributed in a network

enhancing network lifetime

-roughputpacket deliveryratio energy

MAC-PHY MACPHYMeant to enable PHY layer

cooperation and maximize gains ofcooperation at the MAC layer

Leverages both spatial diversity andcoding gain -roughput delay

MAC-centric MACAPPL

Meant to support QoS needs in newvideo apps and enable MAC layer

differentiation for H264partitioning

Targets multimedia applications byusing the MPEG-4 scheme

Delay packet lossrate

ACFEC MACnetwork(UDP)

Meant to enhance the quality ofvideo streaming over 80211WLANs

and overcome packet losses

Adjusts redundancy rates toovercome channel fluctuations and

detect and reduce packet lossPacket loss rate

MHDMwTS mdash

Meant to provide reliable datatransfer with the multipath routingand load-balancing congestioncontrol method in WMSNs

More reliable than basic routingschemes for transport multimedia

data

Latency packagetransmit rate

BCFL mdash

Introduces dispersion into fuzzylogic-based routing and sets every

cross-layer parameter with adynamic weight

Can be used to select a CH in cluster-based routing protocols andproposes a dispersion formula

Node utilitydispersion

CLODDatalinknetworktransport

Prolongs network lifetime andachieves congestion control anddesigned for lightly loaded WSNs

Assumes fixed link capacity andintegrates compressed sensing

technology

-roughputaverage energy CS

error ratio

Journal of Computer Networks and Communications 29

the purposefulness of the protocols at the networklayer CLD offers solutions to conserving powerminimizing energy and controlling both the flowand congestion in the network as well as fault tol-erance so it is an opportunity for designers con-sidering other layers We further desire to developCLDs that shall make an impact on network oper-ation and get close attention [18]

(iii) CLD interface standardization the architecturemust deliver module functionality although thereexist some queries amongst modules about possibleinterfaces predicted by the necessity to share in-formation among remote protocol layers Technicalchallenges include developing designing andstandardizing of cross-layer interfaces and algo-rithms that satisfy the cross-layer optimization re-quirements among protocol layers

Table 4 summarizes these cross-layer models with a fewcomments

5 Future Research Direction

51 At theMACLayer In WMSNs achieving duty cycling istricky since video traffic is volatile Listening to adjacentnodes wastes energy We need to switch states from awake tosleep -is requires further research

Self-organizing networks and green communicationalthough most CRWSN energy-saving algorithms aredesigned to reduce node transmission power energyconsumption is inevitable during the operational mode ofBSs because of internal processing Scholars are trying withpolicy-makers to encourage a shift to greenWSNs to reduceoperational costs and carbon footprints With self-orga-nizing CRWSNs for automatic switch-off when idle energyis saved which calls for cognitive switching algorithms Sointegrating cognitive switching with radio resource allo-cation in CRWSNs is a promising research area [78]

52 At the Application Layer Considering a WMSN ar-chitecture end users access the network via the BS to andfrom which they send their sensed data We desire to getguaranteed QoS and energy efficiency by designing a newarchitecture in the form of hardware and software whichneeds to be investigated further for a fix of the same [1]

QoS and energy efficiency are very crucial in WMSNsand mostly real-time apps that require guaranteed band-width and throughput in their network lifetime Mostprotocols ignore base station (BS) mobility and WMSNnodes Traffic management telemedicine and battlefieldsurveillance apps require mobile nodes or BSs so designingdynamic routing protocols to be adaptable in these cir-cumstances is necessary

53 Cross-Layer Security of data transmitted on WMSNseg in military surveillance and e-commerce is key withQoS and energy efficiency WMSNs are vulnerable to attacks

like worms and sinkhole Moreover their computationpower is not that high complicating implementation ofstrong secure protocols -is needs further attention

For fairness and priority issues in CRWSNs nodes areassigned diverse priorities depending on importance andurgency of their data For better network performance thereis a need to develop fair resource allocation mechanismswith priority-based fairness in sensors-ere is still potentialin this research area [78]

We must develop multiobjective adaptive protocols tooptimize QoS metrics involved in routing for best trans-mission results to get a trade-off between different opti-mization metrics We need to design standardized cross-layer algorithms which are able to meet cross-layer opti-mization standards

6 Conclusion

We have reviewed the different QoS strategies for WSNs inthe context of IoT from the MAC layer and application layeras well as the cross-layer paradigm For the MAC layer wehave reviewed protocols for WSNs such as the hybrid of thecontention-free and contention-based MAC protocols andthose for WMSNs We further reviewed a number of ap-plication layer protocols including many machine-to-ma-chine requestresponse and publishsubscribe protocols Forsystem optimization cross-layer QoS strategies are veryimportant in wireless communication We have reviewed anumber of cross-layer strategies For all categories reviewedchallenges and opportunities are discussed Finally somepossible future directions are discussed for research andapplication showing a promising potential for future re-search in this relatively new area especially as more WMSNapplications emerge recently

Conflicts of Interest

-e authors hereby declare no conflicts of interest regardingthe publication of this work

Acknowledgments

-is research was funded by the National Natural ScienceFoundation of China (Nos 61602398 61672447 and61711540306)

References

[1] V Bhandary A Malik and S Kumar ldquoRouting in wirelessmultimedia sensor networks a survey of existing protocolsand open research issuesrdquo Journal of Engineering vol 2016Article ID 9608757 27 pages 2016

[2] I F Akyildiz T Melodia and K R Chowdhury ldquoWirelessmultimedia sensor networks applications and testbedsrdquoProceedings of the IEEE vol 96 no 10 pp 1588ndash1605 2008

[3] M Iqbal M Naeem A Ahmed M Awais A Anpalagan andA Ahmad ldquoSwarm intelligence based resource managementfor cooperative cognitive radio network in smart hospitalsrdquoWireless Personal Communications vol 98 no 1 pp 571ndash5922018

30 Journal of Computer Networks and Communications

[4] S A Alvi B Afzal G A Shah L Atzori and W MahmoodldquoInternet of multimedia things vision and challengesrdquo AdHoc Networks vol 33 pp 87ndash111 2015

[5] Q F Hassan A U R Khan and S A Madani Internet ofings Challenges Advances and Applications CRC PressTaylor amp Francis Group A Chapman amp Hall Group BocaRaton FL USA 2018

[6] Q Wang Y Zhao W Wang et al ldquoMultimedia IoT systemsand applicationsrdquo in Proceedings of the 2017 Global Internet ofings Summit (GIoTS) vol 2 Geneva Switzerland June 2017

[7] A Aslam and E Curry ldquoTowards a generalized approach fordeep neural network based event processing for the internet ofmultimedia thingsrdquo IEEE Access vol 6 pp 25573ndash25587 2018

[8] M Jridi T Chapel V Dorez L Bougeant and A Le BotlanldquoSoC-based edge computing gateway in the context of theinternet of multimedia things experimental platformrdquoJournal of Low Power Electronics and Applications vol 8no 1 p 1 2018

[9] H Noura A Chehab L Sleem andM N Rapha ldquoOne roundcipher algorithm for multimedia IoT devicesrdquo MultimediaTools and Applications vol 77 no 14 pp 18383ndash18413 2018

[10] T Balan D Robu and F Sandu ldquoMultihoming for mobileinternet of multimedia thingsrdquo Mobile Information Systemsvol 2017 Article ID 6965028 16 pages 2017

[11] H Shen and G Bai ldquoRouting in wireless multimedia sensornetworks a survey and challenges aheadrdquo Journal of Networkand Computer Applications vol 71 pp 30ndash49 2016

[12] A Ahmad M H Rehmani H Tembine O A Mohammedand A Jamalipour ldquoIEEE access special section Editorial optimization for emerging wireless Networks IoT 5G andsmart grid communication networksrdquo IEEE Access vol 5pp 2096ndash2100 2017

[13] B A Muzakkari M A Mohamed M F A Kadir andZ Mohamad ldquoRecent advances in energy efficient-QoS awareMAC protocols for wireless sensor networksrdquo InternationalJournal of Advanced Computer Research vol 8 no 38pp 212ndash228 2018

[14] M A Yigitel O D Incel and C Ersoy ldquoQoS-aware MACprotocols for wireless sensor networks a surveyrdquo ComputerNetworks vol 55 no 8 pp 1982ndash2004 2011

[15] A M Abbas and O Kure ldquoQuality of Service in mobile ad hocnetworks a surveyrdquo International Journal of Ad Hoc andUbiquitous Computing vol 6 no 2 p 75 2010

[16] M Q Shatnawi ldquoApplication layer protocols for the internetof things a surveyrdquo in Proceedings of the InternationalConference on Engineering amp MIS (ICEMIS) Agadir Mo-rocco September 2016

[17] V Karagiannis P Chatzimisios F Vazquez-gallego andJ Alonso-zarate ldquoA survey on application layer protocols forthe internet of things research motivationrdquo InternationalJournal of Advanced Research in Computer Science vol 8no 3 pp 1ndash10 2015

[18] M S Bernard T Pei Z Li and K Li ldquoQoS strategies forwireless multimedia sensor networks in the context of IoTrdquoLecture Notes of the Institute for Computer Sciences SocialInformatics and Telecommunications Engineering vol 4pp 228ndash253 2019

[19] A AlAmri and M Abdullah ldquoCross-layer quality of serviceprotocols for wireless multimedia sensor networksrdquo in Pro-ceedings of the International Conference on CommunicationManagement and Information Technology pp 649ndash658Cosenza Italy April 2017

[20] J Guo L Sun and R Wang ldquoA cross-layer and multipathbased video transmission scheme for wireless multimedia

sensor networksrdquo Journal of Networks vol 7 no 9pp 1334ndash1340 2012

[21] A Ksentini A Naimi and M Gueroui ldquoToward an im-provement of H264 video transmission over IEEE 80211ethrough a cross-layer architecturerdquo IEEE CommunicationsMagazine vol 44 no 1 pp 107ndash114 2006

[22] W Ye J Heidemann and D Estrin ldquoAn energy-efficientMAC protocol for wireless sensor networksrdquo in Proceedings ofthe 21st International Annual Joint Conference of the IEEEComputer and Communications Societies (INFOCOM 2002)New York NY USA June 2002

[23] J Polastre J Hill and D Culler ldquoVersatile low power mediaaccess for wireless sensor networks categories and subjectdescriptorsrdquo in Proceedings of the Second InternationalConference on Embedded Networked Sensor Systems Sen-Sysrsquo04 pp 95ndash107 Baltimore MA USA November 2004

[24] T Danmanee K N Nakorn and K Rojviboonchai ldquoCU-MAC a duty-cycle MAC protocol for internet of things inwireless sensor networksrdquo ECTI Transactions on ElectricalEngineering Electronics and Communications vol 16 no 2pp 30ndash43 2018

[25] T Van Dam and K Langendoen ldquoAn adaptive energy-efficientMAC protocol for wireless sensor networksrdquo in Proceedings ofthe 1st International Conference on Embedded NetworkedSensor Systems Los Angeles CA USA November 2003

[26] P Xie and J Cui ldquoR-MAC an energy-efficient MAC protocolfor underwater sensor networksrdquo in International Conferenceon Wireless Algorithms Systems and Applications (WASA2007) Chicago IL USA August 2007

[27] A Rahim N Javaid M Aslam U Qasim and Z A KhanldquoAdaptive-reliable medium access control protocol forwireless body area networksrdquo in Proceedings of the AnnualIEEE Communications Society Conference on Sensor Meshand Ad Hoc Communications and Networks (SECON) SeoulKorea June 2012

[28] H S Kwak S Ullah D H Kwak and C H Lee ldquoA powerefficient MAC protocol for wireless body area networksrdquoJournal of the Korean Institute of Intelligent Systems vol 8no 6 pp 131ndash140 2009

[29] C Omeni A J Burdett and M Park ldquoEnergy efficientmedium access protocol for wireless medical body area sensornetworksrdquo in Proceedings of the 4th IEEEEMBS InternationalSummer School and Symposium on Medical Devices andBiosensors pp 29ndash32 Cambridge UK August 2007

[30] L Sitanayah C J Sreenan and K N Brown ldquoER-MAC ahybrid MAC protocol for emergency response wireless sensornetworksrdquo in Proceedings of the Fourth International Con-ference on Sensor Technologies and Applications pp 244ndash249Mestre Italy July 2010

[31] Y Sun and D B Johnson ldquoRI-MAC a receiver-initiatedasynchronous duty cycle MAC protocol for dynamic trafficloads in wireless sensor networksrdquo in Proceedings of the 6thACM conference on Embedded network sensorsystemsmdashSenSysrsquo08 pp 1ndash14 Raleigh NC USA November2008

[32] M Buettner G V Yee E Anderson and R Han ldquoX-MAC ashort preamble MAC protocol for duty-cycled wireless sensornetworksrdquo in Proceedings of the 4th International Conferenceon Embedded Networked Sensor Systems SenSys 2006pp 307ndash320 Boulder Colorado USA October 2006

[33] S Boulfekhar and M Benmohammed ldquoSynchronous receiverinitiated MAC protocol for long-lived sensor networksrdquoComputers amp Electrical Engineering vol 40 no 2 pp 504ndash516 2014

Journal of Computer Networks and Communications 31

[34] S Mehta and K S Kwak ldquoH-MAC a hybrid Mac protocol forwireless sensor networksrdquo International Journal of ComputerNetworks amp Communications vol 2 no 2 pp 108ndash117 2010

[35] I Rhee A Warrier M Aia J Min and M L Sichitiu ldquoZ-MAC a hybrid MAC for wireless sensor networksrdquo IEEEACM Transactions on Networking vol 16 no 3 pp 511ndash5242008

[36] S Arshad A Al-sadi and A Barnawi ldquoZ-MAC performanceevaluation and enhancementsrdquo Procedia Computer Sciencevol 21 pp 485ndash490 2013

[37] Y Liu I Elhanany and H Qi ldquoAn energy-efficient QoS-aware media access control protocol for wireless sensornetworksrdquo in Proceedings of the IEEE International Confer-ence on Mobile ad hoc and Sensor Systems pp 9ndash11 Wash-ington DC USA December 2005

[38] J Decotignie ldquoWiseMAC an ultra low power MAC protocolfor the downlink of infrastructure wireless sensor networksrdquoin Proceedings of the Ninth International Symposium onComputers And Communications pp 244ndash251 AlexandriaEgypt June 2004

[39] R A Rashid ldquoDevelopment of energy aware TDMA-basedMAC protocol for wireless sensor network systemrdquo EuropeanJournal of Scientific Research vol 30 no 4 pp 571ndash578 2009

[40] H P Van Hoesel ldquoA lightweight medium access protocol(LMAC) for wireless sensor networksrdquo in Proceedings of theInternational Workshop on Networked Sensing Systems Bal-timore MD USA November 2004

[41] C Li P Wang H Chen andM Guizani ldquoA cluster based on-demand multi-channel MAC protocol for wireless multi-media sensor networksrdquo in Proceedings of the IEEE In-ternational Conference on Communications pp 2371ndash2376Beijing China May 2008

[42] J L Sobrinho and A S Krishnakumar ldquoQuality-of-service inad hoc carrier sense multiple access wireless networksrdquo IEEEJournal on Selected Areas in Communications vol 17 no 8pp 1353ndash1368 1999

[43] J P John ldquoProviding MAC QoS for multimedia traffic in80211e based multi-hop ad hoc wireless networksrdquo ComputerNetworks vol 51 no 1 pp 153ndash176 2007

[44] S Lofty and N Padmavati ldquoA survey on mobility basedprotocols in WSNsrdquo in Proceedings of the InternationalConference Communication and Manufacturing pp 12ndash15Hangzhou China July 2014

[45] S Deng L Shen and J Li ldquoMobility-based clustering pro-tocol for wireless sensor networks with mobile nodesrdquo IETWireless Sensor Systems vol 1 no 1 pp 39ndash47 2011

[46] M Khalil A Khalid F U Khan and A Shabbir ldquoA review ofrouting protocol selection for wireless sensor networks insmart citiesrdquo in Proceedings of the 24th Asia-Pacific Confer-ence on Communications (APCC) Ningbo China November2014

[47] O Message Q Telemetry and T Mqtt Committee Specifi-cation Draft 02 April 2014

[48] M Collina M Bartolucci A Vanelli-coralli andG E Corazza ldquoInternet of things application layer protocolanalysis over error and delay prone linksrdquo in Proceedings ofthe 7th Advanced Satellite Multimedia Systems Conference andthe 13th Signal Processing for Space CommunicationsWorkshop Livorno Italy September 2014

[49] A Talaminos-barroso M A Estudillo-valderrama L M RoaJ Reina-tosina and F Ortega-ruiz ldquoA machine-to-machineprotocol benchmark for e health applicationsmdashuse case re-spiratory rehabilitationrdquo Computer Methods and Programs inBiomedicine vol 129 pp 1ndash11 2016

[50] V Goyal O Pandey A Sahai and B Waters ldquoAttribute-based encryption for fine-grained access control of encrypteddatardquo in Proceedings of the 13th ACM Conference on Com-puter and Communications Security pp 89ndash98 AlexandriaVA USA October 2006

[51] M Singh M A Rajan V L Shivraj and P BalamuralidharldquoSecure MQTT for internet of things (IoT)rdquo in Proceedings ofthe Fifth International Conference on Communication Systemsand Network Technologies Gwalior India April 2015

[52] A Stanford-Clark and H L Truong MQTT for Sensor Net-works (MQTT-SN) Protocol Specification httpmqttorgnewwp-contentuploads200906MQTT-SN_spec_v12pdf 2013

[53] RFC ldquoExtensible messaging and presence protocol (XMPP)rdquoin Proceedings of the IETF pp 1ndash9 San Diego CA USAAugust 2005

[54] S Vinoski and I Technologies ldquoQueuing protocolrdquo IEEEInternet Computing vol 10 no 6 pp 87ndash89 2006

[55] J L Fernandes I C Lopes J J P C Rodrigues and S UllahldquoPerformance evaluation of restful web services and AMQPprotocolrdquo in Proceedings of the Fifth International Conferenceon Ubiquitous and Future Networks (ICUFN) pp 2ndash7 DaNang Vietnam July 2013

[56] M Asim ldquoA survey on application layer protocols for internetof things (IoT)rdquo International Journal of Advanced Researchin Computer Science vol 8 no 3 pp 996ndash1000 2017

[57] K An A Gokhale D Schmidt S Tambe P Pazandak andG Pardo-castellote ldquoContent-based filtering discovery pro-tocol (CFDP) scalable and efficient OMG DDS discoveryprotocolrdquo in Proceedings of the 8th ACM InternationalConference on Distributed Event-Based Systems New YorkNY USA July 2014

[58] E Web What is a web service pp 52ndash57 2010[59] M Laine RESTful Web Services for the Internet of ings

pp 2ndash4 2012 httpmediaTkk[60] IETF -e WebSocket protocol pp 1ndash71 2011[61] V Pimentel ldquoCommunicating and displaying real-time data

with websocketrdquo IEEE Internet Computing vol 16 no 4pp 45ndash53 2012

[62] Wikipedia 2019 httpsenwikipediaorgwikiStreaming_Text_Oriented_Messaging_Protocol2019

[63] K Raghavendra and S Nireshwalya ldquoApplication layer se-curity issues and its solutionsrdquo IJCSET vol 2 no 6pp 1266ndash1269 2012

[64] Y Xue B Ramamurthy andM C Vuran ldquoSDRCS a service-differentiated real-time communication scheme for eventsensing in wireless sensor networksrdquo Computer Networksvol 55 no 15 pp 3287ndash3302 2011

[65] C Sonmez S Isik M Y Donmez O D Incel and C ErsoyldquoSUIT a cross layer image transport protocol with fuzzy logicbased congestion control for wireless multimedia sensornetworksrdquo in Proceedings of the 5th International Conferenceon New Technologies Mobility and Security (NTMS) pp 1ndash6IEEE Istanbul Turkey May 2012

[66] T Melodia and I Akyildiz ldquoCross-layer QoS-aware com-munication for ultra wide band wireless multimedia sensornetworksrdquo IEEE Journal on Selected Areas in Communica-tions vol 28 no 5 pp 653ndash663 2010

[67] Q Wang and M A Abu-Rgheff ldquoCross-layer signalling fornext-generation wireless systemsrdquo in Proceedings of the 2003IEEE Wireless Communications and Networking 2003WCNC 2003 vol 2 pp 1084ndash1089 IEEE New Orleans LAUSA March 2003

[68] P Liu Z Tao S Narayanan T Korakis and S PanwarldquoCoopMAC a cooperative MAC for wireless LANsrdquo IEEE

32 Journal of Computer Networks and Communications

Journal on Selected Areas in Communications vol 25 no 2pp 340ndash354 2007

[69] H Shan H T Cheng and W Zhuang ldquoCross-layer co-operative MAC protocol in distributed wireless networksrdquoIEEE Transactions onWireless Communications vol 10 no 8pp 2603ndash2615 2011

[70] F Liu T Korakis Z Tao and S Panwar ldquoAMAC-PHY cross-layer protocol for wireless Ad-Hoc networksrdquo in Proceedingsof the 2008 IEEE Wireless Communications and NetworkingConference pp 1792ndash1797 IEEE Las Vegas NV USA March2008

[71] L Jacob and H R Shamna ldquoEfficient cooperative MAC androuting in wireless networksrdquo Transactions on Networks andCommunications vol 3 no 5 2015

[72] S Rao and K Shama ldquoCross layer protocols for multimediatransmission in wireless networksrdquo International Journal ofComputer Science amp Engineering Survey vol 3 no 3pp 15ndash28 2012

[73] L Han S Park S Kang and H Peter ldquoAn adaptive cross-layer FEC mechanism for video transmission over 80211WLANsrdquo in Proceedings of the 2009 International Conferenceon Internet pp 209ndash215 Las Vegas NV USA December2009

[74] N Li J Martınez and V H Dıaz ldquo-e balanced cross-layerdesign routing algorithm in wireless sensor networks usingfuzzy logicrdquo Sensors vol 15 no 8 pp 19541ndash19559 2015

[75] G Sun J Qi Z Zang and Q Xu ldquoA reliable multipathrouting algorithm with related congestion control scheme inwireless multimedia sensor networksrdquo in Proceedings of the2011 3rd International Conference on Computer Research andDevelopment vol 4 pp 229ndash233 IEEE Shanghai ChinaMarch 2011

[76] M Li Y Jing and C Li ldquoA robust and efficient cross-layeroptimal design in wireless sensor networksrdquoWireless PersonalCommunications vol 72 no 4 pp 1889ndash1902 2013

[77] J Yan M Zhou and Z Ding ldquoRecent advances in energy-efficient routing protocols for wireless sensor networks areviewrdquo IEEE Access vol 4 pp 5673ndash5686 2016

[78] A Ahmad S Ahmad M H Rehmani and N U Hassan ldquoAsurvey on radio resource allocation in cognitive radio sensornetworksrdquo IEEE Communications Surveys amp Tutorials vol 17no 2 pp 888ndash917 2015

Journal of Computer Networks and Communications 33

International Journal of

AerospaceEngineeringHindawiwwwhindawicom Volume 2018

RoboticsJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Active and Passive Electronic Components

VLSI Design

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Shock and Vibration

Hindawiwwwhindawicom Volume 2018

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawiwwwhindawicom

Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Control Scienceand Engineering

Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Journal ofEngineeringVolume 2018

SensorsJournal of

Hindawiwwwhindawicom Volume 2018

International Journal of

RotatingMachinery

Hindawiwwwhindawicom Volume 2018

Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Navigation and Observation

International Journal of

Hindawi

wwwhindawicom Volume 2018

Advances in

Multimedia

Submit your manuscripts atwwwhindawicom

Page 14: ReviewArticle - downloads.hindawi.com · ReviewArticle QoSStrategiesforWirelessMultimediaSensorNetworksinthe ContextofIoTattheMACLayer,ApplicationLayer,and Cross-LayerAlgorithms MuwongeSsajjabbiBernard

are able to communicate without colliding-is saves energyand increases the lifetime of the network A node in L-MACcan only transmit one message per frame [13] -e networkis set up in such a way that all nodes are not synchronized onbeing switched on-e gateway takes control of a timeslot soas to have them synchronized and its immediate neighborgets the control messages thereby synchronizing their timerswith the gatewayrsquos Neighboring nodes learn of multiplegatewaysrsquo time slots in their vicinity just after one frame-ejust synchronized nodes randomly choose time slots forcontrolling as long as they are not occupied -e nodes thuspreserve a table for the neighboring nodes which makes atimeslot reusable after a minimum of 3 hops and preventsmessage collision All nodes keep their slots before thebattery is down or learn of an impending collision All nodescan track their hop distances to a selected gateway node andthe information is communicated in control messages [40]

Advantages

(i) It minimizes the number of transceiver switches(ii) It adapts the nodal sleep interval to data traffic

amount(ii) It makes implementation less complex(iii) It offers improved network lifetime(iv) Energy wastage through idle listening in un-

controlled time slots is avoided via a short time-out interval

Disadvantages

(i) L-MAC is not very good for mobile SNs since slotsare calculated only once

(ii) It suffers idle listening because nodes keep listeningto slotsrsquo control parts so as to get data and enableother nodes to join the network all the time [13]

2117 C-MAC -is was designed for WSNs with 3 seg-ments namely (i) aggressive RTS having double channelcheck to assess the channel (ii) anycast to initialize the flowand (iii) convergent packet forwarding to stabilize the flowIt promises an improvement in energy efficiency lowerlatency and enhanced throughput while avoiding syn-chronization overhead though it operates only in low duty-cycling apps especially where there is less traffic But whenthere is more traffic the protocol employs anycast-basedpacket forwarding for waking the nodes up or finding aforwarder quickly and converges from route-suboptimal toroute-optimal unicast After this energy could be preservedby the nodes using a synchronized wake-up plan [13]

Advantages

(i) It improves energy efficiency(ii) It lowers latency(iii) It enhances throughput avoiding synchronization

overhead

Disadvantages

(i) It operates only in low duty-cycling applicationsespecially where there is less traffic

22 MAC Protocols for WMSNs

221 Diff-MAC -is is a CSMACA-based QoS-awarehybrid-priority-based MAC protocol for WMSNs It targetsenhancing the channel utilization while making use of ef-ficient service differentiation to do coordination for mediumaccess of each traffic class while giving fair quick datatransfer It offers QoS using the following features [14]

(i) It balances energy consumption and delay throughsensor node duty-cycling adaptation depending onthe dominant traffic class

(ii) It has a feature for fragmentation and messagepassing that breaks lengthy video frames intosmaller packets that are transmitted as a burst thusreducing retransmission costs if there are any MACfailures

(iii) -ere is fair data delivery amongst the nodes andtraffic classes respectively resulting from intranodeand intraqueue prioritization reducing unbearableperformance

(iv) -e number of collisions can be reduced and packetlatencies are kept within limits by adjusting the sizeof the congestion window in this protocol as per thetraffic needsAdvantages

(i) Diff-MAC is very suitable for WMSNs since itadapts quickly to changes in networkconditions

(ii) It is a fair protocol(iii) It is scalable

Disadvantages

(i) It is quite difficult to monitor the statistics ofthe network

(ii) Adapting dynamically to the network condi-tions is not an easy task

(iii) It exhibits some degree of latency in packetdelivery

-e implementation of Diff-MAC on Imote2 indicatesthat resource requirements can be met on the sensorhardware Results of extensive simulation runs showed thatDiff-MAC outperforms its competitors

222 Cluster-Based On-Demand Multichannel MAC Pro-tocol for WMSN (COM-MAC) -is is an energy-efficientprotocol in WMSNs with high throughput and reliable datatransmission It achieves energy efficiency by schedulingmultichannel media access to reduce channel contentionand remove collision idle listening and overhearing [41]

Design it has a three-phased operation with the (i)request phase (ii) scheduling phase and (iii) datatransmission phaseRequest phase here all nodes have two protocols eachfor sending requests to the cluster head (CH) ie thecontention-based and contention-free TDMAFDMA

14 Journal of Computer Networks and Communications

protocols -e contention-based protocol allocates achannel per node for sending the request message andfree channels are useable as control channels Forcontention-free protocols a control slot is allocatedwhen a network is deployed Here a slot per channel isallocated to a different node enabling requests to be sentwithout interfering with transmissions from othernodes -e contention-based protocol operates undertwo steps (i) Control channel assignment phase achannel is assigned per node for transmitting a request(REQ) message A channel is assigned to multiplenodes and nodes are evenly distributed to availablechannels for congestion control (ii) Request trans-mission phase nodes with data send a REQmessage viathe assigned control channel to notify the CH whichresponds with an ACK message -e contention-freeTDMAFDMA protocol too has two phases includingthe control slot assignment and request transmissionphases For unreliable channel conditions and lowtraffic load the contention-based protocol is fit for usebecause of less collisions and congestion whereas thecontention-free protocol is preferable in heavy trafficload and reliable channel applicationsScheduling phase from the above phase a schedule isgenerated by the CH for coordination and this enablesnodes to transmit and broadcast their data via controlchannels-e schedule for broadcasting only comprisesdata for nodes assigned to a channel -e schedule isrebroadcast for transmission reliability enhancementbut for energy efficiency the nodersquos transceiver isswitched off when all the schedule is completely de-livered till it is the time for transmitting again-e timeslot and data communication radio channel are allincluded in the schedule generated for each nodeData transmission phase on getting the schedule thesensor node sends its data as per the time slot and thechannel assigned -e time slot is divided into datatransmission andACK sections ACK supports link layererror control Using implicit selective repeat ARQ theCH sends an ACK message for all received packets Ifthere is no ACK the packet is assumed lost and hence aretransmission during the subsequent interval is done-is enhances reliability of transmissions though excessdelay is possible here which is also intolerable for real-time critical apps like video or audio -is gives rise to ahybrid MAC protocol to exploit the unused spectrumduring data transmission sessions [41]Advantages

(i) It achieves energy efficiency by scheduling mul-tichannel media access to reduce channel con-tention and remove collision idle listening andoverhearing

(ii) It maximizes network throughput by dynamicallyallocating time slots and channels for nodes byexecution of a traffic-adaptive QoS-aware sched-uling algorithm depending on needs for QoS andtraffic data information

(iii) It increases transmission reliability by in-corporating spectrum-aware ARQ to avoid spec-trum wastage

223 EQoSA -is protocol is developed for video andimage transfer across WMSNs in a hybrid mode withprovision of QoS Concerning the active sensor nodes andtheir traffic loads it employs dynamic session sizes andchanges bitmap-assisted (BMA) MACrsquos fixed session sizeNodes declare any data available for transmission duringcontention time -is forces the cluster head to make aschedule of how slots have been assigned It communicatesthis schedule to the nodes giving each node a given numberof slots per session and so enabling the busty traffic to beaccommodated [14] -e protocol does not perform wellwith DiffServe mechanisms but does well with schemessupporting busty traffic loads and strong cluster heads arenecessary in WMSNs that can assign and broadcast slots

224 QoS-Supported Energy-Efficient MAC (QE-MAC)Based on the IEEE 80211e standard this protocol ensuresfairness and has low latency and jitter and better efficiencyon energy usage than other QoS-aware MAC protocols forWMSNs It operates under a two-phased routine

-e first phase introduces the innovating prioritymechanism in IEEE 80211e QoS and several data types areassigned priorities Its coordination function is a combi-nation of distributed contention-based channel allocationand centralized polling-based channel access mechanisms

-e second phase preserves the nodesrsquo energy usingdynamic duty cycling -ere is an exchange of RTSCTSpackets to supplement CSMACA by muting nodes near thesenderreceiver during packet duration [42]

Advantages

(i) It ensures fairness(ii) It has low latency and jitter and better efficiency on

energy usage than other QoS-awareMAC protocolsfor WMSNs

225 Black-Burst (BB) Contention A contention-basedMAC scheme for enhancing instantaneous QoS access tocarrier-sense WSNs is proposed in [42] In this schemenodes contend for the channel until it is free Real-timepackets are given higher priority over non-real-time packetsand if a node is transmitting real-time packets it is prior-itized over other nodes [15] Voice and video are some of thereal-time apps in consideration here which periodicallyaccess the common radio channel when transmitting -eseneed constrained end-to-end delay of packets at the MAClayer -is protocol majorly does the following [42]

(i) Prioritization of real-time traffic(ii) Implementation of the round-robin mechanism in

real-time nodes(iii) Bounded access delays among real-time packets

Journal of Computer Networks and Communications 15

(iv) Implementation of real-time communication withvarying needs of bandwidth good for multimediatraffic

Amechanism for dynamically assigning packet prioritiesbased on their deadlines and the traversed hops is proposedin [43] -e ReAllocative Priority (ReAP) mechanism ismeant to provide QoS at the MAC layer to video traffic -isscheme introduces adaptive TXOP by modifying the TXOPdynamically based on the queue length [15] Other protocolsin this category include dynamic duty cycle and adaptivecontention window-based MAC protocol frame sharing(FRASH) MAC protocol and real-time independentchannels (RICH) MAC protocol

Advantages

(i) It can implement real-time communication withvarying needs of bandwidth

23 Challenges and Opportunities(a) One of the challenges in satellite transmission is

cross-talk leading to interference in different fre-quency bands It is necessary to increase the rise andfall times as well as protect sensitive nodes as wecircumvent floating ones if we are to regulate cross-talk as in satellite communication therefore Wefurther need to investigate the possibility of usingdifferential signaling -is constitutes an opportu-nity for further research

(b) Another challenge is that the hidden and exposedterminal problems are still noticeable despite the useof RTSCTS solutions in MAC since collisions arestill possible from transmissions of separate nodesAt the MAC layer ACKs might block efficientchannel usage by prohibiting exposed terminalsfrom reusing the channel and may be eliminatedleaving no proof of sender data packet receptionUsing a directional antenna is one of the proposedtechniques of high-speed WSNs like IEEE 80211-is constitutes an opportunity for further study intothese issues for better solutions though

(c) Another challenge is that since nodes in contention-based MAC protocols are mobile there are manycollisions by the packets and schedule in-consistencies in schedule-based protocols in the two-hop neighborhood information when nodes areentering or leaving [44] Deng et al [45] propose amobility-based clustering (MBC) protocol for mo-bile nodes to enhance the packet delivery rate -isprotocol in comparison with CBR has 25 chancesof decreasing collision-induced packet losses and50 chances in comparison with low-energyadaptive clustering hierarchy (LEACH)

(d) -e process of analyzing and evaluating routingperformance for WMSNs is quite complex becauseof the bursty multimedia traffic Because of the wayencoding scheme frames are structured as well as thechanges in between the scenes that occur naturally

the compressed video frames display some substantialburstiness on several time scales -ere will also befrequent route recalculation that is as well energy in-tensive for nodes frequently capturing and transmittingenergy -e solution here could as well possibly be theuse of the MBC protocol Route recalculation tooshould be dynamic to cater for the mobile nodes whichmight be in and out of the range at different instants oftime Recently the use of smart dumpsters for effectivewaste management in smart cities has also been pro-posed as a possible solution [46]

(e) For WSNs their routing protocols do not look atmultimedia applications which require enormousbandwidth and processing energy and should betransmitted in real time with utmost fidelity Extrachallenges in routing are imposed in WMSNs indeployment of heterogeneous sensors since therecould be audio video and still pictures in addition toscalar data during transmission Heterogeneous datahandled also have different QoS requirements andwith many different business needs different ser-vices with varying requirements give rise to a bigchallenge for the routing design -ere is a need todesign protocols that can support WMSNs such asEQoSA [14] for this purpose

Table 2 summarizes the reviewed MAC protocols forWSNs and the ones designed for suitability in WMSNs

3 QoS Strategies at the Application Layer

We discuss many protocols used to solve different needs ofcommunication between machines

31 ConstrainedApplication Protocol (CoAP) It is a requestresponse protocol designed by the Internet Engineering TaskForce (IETF) for constrained devices such as those havinginsufficient RAM or CPU and WPANs with low-powerconstraints resulting in poor transfer of packets andenormous overhead -e protocol was designed for ma-chine-to-machine (M2M) apps and system automation toreduce bandwidth requirements increase packet transferdecrease the overhead and make work simpler leading tolightweight implementation HTTP commands like GETPOST PUT and DELETE are used in a client-server ar-chitecture Using the publishsubscribe architecture CoAPcan support multiple users and hence yield better perfor-mance via multicasting enabling asynchronous messageexchange -is protocol can support both multicasting andunicasting via UDP and hence is a reliable and simpleprotocol -is is done via the messaging sublayer (for re-liability based on UDP) and the requestresponse sublayer(for communication and interaction) [16]

It uses the following messages

(i) Confirmable message this guarantees reliabilityusing ACK messages sent by the receiver node-ese could be synchronous or asynchronous ifthere is a need for more computational time

16 Journal of Computer Networks and Communications

(ii) Nonconfirmable message here no ACK messagesare sent back

(iii) Acknowledgment (ACK) message this confirmsreceipt of a confirmable message

(iv) Reset message this confirms receipt of an un-processed message in case any critical parts nec-essary for interpreting the message are missing

(v) Piggybacked response the receiver directly re-sponds while getting the ACK message

(vi) Separate response the reply by the receiver is dis-crete from the ACK message

CoAP has no in-built security features It is secured bythe datagram transport layer security (DTLS) that runs ontop of UDP though it was not designed for the IoT -ismakes its suitability debatable since DTLS offers no supportfor multicasting It also needs more packets for handshakingthereby increasing the traffic volume and computationalresources and so shortening the lifespan of mobile devicessince they use batteries [17]

Advantages

(i) It can support multiple users and hence yieldbetter performance via multicasting

Table 2 Comparison of MAC layer protocols used in QoS provisioning

Protocol Features QoS parameter Comments Access Priorityassignment Energy eff Designed for

S-MAC[22] Collision avoidance Energy latency Requires periodic

synchronization CSMA Hybrid Medium WSNs

B-MAC[23] Collision avoidance Energy delay

throughput Flexible interface CSMA Dynamic High WSNs

T-MAC[25] Collision avoidance Energy Complex nonscalable CSMA Dynamic High WSNs

C-MAC[13]

Dynamic channelallocation Energy Multichannel MAC CSMA Dynamic Medium WSNs

CU-MAC[24]

Multichannelmechanism

-roughputlatency High packet delivery ratio TDMA Dynamic mdash WSNs

A-MAC[39]

Per flow serviceguarantees Delay m-ary tree model

dynamic prioritiesTDMACSMA Dynamic Medium WSNs

ER-MAC[30]

Emergencyresponse

Energy delaydelivery ratio

Adapts to traffic andtopology changes

TDMACSMA Hybrid Higher than Z-

MAC WSNs

RI-MAC[31] Collision detection Delay packet

delivery ratioHigh packet delivery ratio

and throughput CSMA Dynamic Medium WSNs

SRI-MAC[33] Collision avoidance Energy delay Good data transfer and

energy consumption TDMA Dynamic High WSNs

H-MAC[34] Slotted sleep time -roughput

latencyData transmission only in

sleep timeCSMAAloha Dynamic Medium WSNs

Z-MAC[36] Collision avoidance Energy delay

throughputVery good channel usage

and error syncCSMATDMA Dynamic Medium WSNs

Q-MAC[37]

Intrainternodescheduling Energy High QoS improved

energy efficiency CSMA Hybrid High WSNs

X-MAC[32]

Overhearingavoidance

Energy delaythroughput

Reduces excessivepreamble by strobing CSMA Adaptive

static Medium WSNs

L-MAC[40] Collision avoidance Network

lifetime energyReduces the number oftransceiver switches TDMA Static Medium WSNs

WiseMAC[38]

No downlinkchannel collision Power delay Significantly low-power

consumptionAlohaCSMA Static Higher than

ZigbeeInfrastructure

WSNsAR-MAC[27] Collision avoidance Energy delay Uses data packets and

control packets TDMA Static Higher thanIEEE 802154 WBANs

BB-MAC[42]

Real-time trafficprioritization Delay Employs round robin in

real-time nodesCSMATDMA Dynamic High WMSNs

COM-MAC [41] Collision avoidance Packet delay

throughputBalances reliability and

retransmissionTDMAFDMA Dynamic High WMSNs

EQoSA[14]

Busty traffic loadaccommodation Energy Provides QoS support for

video and imagesTDMACDMA mdash High WMSNs

QE-MAC[42] Collision avoidance Energy jitter

latencyRuns central polling-like

channel access CSMA Dynamic High WMSNs

R-MAC[26] Collision avoidance Energy delay Reservation based mdash Static High Underwater

SNs

Journal of Computer Networks and Communications 17

(ii) It enables asynchronous message exchange(iii) It is a reliable and simple protocol

32 Message Queue Telemetry Transport (MQTT)Developed by IBM and later improved by OASIS with anintention of decreasing bandwidth MQTT is a simple open-source publishsubscribe protocol analogous to the client-server version It is appropriate for lightweight IoT appli-cations and M2M transmissions in constrained situationslike limited power computation capability and memory orbandwidth -e network bandwidth and the computationalresource usage are decreasing since clients are not updatedmaking publishsubscribe protocols outperform requestresponse protocols via IoT requirements [16] -is protocolcan run over TCPIP and so has the following advantages

(i) Delivering of packets with reliable guarantee(ii) Enabling multicasting(iii) Remote device session establishment(iv) Reduction of transport overhead and protocol ex-

change to minimize network traffic(v) Provision of the notification mechanism when

anomalies occur [47]

MQTTrsquos design targets economical use of bandwidthand battery so it has a major application in FacebookMessenger MQTT ensures QoS and reliability of messagedelivery by providing the following three QoS levels

(1) Fire and forget a message is sent once depending onthe networkrsquos best effort and there is no need forACK (lowest QoS level)

(2) Delivered at least once a message is sent at least onceand an ACK message is needed to avoid duplicates

(3) Delivered exactly once by a four-way handshake itensures a message is sent exactly one time (highestQoS level)-e architecture it has three key components (i)publishersmdashlightweight nodes sending data to thebroker before returning to sleep (ii) broker whichorganizes sensory data in topics and sends them tointerested subscribers and (iii) subscribersmdashIoTapps interested in data sent by sensors through thebroker as shown in Figure 3

MQTT is a simple lightweight protocol that is betterthan CoAP via low and high packet loss probability interms of throughput and exhibits lower latency in lowsampling rates [48] It supports a number of imple-mentations for embedded devices and mobile applicationsand offers web-socket support too Nevertheless it hassome shortcomings

(i) It is only employed in simple data types(ii) It exhibits high latency in high sampling rates(iii) It is unsuitable for real-time apps in such cases(iv) It does not provide automatic discovery andor

sufficient security at the protocol level [49]

33 Secure Message Queue Telemetry Transport (SMQTT)-is is an extension of MQTT that is based on lightweightattribute-based encryption using elliptic curve cryptogra-phy -e protocol has three entities as shown in Figure 4(i) publisher that publishes information under a certaintopic (ii) subscriber that obtains the same topic in-formation via a broker and (iii) public key generator(PKG) or broker that acts as the trusted third party -eprotocol has four phases namely (a) set-up phase (b)encrypt phase (c) publish phase and (d) decrypt phase

(a) Set-up phase publisher and subscriber provide aunique ID with attributes for registration with thePKG -is generates a secret master key set andpublic parameters that it publishes with set U A1A2 An For ciphertext policy-attribute-basedencryption (CP-ABE) the PKG gives a key set todevices with attributes [50]

(b) Encrypt phase based on the access tree the publishercreates an access policy An access tree is sent by thepublisher to the PKG for generating the key-gen-eration policy in the key policy- (KP-) based ABEalgorithm -e publishers know beforehand thetopics and subscribers to access as well as the rulesfor access Subscribers are issued with private keysfor all communication sessions Publishers in theCP-ABE algorithm come up with access trees andrules where the payload is encrypted by the pub-lisher who also issues more info needed for de-cryption and the rules With the help of public keycryptography data are encrypted giving rise to ci-phertext on the basis of KPCP-ABE and the receivergets SUBACK from the broker [51]

(c) Publish phase ciphertext is embedded in theSPublish command as payload and topic name isdefined in the header by the publisher -e brokerreceives the SPublish packet and replies withPUBACK prompting the publisher to respond withthe PUBREL packet -e broker then broadcasts totopic subscribers and deletes the data forwarding thePUBCOMP packet to the sender node

(d) Decrypt phase the encrypted information isdecrypted by the subscriber with the help of a secretkey Considering KP-ABE on verification that itsatisfies the access policy the subscriber invokes thePKG for a corresponding key which it sends afterverifying the request With CP-ABE subscribers usethe private key for decryption after authenticationthereby enabling offline interaction [51]Advantages

(i) -e information being encrypted makes it moresecure

34 MQTT for Sensor Networks (MQTT-SN) -is is a ver-sion of MQTT designed to operate under environments oflow bandwidth high link failures and short message lengthIt is a publishsubscribe (pubsub) protocol designed for

18 Journal of Computer Networks and Communications

UDP-based WSNs to visualize communication for low-power devices [5152] It comprises a publisher device actingas the MQTT-SN client to forward communications to agateway that also changes the MQTT-SN message to theMQTTmessage before delivering it to the broker that sendsit to MQTT-SN subscribers [51]

It has a gateway for changing protocols from MQTT toMQTT-SN (Figure 5) [52] -is protocol supports nodes in

the sleeping mode using an offline keep-alive procedure thatcan be used by battery-operated devices to get into thesleeping mode To support sleeping clients MQTT-SN has anew offline keep-alive procedure with which battery-oper-ated devices can go to the sleeping state in which theirmessages are buffered at the server until they are awake InMQTT-SN clients with no preconfigured gateway addresscan figure out the real address of their operational gateway

Publisher

Subscriber

Subscriber

Broker

Publisher

4G

Figure 3 Architecture of the MQTT protocol

Register by mentioning URI Register by mentioning URI

Access tree

Set-up

Publisher ReceiverPKG(broker)

Key management Key management

Encrypt data using public parametersand random key This is done using

CP-ABE Publish it as

Subscribe (topics)SUBACK

S-Publish (TopicEncrypt (data))S-Publish (TopicEncrypt (data))

Encryptand publish

PUBACK

PUBACK

PUBRELPUBRELPUBCOMP

PUBCOMP

Figure 4 SMQTT protocol

Journal of Computer Networks and Communications 19

within the network through a discovery procedure Simul-taneously we may have several gateways in a single wirelessnetwork that can be cooperative via load sharing

MQTT-SN architecture it is composed of MQTT-SN clients gateways and forwarders (Figure 5)rough the MQTT-SN gateway (GW) the MQTT-SN protocol is used by the MQTT-SN clients toconnect to the MQTT server Integration betweenthe MQTT-SN GW and the MQTT-SN server mightbe possible or not Translating between the MQTT-SN and the MQTT for a stand-alone GW is doneusing the MQTT-SN protocol If a GW is detachedfrom the network it can be accessed by the MQTT-SN clients through a forwarder is just encapsu-lates the received MQTT-SN frames that are on thewireless side before forwarding them to the GW un-altered It then decapsulates the received frames tooresending them to the client as sent by the GW [52]ere are 2 types of gateways (a) transparent gatewayand (b) aggregating gateway as shown in Figure 6

(i) Transparent gateway this launches and sustains adirect MQTT connection between the MQTT-SNclient and the MQTT server which is entirely re-served for end-to-end client-server messaging enumber of servers and that of MQTT-SN clientsconnected to the GW are the same is transparent

GW translates syntax between these protocols and iseasier to implement compared to the aggregatingGW But the MQTT server must support dierentsessions for all connected clients Some MQTTservers are congured to enable few parallel sessions

(ii) Aggregating gateway this maintains a single MQTTconnection to the server on behalf of all clients Allexchanged messages from the MQTT-SN client to theaggregating GW are directed to the GW which sortsthe information to forward to the server e aggre-gating GW decreases the number of MQTT sessionssimultaneously supported by the server in WSNs withmany battery-operated sensors and actuators (SAs)having limited storage and processing capabilities It ishowever more complex to implement compared to atransparent GWe general format of the MQTT-SNprotocol message has two or four message headeroctets and n octets of variable header that is optionaldepending on the message in question MQTT-SN has256 message types (MsgType) with those from0X1Eminus 0XFD 0X19 and 0XFF reserved for futureusage [51]

Advantages

(i) e protocol can operate under environmentsof low bandwidth high link failures and shortmessage length

MQTT-SN gateway

MQTT-SN

MQTT

MQTT-SN

MQTT-SN

MQTT-SN

MQTT-SN client

MQTT-SN clients

MQTT broker

MQTT-SN client

MQTT-SN client

MQTT-SN gateway

MQTT-SN forwarder

Figure 5 MQTT-SN architecture

20 Journal of Computer Networks and Communications

(ii) It is a pubsub protocol designed for UDP-basedWSNs to visualize communication for low-power devices

(iii) It can be optimal for low-cost battery-operatednodes having low processing power and lessstorage

(iv) It supports nodes in the sleeping mode using theobrvbarine keep-alive procedure used by battery-operated devices

35e Extensible Messaging and Presence Protocol (XMPP)e standardization of this protocol by the IETF made itpopular for usage over the Internet especially in instantmessaging and chatting platforms It is however susceptibleto spamming attacks and its failure to have worldwidesupport has made Google too withdraw its support of thestandard [17] e protocol is mainly useable in the client-server architecture but can also be applicable in client-clientXMPP communication [53] Among the application layerprotocols XMPP is the only one that supports both asyn-chronous (publishsubscribe) and synchronous (requestresponse) services while running over TCP but can still runover other protocols

Given the simplicity of its implementation and ease ofusability in XMPP instant messaging (IM) is suitable forXMPP and is implemented using the message stanzaltmessagegt ese stanzas are usable in one-to-one ormultiuser sessions or chatting or in error reporting andsending headlines Attribute type of the stanza denes themessage type e attribute type could be a chat group chatheadline normal or error An instant messaging network inXMPP can support many XMPP clients and servers with agateway congured to enable other IM network users toconnect to other XMPP users e XMPP server mainlycontrols sessions and XML streams to clients and servers inaddition to routing XML stanzas in the streams [53] eTCP ports 5269 and 5222 have been respectively registeredfor the XMPP server-to-server connections and XMPP clientconnections by the Internet Assigned Numbers Authority(IANA) XMPP enables specication of XMPP ExtensionProtocols (XEP) to enhance its usefulness [17]

XMPP security clients connect to the server usingtransport layer security (TLS) and simple authentication andsecurity layer (SASL) protocols which are at the same timeused by the servers for interdomain communications Fordata condentiality and integrity TLS will encrypt thestreams of XML for all sessions To conrm the identity of anyclient that needs to access the server the SASL protocol doesthe authentication of the XML stream For any communi-cating session with a server to start the client needs to resolvethe serverrsquos DNS hostname For protection of the server of theclient from third-party attacks the server maintains the au-thentication details secret ie IP address and access methodand only original connections of the server are required

In server-to-server communications the SASL protocolis employed for authentication and condentiality Server-to-server communications may as well be deactivated from aserver by an administrator depending on the organizationalpolicy To guard against domain-spoong attacks serverdialback can be a good solution since servers support itsecuring more the XML stanzas Nevertheless it cannot beused to authenticate secure or encrypt in-server streamsand the server identication resulting therein is not strongenough to be relied on It further cannot secure against DNSpoisoning attacks and IP session hijacking for remote do-mains [53] High-security domains can use TLS and SASL

Advantages

(i) It is popular for usage over the Internet especially ininstant messaging and chatting platforms

Disadvantages

(i) It is susceptible to spamming attacks(ii) It has failed to have worldwide support making

Google withdraw support of the standard(iii) It is not secure against DNS poisoning attacks and

IP session hijacking for remote domains

36 Advanced Message Queuing Protocol (AMQP) isprotocol can be utilized by a number of platforms formessage exchange since it is open source and written inmany languages It is a publishsubscribe model depending

MQTT-SNclients Broker

MQTT

(a)

MQTT-SNclients Broker

(b)

Figure 6 (a) Transparent gateway and (b) aggregating gateway

Journal of Computer Networks and Communications 21

on a reliable and efficient messaging queue with capacity forinstant messaging or e-mail for different applications es-pecially in IoT environments It differs from others in thesense that message types can be specified and have theirsource traceable and how to do trade-off between perfor-mance security and reliability For reliability of messagetransmission AMQPmakes use of some delivery guaranteeslike at most once at least once and exactly once It can alsohowever utilize the TCP transport layer for guaranteeingreliability -e AMQPrsquos publishsubscribe method has twocomponents namely (i) exchange queue meant for routingof messages to an appropriate order in the queue and (ii)message queue that acts as storage pending their delivery toreceivers [16] In this protocol model the broker can makerouting decisions contrary to othermessaging systems whereapps using the queue have in them embedded the decision-making logic for recipients or senders -e developer mayneed to check each of the affected apps so as to change thelogic for routing and delivery of the messages [54]

Advantages

(i) It is convenient(ii) It has capacity for interoperability between dif-

ferent vendors(iii) It can use open standards to enable connection

between business partners(iv) It can enable innovations based on AMQP foun-

dations [55](v) It can enable supporting of mission-critical ap-

plications in e-commerce(vi) It can guarantee reliable delivery of messages

Disadvantages

(i) It is not very appropriate for real-timeapplications

(ii) It may not offer automatic discovery(iii) Interoperability is not guaranteed(iv) It lacks open-source libraries for constrained

gadgets [49]

37 Data Distribution Service (DDS) A product of theObject Management Group (OMG) DDS is a pubsubprotocol meant for M2M communications with two sub-layers -ese are the data-centric pubsub sublayer re-sponsible for connecting anonymous data publishers tosubscribers and the data-local reconstruction sublayer that isoptional and integrates DDS into the application layer [56]-e DDS architecture connects system components asshown in Figure 7 Publishers and subscribers are discon-nected with respect to (i) time (ii) space (iii) flow andbehavior (iv) platforms and (v) programming languages[57] A DDS data publisher generates topics and publisher-subscriber coupling is represented as topic name data typeschema and publishers and subscribersrsquo QoS attributesDelivering of data is done by the publisher layer -e datawriter and publishers together make a decision about anynecessary alterations for subsequent forwarding to thesubscribers who need to send data to IoT applications Data

readers are responsible for delivering topics to the sub-scribers after reading through them [56]

Some of the major components of the DDS architectureinclude the following [57]

(i) Topic this is a logical path specifying publicationand subscriptionrsquos data type between DataWritersand DataReaders For a successful session to begintopic names types and DataWriters and Data-Readersrsquo QoS should correspond

(ii) Domain this is a logical setting for transmissionthat can be utilized for separating and optimizingnetwork transmission in a setup of similar appli-cations If DDS apps share a domain ID then theycan exchange data among themselves

(iii) DataWriter and DataReader these are terminalobjects for writing and reading messages re-spectively to and from a universal data space

(iv) Participant this is an object that stands for apublisher or subscriber of the DDS app in a do-main acting like a container for other objects

Advantages

(i) It is highly reliable and offers QoS support(ii) It is highly scalable(iii) It is appropriate for real-time apps(iv) It is fault tolerant(v) It has high level of interoperability(vi) It does not need brokers that may act as

bottlenecks(vii) It offers automatic discovery(viii) It has a user-defined data structure for topic(ix) It supports publisher and subscriber decoupling

Disadvantages

(i) It is originally developed for only stand-aloneLANs

(ii) A lot of memory is needed(iii) It has no open-source libraries to support con-

strained devices(iv) It is hard to design and configure compared to

other protocols [49]

38 RESTful Services REpresentational State Transfer (REST)-is is an architectural style based on a set of principles thatdefine and address networked resources It offers web servicesthat enable transmission from the HTTP device to the devicewithin the IoT architecture and enables clients to reach theserver [16]-is protocol utilizes HTTP commands to backupthe requestresponse messaging model HTTP is a popularand secure protocol in the WWW using TLSSSL [16] ARESTful application has the following characteristics

(i) State and functionality are separated into distrib-uted resources

(ii) All resources are distinctively named by the use ofunique HTTP commands (GET POST PUT orDELETE over the Internet)

22 Journal of Computer Networks and Communications

(iii) It is a clientserver stateless layered protocol thatsupports caching REST uses the same methods asHTTP for all requests or responses with POST andGET for creating and retrieving resources PUT forupdating and changing the state of resources andDELETE for removing resourcesAdvantages of REST over arbitrary web services(SOAP) in the IoT context

(i) It reduces overhead(ii) It reduces parsing complexity(iii) It is stateless(iii) It integrates more tightly with HTTP [58](iv) It is easy to implement and learnis together

with its natural t over HTTP makes it apreferred method for Web 20 applications toshowcase their data and secure M2M sessionsin the IoT

(v) Applications that support RESTful web ser-vices exhibit better performance than the re-source-constrained WSN nodes [59]

(vi) REST is important for the IoT since it supportscommercial M2M cloud platforms

(vii) Its implementation in smart phones and tabletapps is also easy since it needs only an HTTPlibrary that is accessible to all operating systemplatforms [56]

Disadvantages

(i) Although it enjoys wide usage in commercialM2M platforms it is not likely to dominatebecause it is hard to implement

(ii) Its HTTP usage makes it incompatible withconstrained M2M devices leaving its usage toend applications

39 WebSockets It is a web-based protocol operating on asingle TCP channel and providing full-duplex transmissions

It permits secure duplex communication between a clientexecuting unreliable code in a restricted environment and adistant node communicating from that code while using theorigin-based security model employed by web browsers overTCP [60] e protocol is not a requestresponse model andis not a publishsubscribe model So unlike other protocolsa web-socket session is established by the client initiating ahandshake to the server to launch the session is onceestablished a full-duplex client-to-server connection beginsasynchronously and runs until terminated by either party[56] It gives a model for browser-based apps utilizing bi-directional communication with no need of many HTTPconnections to servers

WebSocket comprises two parts ie (i) the handshakehaving a clientrsquos message and the serverrsquos handshake responseand (ii) the data transfer HTTP headers should be repeated byapplications in all client requests and server responses arisingfrom the endless polling which might increase the commu-nication overhead subject to the application [61] WebSocketcan be used to come up with scalable real-time web appli-cations since it can enable full-duplex transmission via onesocket across the Web Its security can be ensured with thehelp of TLSSSL is protocol is suitable for resource-con-strained devices and IoTapplications will not have a good userexperience with it since it operates a clientserver architectureNevertheless it can outcompete all TCP-based protocolsbecause of the following advantages

(i) It is t for real-time transmission(ii) It has improved security(iii) It minimizes transmission overhead(iv) It can provide ecient messaging systems using

WebSocket ApplicationMessaging Protocol (WAMP)[17]

310 Streaming (Simple) Text-Oriented Messaging Protocol(STOMP) is is a TCP-based protocol designed for textmessage-oriented middleware (MOM) with the use of

Topic

Datareader

Datareader

Datareader

Datawriter

Datawriter

Datawriter

Subscriber PublisherPublisher

Data bus (DDS domain)

Subscriber

Topic Topic

Figure 7 DDS architecture

Journal of Computer Networks and Communications 23

HTTP-like commands for interoperability among platformslanguages and brokers Data communication takes placebetween the client and the broker in line with commands(like CONNECT DISCONNECT ACK NACK SUB-SCRIBE UNSUBSCRIBE SEND BEGIN COMMIT orABORT) then headers in the form ltkeygt ltvaluegt (one perline) then a blank line and then body content terminatingin a null character Server-client communication takes placeover a MESSAGE RECEIPT or ERROR frame with headerand body content format being analogous MOM productssupporting STOMP include Apache ActiveMQ or FuseMessage Broker HornetQ Open Message Queue(OpenMQ) RabbitMQ and syslog-ng through its STOMPdestination plugin [62] In addition to these other appli-cation layer protocols include simple media control protocol(SMCP) lightweight local automation protocol (LLAP)simple sensor interface (SSI) lightweight M2M (LWM2M)XMPP-IOT and simple object access protocol (SOAP)

Advantages

(i) It is lightweight for constrained networks(ii) It has flexibility to choose quality of services with

the given functionality(iii) It is standardized by OASIS Technical Committee(iv) It is easy and quick to implement

Challenges and opportunities

Here security threats are imminent because of the open-endedness of the application layer [63] -ese could includethe following

(i) -ere is poor or lack of a security design for anapplicationrsquos function Coming up with a gooddesign to address this could be a good ideaopportunity

(ii) Some of the programs might have backdoorsevading otherwise secure controls to allow un-authorized access to network resources We requiremechanisms to patch up such loopholes

(iii) Applications that lack or have weak authenticationmechanisms can be easily targeted by unlicensedusers to abuse the network Such mechanisms needto be strengthened further

(iv) Excessively complicated access control rules maybecause of their misunderstanding or poor writinggive unauthorized access allowing dubious activityrather than protecting the network they purport toprotect -ese should be simple and clear to un-derstand by all users

(v) In booting workstations without disks and managingnetwork devices we usually use the TFTP protocolalthough access does not require authenticating theuser by usernamepassword -is makes an intruderable to gain easy access to configaccess info as longas he can guess the filenames -is needs to be fixed

A few steps can be used to secure the application layerTo protect network data transmitted across the network

applications need to strengthen their encryption and au-thentication mechanisms Apps too need to enable strictcontrols of privileges for data access using good mechanismsto balance between usability and effectiveness [63] Appli-cations handling sensitive data need detailed logging andaudit capability as well as testing and reviewing

Table 3 summarizes the application layer protocols dis-cussed with their weaknesses and differences among them

4 Cross-Layer QoS Strategies

41 Service-Differentiated Real-Time Communication Scheme(SDRCS) -is is an event-driven routing protocol that routesreal-time traffic by featuring a cross-layer packet-forwardingdesign in which it integrates the real-time routing functionalitywith a new priority-based MAC scheme -e protocol ap-proximates distributed packet traversal speed for traffic clas-sification and admission control on the basis of this design Itfurther localizes decision-making by prioritizing packet for-warding to maximize the speed of packets [64]

Advantages

(i) It improves bandwidth utilization bandwidthdegradation due to unschedulable packets can beavoided

(ii) It offers a lightweight packet schedulability estima-tion mechanism Here it uses received signalstrength and admission control and then earlymissed-deadline packet-dropping policies are made

(iii) It delivers high end-to-end throughput via highersource data rates alongside strict end-to-end latencyneeds

(iv) Nodes which meet real-time requirements of lesstraffic load and better channel quality are givenhighest priority when it comes to packet forwarding

-e SDRCS operates via 5 mechanisms namely (1) RSS-(Rich Site SummaryReally Simple Syndication-) basedgrouping (2) admission control (3) prioritized queuing(4) real-time MAC and (5) dynamic forwarding -isprotocol has a queuing policy that employs per-hopdeadline-based queues and the nodes schedule theirpackets using FIFO priority-based queues [1] -e prin-cipal features of this protocol are mentioned as follows

(i) It is event driven and is easily adaptable to anynetwork changes

(ii) It does not need additional hardware for locali-zation or multichannel transmission

(iii) -e cross-layer packet-forwarding strategy can beused when transmitting multimedia traffic

(iv) It is able to efficiently circumvent any voids therein

Early deadline miss (EDM) policy drops all unscheduledpackets -e SDRCS circumvents wrong packet dropsdue to EDM since it adjusts to network changes easily

42 Sensor Fuzzy-Based Image Transport (SUIT) -is pro-tocol regulates congestion using a scheme based on fuzzy

24 Journal of Computer Networks and Communications

logic It transmits packets with reduced quality to allowablelevels if there is congestion enabling it to transmit lower-quality packetsframes without being dropped-e numberof packets delivered per second thus rises and while thisenhances video streaming apps it is more appropriatelymade for those that quickly transmit non-real-time videosuch as in video surveillance SUIT can stream JPEG pic-tures other than the traditional predictive video codingmethods that are not suitable for sensor devices because oftheir constraints of energy memory and computationalspeed Progressive JPEG-PJPEG performs better via packettransmission rate and delay though it degrades the imagequality because of congestion [65] -is protocol employscross-layer methods for interlayer data exchange though itstill employs some layering making it still possible topreserve all advantages of the layered design since it hasapplication and transport layers but has noMAC or routinglayer

For congestion detection and control SUIT bases on thepercentage number of in-bound and out-bound frames perwindow number of contenders and buffer occupancy of thenext-hop node Since buffer management has an effect oneffectiveness and QoS inWMSNs it is useful in the transportlayer so the protocol prioritizes packets for better QoS SUITreorders packets by tracking the sequence numbers of thedelivered packets since they all have a source ID framenumber and sequence number that it uses to detect anymissing packets -e protocol however does not providepacket reliability [65]

Advantages

(i) -e protocol prioritizes packets for better QoS(ii) It can transmit lower-quality packetsframes

without being dropped

Disadvantages

(i) It does not provide packet reliability

43 Network Layer QoS Support Enforced by a Cross-LayerController (NLQS) -is scheme permits packet-level servicedifferentiation as a function of throughput end-to-endpacket error rate and delay [66] (see Figure 8) -is en-hances network layer QoS It comprises a cross-layer controlunit (XLCU) for configuring and controlling networkingfunctionalities at physical MAC and network layers [18]-is depends upon the unified logic which affects choices forapplication layer needs and status of functional blocks whichdo implementation of networking functions [66]

Advantages

(i) Cross-layer interactions can be controlled withoutweakening the upgradability modularity andsimplicity of designing the system

44 Cross-Layer Signaling Shortcuts (CLASS) CLASS has ahigh level of effectiveness flexibility and comprehensive-ness According to [67] it has the following features

(i) Direct signaling among remote layers leading tofaster transmission

(ii) Lightweighted internal message format standard-ized protocols are not suitable for use in internalsignaling since they are not usually lightweight forinstance transmitting against faults in the networkCLASS requires only three fields namely destina-tion address event type and event contents

(iii) -e setup of its exterior messages is standardconsidering external signaling ICMP and TCPIPare suitable for general messages and short notifi-cations respectively

(iv) Message control protocol should ensure optimizedand organized exchange of dense simultaneousmessages across layers for high efficiency andavoidance of possible conflicts -e protocol appliesto many cross-layer signaling situations

Table 3 Differences and problems with application layer protocols

Protocol Architecture Transport QoS options Security Weaknesses

CoAP Requestresponse UDP DTLS No in-built security features DTLS does not supportmulticast

MQTT Publishsubscribe TCP TLSSSL Insufficient security at the protocol level sincepayload values are not encrypted

SMQTT Publishsubscribe TCP CPKP-ABE Key revocation and group pubsub for distributedSMQTT are still a challenge

MQTT-SN Publishsubscribe UDP TLSSSL SSLTLS suffers from attacks like BEAST CRIMERC4 and Heartbleed

XMPP RequestresponsePublishsubscribe TCP X TLSSSL Susceptible to spamming attacks and lacks worldwide

supportAMQP Publishsubscribe TCP TLSSSL Inappropriate for real-time applications

DDS Publishsubscribe TCPUDP TLSSSL Memory intensive and no open-source libraries forconstrained devices

REST Requestresponse HTTP X HTTPS Hard to implement Uses HTTP so incompatiblewith constrained apps

WebSockets PublishsubscribeClientserver TCP X TLSSSL Bad user experience for IoT apps since it runs the

clientserver architectureSTOMP Clientserver TCP HTTP -e broker can act as a bottleneck

Journal of Computer Networks and Communications 25

Advantages

(i) e protocol is scalable and has a very lowpropagation delay

(ii) It is very ecient and exible since it uses directsignaling between any two layers

(iii) It is very comprehensive and eective

45 Cross-Layer and Multipath-Based Video Transmission(CMVT) CMVT is a hybrid of application and networklayers At the application layer it carries out the encoding ofvideo streams to video data frames (I- P- and B-frames) bymeans of the MPEG-4 encoding format At the networklayer route discovery and data transfer take place Twoalgorithms greedy forwarding and rollback are used todiscover a number of paths from the source to the sink nodevia the route discovery method [20] A specied node icalculates the evaluation of its adjacent node j using thefollowing equation

fij (1 minus α)d2(j D) minus d2min(i)d2max(i) minus d

2min(i)

+ αeinit(j) minus eres(j)

einit(j)

(19)

where fij evaluation value for the node i to j d2(j D) isthe distance from the node j to the destination node D

d2min(i) and d2max(i) are the minimum and maximum dis-

tances of neighbors of the node i to D respectively einit(j) isthe initial energy of the node j eres(j) is the current residualenergy of the node j and α is the energy coecient given asfollows

α emax(i) minus emin(i)

emax(i) (20)

where emax(i) and emin(i) are respectively the maximumand minimum energy left of all node irsquos neighbors

e network layer further transmits video streams withCMVT doing status evaluation choosing an appropriatecommunication path for any packets and the QoS guaranteelevel for the path i is calculated as

fi (1 minus ω)hisum hi

+ ωnisum ni

(21)

where fi is the evaluation value for the path i hi are the hopsfor the path i ni is the summation of packets sent via isum ni isthe summation of packets sent by sources and ω is theenergy consumption factor [18]

Advantages

(i) e CMVT protocol is very superior in mediatransmission particularly large-scale WMSNs

To network peers

Flow requirements

-

Atimes

Admissioncontrol

Capacityassignment

Next-hopselection

Coding rate

Data rate

QoScontracts

Cross-layer controlunit (XLCU)

Localization subsystem

Synchronizationsubsystem

Networkcoordinationunit (NCU)

UWB modulator

Channel coder

QoS scheduler

Georouting

Source coder

Rate control

Figure 8 Cross-layer controller architecture

26 Journal of Computer Networks and Communications

46 Cross-Layer Cooperative MAC (CoopMAC)CoopMAC [68] is based on the IEEE 80211 DCF approach-ere is a reduction in interference of neighboring cellsleading to uniform coverage in densely deployed networks[68] A connection is established by two-way authenticationbetween the sender and the receiver When a Clear-to-Send(CTS) packet and short interframe space (SIFS) are obtainedby the sender packets are transmitted right to the receiverwith no valuable cooperation If available the two nodes findout whether they can exchange data -is is done by the useof a helper identification signal to establish how feasiblecooperative communication is It is established in thepresence of a signal and if it is not available direct com-munication is activated For distributed systems helper-initiated cooperation is preferable because of the presence ofthe RTSCTS packets [69]

Advantages

(i) CoopMAC has an advantage over others due to thespatial diversity amongst the three nodes [70]

(ii) It has the potential to achieve significantthroughput and delay with no complexity in thedesign of the system

47 Cooperative MAC Protocol for Multihop Networks (M-CMAC) Just as CoopMAC M-CMAC is made in such away that high data rate stations help low data rate stations toforward traffic for broadcasting [71] -ere are also helperschosen in such a way that 2 fast-hop transmissions replace aslow hop transmission -e helper via whom there existsleast delay from the sender to the receiver is considered bestbecause of theminimum two-hop transmission rate and so isthe chosen neighbor to the duo Here the working as-sumption is that all nodes have their position coordinatesknown and thus the Euclidean distances between all pairscan be calculated and converted to the data rate for that linkAll nodes have cooperative tables (CT) of potential helperswith destination and helper MAC address Euclidean dis-tance and total distance through the helper Any sender withdata to transmit checks out for the existence of any helper inthe CT for the destination and if available forwards RTS forchannel reservation for single-hop duration M-CMAC hasgot an RTS format having a shape with five fields similar tothe one given as follows

Framecontrol Duration Source

addressDestination

addressHelperaddress

-e destination nodersquos address is kept by the source in thepart for the helper address that is in turn saved as the desti-nation address Nodes inspect helper and destination addressfields on receipt of RTS whereby the node acts as the othernodersquos helper in case the helper address field is not the sameand the helper transmits CTS back to the source if at all it wantsto forward data On delivery of the CTS packet the sourcetransmits to the helper which in turn sends the packets to thedestination that also sends ACK to the helper on receipt of thepacket In M-CMAC there exists a higher level of channelreuse (parallel transmissions) due to the increased number of

nodes therein that subsequently raises availability of helpers fordata forwarding -is results in increased throughput incomparison with CoopMAC and IEEE 80211 DCF [71]

Advantages

(i) It promises a higher throughput compared toCoopMAC and IEEE 80211 DCF

48 Cluster-Based Cooperative Routing (CBCR) Protocol-e CBCR protocol has a multihop data-forwarding func-tion realized at the link layer with cooperative links that useM-CMAC -is protocol encompasses two stages namelythe routing relay selection phase and data forwarding phase

(i) Routing relay selection phase all nodes announcetheir presence to their neighbors by broadcastingperiodical beacon messages that carry the MAC ad-dress of the node Each of these constructs a relaytable that contains all neighboring nodes with whichit is able to communicate -e node further broad-casts its neighbor list in case of any change to itsentries since the previous broadcast -e MAC ad-dresses of nodes next to the node X are contained inits relay tablersquos column one and the neighbor nodersquosrow has MAC addresses for neighbors of the adjacentnode Based on its relay table each node in-dependently chooses routing relays Choosing a nodeas a relay node depends on the number of nodes itconnectsmdashit should be the highest number [71]

(ii) Data forwarding phase a node with packets totransmit needs to first verify if the receiver is in thesame cluster and if that receiver has a helper to whichthe packet is thence forwarded but in whose absencethe packet is delivered right to the receiver In casethe intended receiver is in a different cluster therelay table will be checked to see whether that re-ceiver is reachable via other routing relays If it isreachable packets will be sent to the routing relaydirectly or via the helper if at all the relay has one Incase of destination unreachable via relays packetsare multicast to all relays by the node [71]Advantages

(i) It enables multicasting

49 MAC-PHY Cross-Layer Protocol A cooperative cross-layer standard for cooperation at the physical layer in next-generation WMSNs is developed in [70] It provides acomplete MAC layer algorithm that gives a supportive shellfor the PHY-MAC layer Like CoopMAC its scheme de-pends on the middle node for communication betweennodes-eMAC layer protocol has been changed to regulateinformation exchanges at the physical layer -e receivernode receives duplicate packets from source and helpernodes to decode the data [70]

410 MAC-Centric Approach -is cross-layer protocoltargets multimedia applications by using the MPEG-4

Journal of Computer Networks and Communications 27

scheme [21] It is characterized by 4 access categories(AC3ndashAC0) according to their priority in transmission It ismeant for supporting different QoS needs in upcomingvideo applications and enabling MAC layer differentiationfor H264 partitioning [18] Because of low bandwidth delayand other QoS challenges that cause inefficiency in trans-mittingmultimedia data inWMSNs a number of algorithmsdepending on IEEE 80211e have been proposed to supporttransmission of quality videos [72] Selecting an AC dependson the measures for QoS such as the loss rate and delay-us AC3 having the highest priority is mapped to theparameter set concept since the stream is sensitive to bit lossespecially in video transmission [21]

411 Adaptive Cross-Layer Forward Error Correction(ACFEC) In the ACFEC [73] model data packets are ex-changed among nodes through the access point (AP) whichoperates the infrastructure mode in which it adds FEC to videodata [72] -ese data are dealt by encapsulation using astreaming server to the receiver through thewireless AP as RTPpackets An adaptive FEC controller senses the category ofpackets out of the RTP header and recovers the header ofpackets from UDP -e encoder generates some error-cor-recting packets whose number is determined by the sourcepacket number of the block Multimedia transmissions aremonitored by the controller using MAC failure data and itscounter is increased by one if there is a failed transmission-econtroller uses the failure counter to change the packet numberproduced after transmitting a block [18] In case there are lostpackets it changes the redundancy rates and produces extrapackets to replace the lost ones and satisfy the requirements ofthe receiving node [72] -e FEC packet number is enhancedor reduced to satisfy the receiverrsquos requirements and stoppacket losses-is is done through accurate detection of packetlosses and adjustment of redundancy rates If all video datapackets are well received there will not be generation of FECpackets [73]

Advantages

(i) It promises a good QoS through packet loss re-duction and redundancy rate adjustment

412 Balanced Cross-Layer Fuzzy Logic (BCFL) DesignRouting Algorithm in WSNs A new fuzzy logic-basedrouting algorithm (BCFL) [74] was designed using disper-sion of the cross-layer parameter as the fuzzy logic inferencesystem input Each cross-layer parameter has a dynamicweight depending on the value of dispersion -e designcomes with some innovations as follows (i) for fuzzy logicinference system input the absolute parameter value issubstituted by parameter dispersion thereby significantlyreducing algorithmic complexity (ii) dispersion does notchange with the order of magnitude according to the dis-persion formula and (iii) the weight of the parameter de-pends on the size of its dispersion and the two are inverselyproportional to each other -is enables BCFL to have somenobility unlike other algorithms as per the following dis-tinguishing properties making it advantageous

(i) It has got simple if-then rules which remain con-stant even when the constraints increase

(ii) It is capable of dealing with many constraints withno increment in complexity

(iii) It has capacity to yield a more balanced solutionthan other algorithms

(iv) It is easily adaptable to changes in network con-ditions and topology even when the changes arefrequent like in underwater WSNs -e algorithm isuseable in choosing the CH in cluster-based routingprotocols [74]

413 Minimum Hop Disjoint Multipath Routing Algorithmwith Time Slice Load-Balancing Congestion Control SchemeMHDMwTS is a two-phased routing protocol comprisingpath build-up and path acknowledgment phases It consists ofmultiple sources each of which has three build-up disjointpaths namely primary alternate and back-up paths Onactivation the source node requests to build up a route to thenearest hop neighbor which is the path build-up phase [19] Inthis phase step one has source activation in which therequested node for path building adds its number and time-stamp and sends to the least hop-count neighboring node-isgoes on until the least time latency sink having the informationneeded to construct the primary route is reached [75] In steptwo there is path extraction when the new package fromanother path arrives whereby the extracted path is comparedwith the primary path If the node is shared the package will berejected or an alternative path will be searched for to get abackup through comparison of the preceding two paths Phasetwo is path acknowledgment where we have the third step inwhich the sink returns the ACK packet to the sender with pathinformation having nodes and their time information after ithas computed it using the timestamp [1819]

Advantages

(i) -e protocol minimizes end-to-end latency(ii) It enhances congestion control

414 Cross-Layer Optimal Design (CLOD) Authors in [76]come up with a CLOD for scheduling at the data link layerrouting at the network layer and controlling congestion atthe transport layer with an assumption of fixed link capacity-rough congestion control energy efficiency is improvedTransport layer congestion at the nodes is minimized bycompressed sensing (CS) in which transmitted bits are re-duced whereas optimally allocating resources decreasescongestion on the links at the data link layer CLODpromises minimized computational complexity and betterperformance in light traffic scenarios [77] Generating andstoring CLOD is more efficient compared to Gaussianrandom matrices It prolongs network lifetime and savesenergy

Advantages

(i) It does congestion control subsequently increasingenergy efficiency

28 Journal of Computer Networks and Communications

(ii) It reduces computational complexity(iii) It improves performance in light traffic scenarios

415 Challenges and Opportunities of the Cross-Layer Design(CLD)

(i) Different layers in a real network assume differentfunctions andor services A layer only communi-cates with its neighboring layers -e layered modeldegrades system performance because of many

features in wireless transmission QoS parameters inthe layers of WMSNs are shared by routing pro-tocols to optimize the performance Neverthelessthere is a need to design cross-layer models forincreased efficiency in routing [1]

(ii) -e physical layer plays a highly vital role in CLDRate adaptation and channel allocation take place atthe physical layer via signal processing to enhanceQoS End-to-end performance in wireless media isaffected by the changes therein given the effect on

Table 4 A summary of some of the reviewed cross-layer models

Protocol Layers Aims Comments QoS parameters

SDRCS MACPHY Routing real-time traffic Transmits multimedia traffic viacross-layer packet forwarding

-roughputlatency

SUIT Applicationtransport Transmission of non-real-time video Uses fuzzy logic to regulate

congestionTransmission rate

delay

CLASS Any twoDesign serves as a framework fordifferent implementations ofdifferent application scenarios

Scalable very efficient flexible andhas very low propagation delay

Propagation delayjitter

NLQS PHYMACnetwork

Permits packet-level servicedifferentiation as a function of

throughput packet error rate anddelay

Network layer QoS is enhanced -roughput delaypacket error rate

CMVT Applicationnetwork

Encodes video streams to video dataframes via MPEG-4 encoding anddoes route discovery and data

transfer

Greedy forwarding and rollback areused to find source-to-sink paths Energy

CoopMAC PHYMAC Offers spatial diversity among thethree nodes

Helper ID signal finds out howfeasible cooperative communication

is

-roughput ratedelay

M-CMAC PHYMAC Increases end-to-end throughputand packet delivery ratio

Euclidean distances between nodesare calculated and converted to the

data rate for a link

-roughputpacket delivery

ratio

CBCR PHYMACMinimizes control overhead andtime consumed in establishing thecooperative paths than M-CMAC

Energy consumption is moreuniformly distributed in a network

enhancing network lifetime

-roughputpacket deliveryratio energy

MAC-PHY MACPHYMeant to enable PHY layer

cooperation and maximize gains ofcooperation at the MAC layer

Leverages both spatial diversity andcoding gain -roughput delay

MAC-centric MACAPPL

Meant to support QoS needs in newvideo apps and enable MAC layer

differentiation for H264partitioning

Targets multimedia applications byusing the MPEG-4 scheme

Delay packet lossrate

ACFEC MACnetwork(UDP)

Meant to enhance the quality ofvideo streaming over 80211WLANs

and overcome packet losses

Adjusts redundancy rates toovercome channel fluctuations and

detect and reduce packet lossPacket loss rate

MHDMwTS mdash

Meant to provide reliable datatransfer with the multipath routingand load-balancing congestioncontrol method in WMSNs

More reliable than basic routingschemes for transport multimedia

data

Latency packagetransmit rate

BCFL mdash

Introduces dispersion into fuzzylogic-based routing and sets every

cross-layer parameter with adynamic weight

Can be used to select a CH in cluster-based routing protocols andproposes a dispersion formula

Node utilitydispersion

CLODDatalinknetworktransport

Prolongs network lifetime andachieves congestion control anddesigned for lightly loaded WSNs

Assumes fixed link capacity andintegrates compressed sensing

technology

-roughputaverage energy CS

error ratio

Journal of Computer Networks and Communications 29

the purposefulness of the protocols at the networklayer CLD offers solutions to conserving powerminimizing energy and controlling both the flowand congestion in the network as well as fault tol-erance so it is an opportunity for designers con-sidering other layers We further desire to developCLDs that shall make an impact on network oper-ation and get close attention [18]

(iii) CLD interface standardization the architecturemust deliver module functionality although thereexist some queries amongst modules about possibleinterfaces predicted by the necessity to share in-formation among remote protocol layers Technicalchallenges include developing designing andstandardizing of cross-layer interfaces and algo-rithms that satisfy the cross-layer optimization re-quirements among protocol layers

Table 4 summarizes these cross-layer models with a fewcomments

5 Future Research Direction

51 At theMACLayer In WMSNs achieving duty cycling istricky since video traffic is volatile Listening to adjacentnodes wastes energy We need to switch states from awake tosleep -is requires further research

Self-organizing networks and green communicationalthough most CRWSN energy-saving algorithms aredesigned to reduce node transmission power energyconsumption is inevitable during the operational mode ofBSs because of internal processing Scholars are trying withpolicy-makers to encourage a shift to greenWSNs to reduceoperational costs and carbon footprints With self-orga-nizing CRWSNs for automatic switch-off when idle energyis saved which calls for cognitive switching algorithms Sointegrating cognitive switching with radio resource allo-cation in CRWSNs is a promising research area [78]

52 At the Application Layer Considering a WMSN ar-chitecture end users access the network via the BS to andfrom which they send their sensed data We desire to getguaranteed QoS and energy efficiency by designing a newarchitecture in the form of hardware and software whichneeds to be investigated further for a fix of the same [1]

QoS and energy efficiency are very crucial in WMSNsand mostly real-time apps that require guaranteed band-width and throughput in their network lifetime Mostprotocols ignore base station (BS) mobility and WMSNnodes Traffic management telemedicine and battlefieldsurveillance apps require mobile nodes or BSs so designingdynamic routing protocols to be adaptable in these cir-cumstances is necessary

53 Cross-Layer Security of data transmitted on WMSNseg in military surveillance and e-commerce is key withQoS and energy efficiency WMSNs are vulnerable to attacks

like worms and sinkhole Moreover their computationpower is not that high complicating implementation ofstrong secure protocols -is needs further attention

For fairness and priority issues in CRWSNs nodes areassigned diverse priorities depending on importance andurgency of their data For better network performance thereis a need to develop fair resource allocation mechanismswith priority-based fairness in sensors-ere is still potentialin this research area [78]

We must develop multiobjective adaptive protocols tooptimize QoS metrics involved in routing for best trans-mission results to get a trade-off between different opti-mization metrics We need to design standardized cross-layer algorithms which are able to meet cross-layer opti-mization standards

6 Conclusion

We have reviewed the different QoS strategies for WSNs inthe context of IoT from the MAC layer and application layeras well as the cross-layer paradigm For the MAC layer wehave reviewed protocols for WSNs such as the hybrid of thecontention-free and contention-based MAC protocols andthose for WMSNs We further reviewed a number of ap-plication layer protocols including many machine-to-ma-chine requestresponse and publishsubscribe protocols Forsystem optimization cross-layer QoS strategies are veryimportant in wireless communication We have reviewed anumber of cross-layer strategies For all categories reviewedchallenges and opportunities are discussed Finally somepossible future directions are discussed for research andapplication showing a promising potential for future re-search in this relatively new area especially as more WMSNapplications emerge recently

Conflicts of Interest

-e authors hereby declare no conflicts of interest regardingthe publication of this work

Acknowledgments

-is research was funded by the National Natural ScienceFoundation of China (Nos 61602398 61672447 and61711540306)

References

[1] V Bhandary A Malik and S Kumar ldquoRouting in wirelessmultimedia sensor networks a survey of existing protocolsand open research issuesrdquo Journal of Engineering vol 2016Article ID 9608757 27 pages 2016

[2] I F Akyildiz T Melodia and K R Chowdhury ldquoWirelessmultimedia sensor networks applications and testbedsrdquoProceedings of the IEEE vol 96 no 10 pp 1588ndash1605 2008

[3] M Iqbal M Naeem A Ahmed M Awais A Anpalagan andA Ahmad ldquoSwarm intelligence based resource managementfor cooperative cognitive radio network in smart hospitalsrdquoWireless Personal Communications vol 98 no 1 pp 571ndash5922018

30 Journal of Computer Networks and Communications

[4] S A Alvi B Afzal G A Shah L Atzori and W MahmoodldquoInternet of multimedia things vision and challengesrdquo AdHoc Networks vol 33 pp 87ndash111 2015

[5] Q F Hassan A U R Khan and S A Madani Internet ofings Challenges Advances and Applications CRC PressTaylor amp Francis Group A Chapman amp Hall Group BocaRaton FL USA 2018

[6] Q Wang Y Zhao W Wang et al ldquoMultimedia IoT systemsand applicationsrdquo in Proceedings of the 2017 Global Internet ofings Summit (GIoTS) vol 2 Geneva Switzerland June 2017

[7] A Aslam and E Curry ldquoTowards a generalized approach fordeep neural network based event processing for the internet ofmultimedia thingsrdquo IEEE Access vol 6 pp 25573ndash25587 2018

[8] M Jridi T Chapel V Dorez L Bougeant and A Le BotlanldquoSoC-based edge computing gateway in the context of theinternet of multimedia things experimental platformrdquoJournal of Low Power Electronics and Applications vol 8no 1 p 1 2018

[9] H Noura A Chehab L Sleem andM N Rapha ldquoOne roundcipher algorithm for multimedia IoT devicesrdquo MultimediaTools and Applications vol 77 no 14 pp 18383ndash18413 2018

[10] T Balan D Robu and F Sandu ldquoMultihoming for mobileinternet of multimedia thingsrdquo Mobile Information Systemsvol 2017 Article ID 6965028 16 pages 2017

[11] H Shen and G Bai ldquoRouting in wireless multimedia sensornetworks a survey and challenges aheadrdquo Journal of Networkand Computer Applications vol 71 pp 30ndash49 2016

[12] A Ahmad M H Rehmani H Tembine O A Mohammedand A Jamalipour ldquoIEEE access special section Editorial optimization for emerging wireless Networks IoT 5G andsmart grid communication networksrdquo IEEE Access vol 5pp 2096ndash2100 2017

[13] B A Muzakkari M A Mohamed M F A Kadir andZ Mohamad ldquoRecent advances in energy efficient-QoS awareMAC protocols for wireless sensor networksrdquo InternationalJournal of Advanced Computer Research vol 8 no 38pp 212ndash228 2018

[14] M A Yigitel O D Incel and C Ersoy ldquoQoS-aware MACprotocols for wireless sensor networks a surveyrdquo ComputerNetworks vol 55 no 8 pp 1982ndash2004 2011

[15] A M Abbas and O Kure ldquoQuality of Service in mobile ad hocnetworks a surveyrdquo International Journal of Ad Hoc andUbiquitous Computing vol 6 no 2 p 75 2010

[16] M Q Shatnawi ldquoApplication layer protocols for the internetof things a surveyrdquo in Proceedings of the InternationalConference on Engineering amp MIS (ICEMIS) Agadir Mo-rocco September 2016

[17] V Karagiannis P Chatzimisios F Vazquez-gallego andJ Alonso-zarate ldquoA survey on application layer protocols forthe internet of things research motivationrdquo InternationalJournal of Advanced Research in Computer Science vol 8no 3 pp 1ndash10 2015

[18] M S Bernard T Pei Z Li and K Li ldquoQoS strategies forwireless multimedia sensor networks in the context of IoTrdquoLecture Notes of the Institute for Computer Sciences SocialInformatics and Telecommunications Engineering vol 4pp 228ndash253 2019

[19] A AlAmri and M Abdullah ldquoCross-layer quality of serviceprotocols for wireless multimedia sensor networksrdquo in Pro-ceedings of the International Conference on CommunicationManagement and Information Technology pp 649ndash658Cosenza Italy April 2017

[20] J Guo L Sun and R Wang ldquoA cross-layer and multipathbased video transmission scheme for wireless multimedia

sensor networksrdquo Journal of Networks vol 7 no 9pp 1334ndash1340 2012

[21] A Ksentini A Naimi and M Gueroui ldquoToward an im-provement of H264 video transmission over IEEE 80211ethrough a cross-layer architecturerdquo IEEE CommunicationsMagazine vol 44 no 1 pp 107ndash114 2006

[22] W Ye J Heidemann and D Estrin ldquoAn energy-efficientMAC protocol for wireless sensor networksrdquo in Proceedings ofthe 21st International Annual Joint Conference of the IEEEComputer and Communications Societies (INFOCOM 2002)New York NY USA June 2002

[23] J Polastre J Hill and D Culler ldquoVersatile low power mediaaccess for wireless sensor networks categories and subjectdescriptorsrdquo in Proceedings of the Second InternationalConference on Embedded Networked Sensor Systems Sen-Sysrsquo04 pp 95ndash107 Baltimore MA USA November 2004

[24] T Danmanee K N Nakorn and K Rojviboonchai ldquoCU-MAC a duty-cycle MAC protocol for internet of things inwireless sensor networksrdquo ECTI Transactions on ElectricalEngineering Electronics and Communications vol 16 no 2pp 30ndash43 2018

[25] T Van Dam and K Langendoen ldquoAn adaptive energy-efficientMAC protocol for wireless sensor networksrdquo in Proceedings ofthe 1st International Conference on Embedded NetworkedSensor Systems Los Angeles CA USA November 2003

[26] P Xie and J Cui ldquoR-MAC an energy-efficient MAC protocolfor underwater sensor networksrdquo in International Conferenceon Wireless Algorithms Systems and Applications (WASA2007) Chicago IL USA August 2007

[27] A Rahim N Javaid M Aslam U Qasim and Z A KhanldquoAdaptive-reliable medium access control protocol forwireless body area networksrdquo in Proceedings of the AnnualIEEE Communications Society Conference on Sensor Meshand Ad Hoc Communications and Networks (SECON) SeoulKorea June 2012

[28] H S Kwak S Ullah D H Kwak and C H Lee ldquoA powerefficient MAC protocol for wireless body area networksrdquoJournal of the Korean Institute of Intelligent Systems vol 8no 6 pp 131ndash140 2009

[29] C Omeni A J Burdett and M Park ldquoEnergy efficientmedium access protocol for wireless medical body area sensornetworksrdquo in Proceedings of the 4th IEEEEMBS InternationalSummer School and Symposium on Medical Devices andBiosensors pp 29ndash32 Cambridge UK August 2007

[30] L Sitanayah C J Sreenan and K N Brown ldquoER-MAC ahybrid MAC protocol for emergency response wireless sensornetworksrdquo in Proceedings of the Fourth International Con-ference on Sensor Technologies and Applications pp 244ndash249Mestre Italy July 2010

[31] Y Sun and D B Johnson ldquoRI-MAC a receiver-initiatedasynchronous duty cycle MAC protocol for dynamic trafficloads in wireless sensor networksrdquo in Proceedings of the 6thACM conference on Embedded network sensorsystemsmdashSenSysrsquo08 pp 1ndash14 Raleigh NC USA November2008

[32] M Buettner G V Yee E Anderson and R Han ldquoX-MAC ashort preamble MAC protocol for duty-cycled wireless sensornetworksrdquo in Proceedings of the 4th International Conferenceon Embedded Networked Sensor Systems SenSys 2006pp 307ndash320 Boulder Colorado USA October 2006

[33] S Boulfekhar and M Benmohammed ldquoSynchronous receiverinitiated MAC protocol for long-lived sensor networksrdquoComputers amp Electrical Engineering vol 40 no 2 pp 504ndash516 2014

Journal of Computer Networks and Communications 31

[34] S Mehta and K S Kwak ldquoH-MAC a hybrid Mac protocol forwireless sensor networksrdquo International Journal of ComputerNetworks amp Communications vol 2 no 2 pp 108ndash117 2010

[35] I Rhee A Warrier M Aia J Min and M L Sichitiu ldquoZ-MAC a hybrid MAC for wireless sensor networksrdquo IEEEACM Transactions on Networking vol 16 no 3 pp 511ndash5242008

[36] S Arshad A Al-sadi and A Barnawi ldquoZ-MAC performanceevaluation and enhancementsrdquo Procedia Computer Sciencevol 21 pp 485ndash490 2013

[37] Y Liu I Elhanany and H Qi ldquoAn energy-efficient QoS-aware media access control protocol for wireless sensornetworksrdquo in Proceedings of the IEEE International Confer-ence on Mobile ad hoc and Sensor Systems pp 9ndash11 Wash-ington DC USA December 2005

[38] J Decotignie ldquoWiseMAC an ultra low power MAC protocolfor the downlink of infrastructure wireless sensor networksrdquoin Proceedings of the Ninth International Symposium onComputers And Communications pp 244ndash251 AlexandriaEgypt June 2004

[39] R A Rashid ldquoDevelopment of energy aware TDMA-basedMAC protocol for wireless sensor network systemrdquo EuropeanJournal of Scientific Research vol 30 no 4 pp 571ndash578 2009

[40] H P Van Hoesel ldquoA lightweight medium access protocol(LMAC) for wireless sensor networksrdquo in Proceedings of theInternational Workshop on Networked Sensing Systems Bal-timore MD USA November 2004

[41] C Li P Wang H Chen andM Guizani ldquoA cluster based on-demand multi-channel MAC protocol for wireless multi-media sensor networksrdquo in Proceedings of the IEEE In-ternational Conference on Communications pp 2371ndash2376Beijing China May 2008

[42] J L Sobrinho and A S Krishnakumar ldquoQuality-of-service inad hoc carrier sense multiple access wireless networksrdquo IEEEJournal on Selected Areas in Communications vol 17 no 8pp 1353ndash1368 1999

[43] J P John ldquoProviding MAC QoS for multimedia traffic in80211e based multi-hop ad hoc wireless networksrdquo ComputerNetworks vol 51 no 1 pp 153ndash176 2007

[44] S Lofty and N Padmavati ldquoA survey on mobility basedprotocols in WSNsrdquo in Proceedings of the InternationalConference Communication and Manufacturing pp 12ndash15Hangzhou China July 2014

[45] S Deng L Shen and J Li ldquoMobility-based clustering pro-tocol for wireless sensor networks with mobile nodesrdquo IETWireless Sensor Systems vol 1 no 1 pp 39ndash47 2011

[46] M Khalil A Khalid F U Khan and A Shabbir ldquoA review ofrouting protocol selection for wireless sensor networks insmart citiesrdquo in Proceedings of the 24th Asia-Pacific Confer-ence on Communications (APCC) Ningbo China November2014

[47] O Message Q Telemetry and T Mqtt Committee Specifi-cation Draft 02 April 2014

[48] M Collina M Bartolucci A Vanelli-coralli andG E Corazza ldquoInternet of things application layer protocolanalysis over error and delay prone linksrdquo in Proceedings ofthe 7th Advanced Satellite Multimedia Systems Conference andthe 13th Signal Processing for Space CommunicationsWorkshop Livorno Italy September 2014

[49] A Talaminos-barroso M A Estudillo-valderrama L M RoaJ Reina-tosina and F Ortega-ruiz ldquoA machine-to-machineprotocol benchmark for e health applicationsmdashuse case re-spiratory rehabilitationrdquo Computer Methods and Programs inBiomedicine vol 129 pp 1ndash11 2016

[50] V Goyal O Pandey A Sahai and B Waters ldquoAttribute-based encryption for fine-grained access control of encrypteddatardquo in Proceedings of the 13th ACM Conference on Com-puter and Communications Security pp 89ndash98 AlexandriaVA USA October 2006

[51] M Singh M A Rajan V L Shivraj and P BalamuralidharldquoSecure MQTT for internet of things (IoT)rdquo in Proceedings ofthe Fifth International Conference on Communication Systemsand Network Technologies Gwalior India April 2015

[52] A Stanford-Clark and H L Truong MQTT for Sensor Net-works (MQTT-SN) Protocol Specification httpmqttorgnewwp-contentuploads200906MQTT-SN_spec_v12pdf 2013

[53] RFC ldquoExtensible messaging and presence protocol (XMPP)rdquoin Proceedings of the IETF pp 1ndash9 San Diego CA USAAugust 2005

[54] S Vinoski and I Technologies ldquoQueuing protocolrdquo IEEEInternet Computing vol 10 no 6 pp 87ndash89 2006

[55] J L Fernandes I C Lopes J J P C Rodrigues and S UllahldquoPerformance evaluation of restful web services and AMQPprotocolrdquo in Proceedings of the Fifth International Conferenceon Ubiquitous and Future Networks (ICUFN) pp 2ndash7 DaNang Vietnam July 2013

[56] M Asim ldquoA survey on application layer protocols for internetof things (IoT)rdquo International Journal of Advanced Researchin Computer Science vol 8 no 3 pp 996ndash1000 2017

[57] K An A Gokhale D Schmidt S Tambe P Pazandak andG Pardo-castellote ldquoContent-based filtering discovery pro-tocol (CFDP) scalable and efficient OMG DDS discoveryprotocolrdquo in Proceedings of the 8th ACM InternationalConference on Distributed Event-Based Systems New YorkNY USA July 2014

[58] E Web What is a web service pp 52ndash57 2010[59] M Laine RESTful Web Services for the Internet of ings

pp 2ndash4 2012 httpmediaTkk[60] IETF -e WebSocket protocol pp 1ndash71 2011[61] V Pimentel ldquoCommunicating and displaying real-time data

with websocketrdquo IEEE Internet Computing vol 16 no 4pp 45ndash53 2012

[62] Wikipedia 2019 httpsenwikipediaorgwikiStreaming_Text_Oriented_Messaging_Protocol2019

[63] K Raghavendra and S Nireshwalya ldquoApplication layer se-curity issues and its solutionsrdquo IJCSET vol 2 no 6pp 1266ndash1269 2012

[64] Y Xue B Ramamurthy andM C Vuran ldquoSDRCS a service-differentiated real-time communication scheme for eventsensing in wireless sensor networksrdquo Computer Networksvol 55 no 15 pp 3287ndash3302 2011

[65] C Sonmez S Isik M Y Donmez O D Incel and C ErsoyldquoSUIT a cross layer image transport protocol with fuzzy logicbased congestion control for wireless multimedia sensornetworksrdquo in Proceedings of the 5th International Conferenceon New Technologies Mobility and Security (NTMS) pp 1ndash6IEEE Istanbul Turkey May 2012

[66] T Melodia and I Akyildiz ldquoCross-layer QoS-aware com-munication for ultra wide band wireless multimedia sensornetworksrdquo IEEE Journal on Selected Areas in Communica-tions vol 28 no 5 pp 653ndash663 2010

[67] Q Wang and M A Abu-Rgheff ldquoCross-layer signalling fornext-generation wireless systemsrdquo in Proceedings of the 2003IEEE Wireless Communications and Networking 2003WCNC 2003 vol 2 pp 1084ndash1089 IEEE New Orleans LAUSA March 2003

[68] P Liu Z Tao S Narayanan T Korakis and S PanwarldquoCoopMAC a cooperative MAC for wireless LANsrdquo IEEE

32 Journal of Computer Networks and Communications

Journal on Selected Areas in Communications vol 25 no 2pp 340ndash354 2007

[69] H Shan H T Cheng and W Zhuang ldquoCross-layer co-operative MAC protocol in distributed wireless networksrdquoIEEE Transactions onWireless Communications vol 10 no 8pp 2603ndash2615 2011

[70] F Liu T Korakis Z Tao and S Panwar ldquoAMAC-PHY cross-layer protocol for wireless Ad-Hoc networksrdquo in Proceedingsof the 2008 IEEE Wireless Communications and NetworkingConference pp 1792ndash1797 IEEE Las Vegas NV USA March2008

[71] L Jacob and H R Shamna ldquoEfficient cooperative MAC androuting in wireless networksrdquo Transactions on Networks andCommunications vol 3 no 5 2015

[72] S Rao and K Shama ldquoCross layer protocols for multimediatransmission in wireless networksrdquo International Journal ofComputer Science amp Engineering Survey vol 3 no 3pp 15ndash28 2012

[73] L Han S Park S Kang and H Peter ldquoAn adaptive cross-layer FEC mechanism for video transmission over 80211WLANsrdquo in Proceedings of the 2009 International Conferenceon Internet pp 209ndash215 Las Vegas NV USA December2009

[74] N Li J Martınez and V H Dıaz ldquo-e balanced cross-layerdesign routing algorithm in wireless sensor networks usingfuzzy logicrdquo Sensors vol 15 no 8 pp 19541ndash19559 2015

[75] G Sun J Qi Z Zang and Q Xu ldquoA reliable multipathrouting algorithm with related congestion control scheme inwireless multimedia sensor networksrdquo in Proceedings of the2011 3rd International Conference on Computer Research andDevelopment vol 4 pp 229ndash233 IEEE Shanghai ChinaMarch 2011

[76] M Li Y Jing and C Li ldquoA robust and efficient cross-layeroptimal design in wireless sensor networksrdquoWireless PersonalCommunications vol 72 no 4 pp 1889ndash1902 2013

[77] J Yan M Zhou and Z Ding ldquoRecent advances in energy-efficient routing protocols for wireless sensor networks areviewrdquo IEEE Access vol 4 pp 5673ndash5686 2016

[78] A Ahmad S Ahmad M H Rehmani and N U Hassan ldquoAsurvey on radio resource allocation in cognitive radio sensornetworksrdquo IEEE Communications Surveys amp Tutorials vol 17no 2 pp 888ndash917 2015

Journal of Computer Networks and Communications 33

International Journal of

AerospaceEngineeringHindawiwwwhindawicom Volume 2018

RoboticsJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Active and Passive Electronic Components

VLSI Design

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Shock and Vibration

Hindawiwwwhindawicom Volume 2018

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawiwwwhindawicom

Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Control Scienceand Engineering

Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Journal ofEngineeringVolume 2018

SensorsJournal of

Hindawiwwwhindawicom Volume 2018

International Journal of

RotatingMachinery

Hindawiwwwhindawicom Volume 2018

Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Navigation and Observation

International Journal of

Hindawi

wwwhindawicom Volume 2018

Advances in

Multimedia

Submit your manuscripts atwwwhindawicom

Page 15: ReviewArticle - downloads.hindawi.com · ReviewArticle QoSStrategiesforWirelessMultimediaSensorNetworksinthe ContextofIoTattheMACLayer,ApplicationLayer,and Cross-LayerAlgorithms MuwongeSsajjabbiBernard

protocols -e contention-based protocol allocates achannel per node for sending the request message andfree channels are useable as control channels Forcontention-free protocols a control slot is allocatedwhen a network is deployed Here a slot per channel isallocated to a different node enabling requests to be sentwithout interfering with transmissions from othernodes -e contention-based protocol operates undertwo steps (i) Control channel assignment phase achannel is assigned per node for transmitting a request(REQ) message A channel is assigned to multiplenodes and nodes are evenly distributed to availablechannels for congestion control (ii) Request trans-mission phase nodes with data send a REQmessage viathe assigned control channel to notify the CH whichresponds with an ACK message -e contention-freeTDMAFDMA protocol too has two phases includingthe control slot assignment and request transmissionphases For unreliable channel conditions and lowtraffic load the contention-based protocol is fit for usebecause of less collisions and congestion whereas thecontention-free protocol is preferable in heavy trafficload and reliable channel applicationsScheduling phase from the above phase a schedule isgenerated by the CH for coordination and this enablesnodes to transmit and broadcast their data via controlchannels-e schedule for broadcasting only comprisesdata for nodes assigned to a channel -e schedule isrebroadcast for transmission reliability enhancementbut for energy efficiency the nodersquos transceiver isswitched off when all the schedule is completely de-livered till it is the time for transmitting again-e timeslot and data communication radio channel are allincluded in the schedule generated for each nodeData transmission phase on getting the schedule thesensor node sends its data as per the time slot and thechannel assigned -e time slot is divided into datatransmission andACK sections ACK supports link layererror control Using implicit selective repeat ARQ theCH sends an ACK message for all received packets Ifthere is no ACK the packet is assumed lost and hence aretransmission during the subsequent interval is done-is enhances reliability of transmissions though excessdelay is possible here which is also intolerable for real-time critical apps like video or audio -is gives rise to ahybrid MAC protocol to exploit the unused spectrumduring data transmission sessions [41]Advantages

(i) It achieves energy efficiency by scheduling mul-tichannel media access to reduce channel con-tention and remove collision idle listening andoverhearing

(ii) It maximizes network throughput by dynamicallyallocating time slots and channels for nodes byexecution of a traffic-adaptive QoS-aware sched-uling algorithm depending on needs for QoS andtraffic data information

(iii) It increases transmission reliability by in-corporating spectrum-aware ARQ to avoid spec-trum wastage

223 EQoSA -is protocol is developed for video andimage transfer across WMSNs in a hybrid mode withprovision of QoS Concerning the active sensor nodes andtheir traffic loads it employs dynamic session sizes andchanges bitmap-assisted (BMA) MACrsquos fixed session sizeNodes declare any data available for transmission duringcontention time -is forces the cluster head to make aschedule of how slots have been assigned It communicatesthis schedule to the nodes giving each node a given numberof slots per session and so enabling the busty traffic to beaccommodated [14] -e protocol does not perform wellwith DiffServe mechanisms but does well with schemessupporting busty traffic loads and strong cluster heads arenecessary in WMSNs that can assign and broadcast slots

224 QoS-Supported Energy-Efficient MAC (QE-MAC)Based on the IEEE 80211e standard this protocol ensuresfairness and has low latency and jitter and better efficiencyon energy usage than other QoS-aware MAC protocols forWMSNs It operates under a two-phased routine

-e first phase introduces the innovating prioritymechanism in IEEE 80211e QoS and several data types areassigned priorities Its coordination function is a combi-nation of distributed contention-based channel allocationand centralized polling-based channel access mechanisms

-e second phase preserves the nodesrsquo energy usingdynamic duty cycling -ere is an exchange of RTSCTSpackets to supplement CSMACA by muting nodes near thesenderreceiver during packet duration [42]

Advantages

(i) It ensures fairness(ii) It has low latency and jitter and better efficiency on

energy usage than other QoS-awareMAC protocolsfor WMSNs

225 Black-Burst (BB) Contention A contention-basedMAC scheme for enhancing instantaneous QoS access tocarrier-sense WSNs is proposed in [42] In this schemenodes contend for the channel until it is free Real-timepackets are given higher priority over non-real-time packetsand if a node is transmitting real-time packets it is prior-itized over other nodes [15] Voice and video are some of thereal-time apps in consideration here which periodicallyaccess the common radio channel when transmitting -eseneed constrained end-to-end delay of packets at the MAClayer -is protocol majorly does the following [42]

(i) Prioritization of real-time traffic(ii) Implementation of the round-robin mechanism in

real-time nodes(iii) Bounded access delays among real-time packets

Journal of Computer Networks and Communications 15

(iv) Implementation of real-time communication withvarying needs of bandwidth good for multimediatraffic

Amechanism for dynamically assigning packet prioritiesbased on their deadlines and the traversed hops is proposedin [43] -e ReAllocative Priority (ReAP) mechanism ismeant to provide QoS at the MAC layer to video traffic -isscheme introduces adaptive TXOP by modifying the TXOPdynamically based on the queue length [15] Other protocolsin this category include dynamic duty cycle and adaptivecontention window-based MAC protocol frame sharing(FRASH) MAC protocol and real-time independentchannels (RICH) MAC protocol

Advantages

(i) It can implement real-time communication withvarying needs of bandwidth

23 Challenges and Opportunities(a) One of the challenges in satellite transmission is

cross-talk leading to interference in different fre-quency bands It is necessary to increase the rise andfall times as well as protect sensitive nodes as wecircumvent floating ones if we are to regulate cross-talk as in satellite communication therefore Wefurther need to investigate the possibility of usingdifferential signaling -is constitutes an opportu-nity for further research

(b) Another challenge is that the hidden and exposedterminal problems are still noticeable despite the useof RTSCTS solutions in MAC since collisions arestill possible from transmissions of separate nodesAt the MAC layer ACKs might block efficientchannel usage by prohibiting exposed terminalsfrom reusing the channel and may be eliminatedleaving no proof of sender data packet receptionUsing a directional antenna is one of the proposedtechniques of high-speed WSNs like IEEE 80211-is constitutes an opportunity for further study intothese issues for better solutions though

(c) Another challenge is that since nodes in contention-based MAC protocols are mobile there are manycollisions by the packets and schedule in-consistencies in schedule-based protocols in the two-hop neighborhood information when nodes areentering or leaving [44] Deng et al [45] propose amobility-based clustering (MBC) protocol for mo-bile nodes to enhance the packet delivery rate -isprotocol in comparison with CBR has 25 chancesof decreasing collision-induced packet losses and50 chances in comparison with low-energyadaptive clustering hierarchy (LEACH)

(d) -e process of analyzing and evaluating routingperformance for WMSNs is quite complex becauseof the bursty multimedia traffic Because of the wayencoding scheme frames are structured as well as thechanges in between the scenes that occur naturally

the compressed video frames display some substantialburstiness on several time scales -ere will also befrequent route recalculation that is as well energy in-tensive for nodes frequently capturing and transmittingenergy -e solution here could as well possibly be theuse of the MBC protocol Route recalculation tooshould be dynamic to cater for the mobile nodes whichmight be in and out of the range at different instants oftime Recently the use of smart dumpsters for effectivewaste management in smart cities has also been pro-posed as a possible solution [46]

(e) For WSNs their routing protocols do not look atmultimedia applications which require enormousbandwidth and processing energy and should betransmitted in real time with utmost fidelity Extrachallenges in routing are imposed in WMSNs indeployment of heterogeneous sensors since therecould be audio video and still pictures in addition toscalar data during transmission Heterogeneous datahandled also have different QoS requirements andwith many different business needs different ser-vices with varying requirements give rise to a bigchallenge for the routing design -ere is a need todesign protocols that can support WMSNs such asEQoSA [14] for this purpose

Table 2 summarizes the reviewed MAC protocols forWSNs and the ones designed for suitability in WMSNs

3 QoS Strategies at the Application Layer

We discuss many protocols used to solve different needs ofcommunication between machines

31 ConstrainedApplication Protocol (CoAP) It is a requestresponse protocol designed by the Internet Engineering TaskForce (IETF) for constrained devices such as those havinginsufficient RAM or CPU and WPANs with low-powerconstraints resulting in poor transfer of packets andenormous overhead -e protocol was designed for ma-chine-to-machine (M2M) apps and system automation toreduce bandwidth requirements increase packet transferdecrease the overhead and make work simpler leading tolightweight implementation HTTP commands like GETPOST PUT and DELETE are used in a client-server ar-chitecture Using the publishsubscribe architecture CoAPcan support multiple users and hence yield better perfor-mance via multicasting enabling asynchronous messageexchange -is protocol can support both multicasting andunicasting via UDP and hence is a reliable and simpleprotocol -is is done via the messaging sublayer (for re-liability based on UDP) and the requestresponse sublayer(for communication and interaction) [16]

It uses the following messages

(i) Confirmable message this guarantees reliabilityusing ACK messages sent by the receiver node-ese could be synchronous or asynchronous ifthere is a need for more computational time

16 Journal of Computer Networks and Communications

(ii) Nonconfirmable message here no ACK messagesare sent back

(iii) Acknowledgment (ACK) message this confirmsreceipt of a confirmable message

(iv) Reset message this confirms receipt of an un-processed message in case any critical parts nec-essary for interpreting the message are missing

(v) Piggybacked response the receiver directly re-sponds while getting the ACK message

(vi) Separate response the reply by the receiver is dis-crete from the ACK message

CoAP has no in-built security features It is secured bythe datagram transport layer security (DTLS) that runs ontop of UDP though it was not designed for the IoT -ismakes its suitability debatable since DTLS offers no supportfor multicasting It also needs more packets for handshakingthereby increasing the traffic volume and computationalresources and so shortening the lifespan of mobile devicessince they use batteries [17]

Advantages

(i) It can support multiple users and hence yieldbetter performance via multicasting

Table 2 Comparison of MAC layer protocols used in QoS provisioning

Protocol Features QoS parameter Comments Access Priorityassignment Energy eff Designed for

S-MAC[22] Collision avoidance Energy latency Requires periodic

synchronization CSMA Hybrid Medium WSNs

B-MAC[23] Collision avoidance Energy delay

throughput Flexible interface CSMA Dynamic High WSNs

T-MAC[25] Collision avoidance Energy Complex nonscalable CSMA Dynamic High WSNs

C-MAC[13]

Dynamic channelallocation Energy Multichannel MAC CSMA Dynamic Medium WSNs

CU-MAC[24]

Multichannelmechanism

-roughputlatency High packet delivery ratio TDMA Dynamic mdash WSNs

A-MAC[39]

Per flow serviceguarantees Delay m-ary tree model

dynamic prioritiesTDMACSMA Dynamic Medium WSNs

ER-MAC[30]

Emergencyresponse

Energy delaydelivery ratio

Adapts to traffic andtopology changes

TDMACSMA Hybrid Higher than Z-

MAC WSNs

RI-MAC[31] Collision detection Delay packet

delivery ratioHigh packet delivery ratio

and throughput CSMA Dynamic Medium WSNs

SRI-MAC[33] Collision avoidance Energy delay Good data transfer and

energy consumption TDMA Dynamic High WSNs

H-MAC[34] Slotted sleep time -roughput

latencyData transmission only in

sleep timeCSMAAloha Dynamic Medium WSNs

Z-MAC[36] Collision avoidance Energy delay

throughputVery good channel usage

and error syncCSMATDMA Dynamic Medium WSNs

Q-MAC[37]

Intrainternodescheduling Energy High QoS improved

energy efficiency CSMA Hybrid High WSNs

X-MAC[32]

Overhearingavoidance

Energy delaythroughput

Reduces excessivepreamble by strobing CSMA Adaptive

static Medium WSNs

L-MAC[40] Collision avoidance Network

lifetime energyReduces the number oftransceiver switches TDMA Static Medium WSNs

WiseMAC[38]

No downlinkchannel collision Power delay Significantly low-power

consumptionAlohaCSMA Static Higher than

ZigbeeInfrastructure

WSNsAR-MAC[27] Collision avoidance Energy delay Uses data packets and

control packets TDMA Static Higher thanIEEE 802154 WBANs

BB-MAC[42]

Real-time trafficprioritization Delay Employs round robin in

real-time nodesCSMATDMA Dynamic High WMSNs

COM-MAC [41] Collision avoidance Packet delay

throughputBalances reliability and

retransmissionTDMAFDMA Dynamic High WMSNs

EQoSA[14]

Busty traffic loadaccommodation Energy Provides QoS support for

video and imagesTDMACDMA mdash High WMSNs

QE-MAC[42] Collision avoidance Energy jitter

latencyRuns central polling-like

channel access CSMA Dynamic High WMSNs

R-MAC[26] Collision avoidance Energy delay Reservation based mdash Static High Underwater

SNs

Journal of Computer Networks and Communications 17

(ii) It enables asynchronous message exchange(iii) It is a reliable and simple protocol

32 Message Queue Telemetry Transport (MQTT)Developed by IBM and later improved by OASIS with anintention of decreasing bandwidth MQTT is a simple open-source publishsubscribe protocol analogous to the client-server version It is appropriate for lightweight IoT appli-cations and M2M transmissions in constrained situationslike limited power computation capability and memory orbandwidth -e network bandwidth and the computationalresource usage are decreasing since clients are not updatedmaking publishsubscribe protocols outperform requestresponse protocols via IoT requirements [16] -is protocolcan run over TCPIP and so has the following advantages

(i) Delivering of packets with reliable guarantee(ii) Enabling multicasting(iii) Remote device session establishment(iv) Reduction of transport overhead and protocol ex-

change to minimize network traffic(v) Provision of the notification mechanism when

anomalies occur [47]

MQTTrsquos design targets economical use of bandwidthand battery so it has a major application in FacebookMessenger MQTT ensures QoS and reliability of messagedelivery by providing the following three QoS levels

(1) Fire and forget a message is sent once depending onthe networkrsquos best effort and there is no need forACK (lowest QoS level)

(2) Delivered at least once a message is sent at least onceand an ACK message is needed to avoid duplicates

(3) Delivered exactly once by a four-way handshake itensures a message is sent exactly one time (highestQoS level)-e architecture it has three key components (i)publishersmdashlightweight nodes sending data to thebroker before returning to sleep (ii) broker whichorganizes sensory data in topics and sends them tointerested subscribers and (iii) subscribersmdashIoTapps interested in data sent by sensors through thebroker as shown in Figure 3

MQTT is a simple lightweight protocol that is betterthan CoAP via low and high packet loss probability interms of throughput and exhibits lower latency in lowsampling rates [48] It supports a number of imple-mentations for embedded devices and mobile applicationsand offers web-socket support too Nevertheless it hassome shortcomings

(i) It is only employed in simple data types(ii) It exhibits high latency in high sampling rates(iii) It is unsuitable for real-time apps in such cases(iv) It does not provide automatic discovery andor

sufficient security at the protocol level [49]

33 Secure Message Queue Telemetry Transport (SMQTT)-is is an extension of MQTT that is based on lightweightattribute-based encryption using elliptic curve cryptogra-phy -e protocol has three entities as shown in Figure 4(i) publisher that publishes information under a certaintopic (ii) subscriber that obtains the same topic in-formation via a broker and (iii) public key generator(PKG) or broker that acts as the trusted third party -eprotocol has four phases namely (a) set-up phase (b)encrypt phase (c) publish phase and (d) decrypt phase

(a) Set-up phase publisher and subscriber provide aunique ID with attributes for registration with thePKG -is generates a secret master key set andpublic parameters that it publishes with set U A1A2 An For ciphertext policy-attribute-basedencryption (CP-ABE) the PKG gives a key set todevices with attributes [50]

(b) Encrypt phase based on the access tree the publishercreates an access policy An access tree is sent by thepublisher to the PKG for generating the key-gen-eration policy in the key policy- (KP-) based ABEalgorithm -e publishers know beforehand thetopics and subscribers to access as well as the rulesfor access Subscribers are issued with private keysfor all communication sessions Publishers in theCP-ABE algorithm come up with access trees andrules where the payload is encrypted by the pub-lisher who also issues more info needed for de-cryption and the rules With the help of public keycryptography data are encrypted giving rise to ci-phertext on the basis of KPCP-ABE and the receivergets SUBACK from the broker [51]

(c) Publish phase ciphertext is embedded in theSPublish command as payload and topic name isdefined in the header by the publisher -e brokerreceives the SPublish packet and replies withPUBACK prompting the publisher to respond withthe PUBREL packet -e broker then broadcasts totopic subscribers and deletes the data forwarding thePUBCOMP packet to the sender node

(d) Decrypt phase the encrypted information isdecrypted by the subscriber with the help of a secretkey Considering KP-ABE on verification that itsatisfies the access policy the subscriber invokes thePKG for a corresponding key which it sends afterverifying the request With CP-ABE subscribers usethe private key for decryption after authenticationthereby enabling offline interaction [51]Advantages

(i) -e information being encrypted makes it moresecure

34 MQTT for Sensor Networks (MQTT-SN) -is is a ver-sion of MQTT designed to operate under environments oflow bandwidth high link failures and short message lengthIt is a publishsubscribe (pubsub) protocol designed for

18 Journal of Computer Networks and Communications

UDP-based WSNs to visualize communication for low-power devices [5152] It comprises a publisher device actingas the MQTT-SN client to forward communications to agateway that also changes the MQTT-SN message to theMQTTmessage before delivering it to the broker that sendsit to MQTT-SN subscribers [51]

It has a gateway for changing protocols from MQTT toMQTT-SN (Figure 5) [52] -is protocol supports nodes in

the sleeping mode using an offline keep-alive procedure thatcan be used by battery-operated devices to get into thesleeping mode To support sleeping clients MQTT-SN has anew offline keep-alive procedure with which battery-oper-ated devices can go to the sleeping state in which theirmessages are buffered at the server until they are awake InMQTT-SN clients with no preconfigured gateway addresscan figure out the real address of their operational gateway

Publisher

Subscriber

Subscriber

Broker

Publisher

4G

Figure 3 Architecture of the MQTT protocol

Register by mentioning URI Register by mentioning URI

Access tree

Set-up

Publisher ReceiverPKG(broker)

Key management Key management

Encrypt data using public parametersand random key This is done using

CP-ABE Publish it as

Subscribe (topics)SUBACK

S-Publish (TopicEncrypt (data))S-Publish (TopicEncrypt (data))

Encryptand publish

PUBACK

PUBACK

PUBRELPUBRELPUBCOMP

PUBCOMP

Figure 4 SMQTT protocol

Journal of Computer Networks and Communications 19

within the network through a discovery procedure Simul-taneously we may have several gateways in a single wirelessnetwork that can be cooperative via load sharing

MQTT-SN architecture it is composed of MQTT-SN clients gateways and forwarders (Figure 5)rough the MQTT-SN gateway (GW) the MQTT-SN protocol is used by the MQTT-SN clients toconnect to the MQTT server Integration betweenthe MQTT-SN GW and the MQTT-SN server mightbe possible or not Translating between the MQTT-SN and the MQTT for a stand-alone GW is doneusing the MQTT-SN protocol If a GW is detachedfrom the network it can be accessed by the MQTT-SN clients through a forwarder is just encapsu-lates the received MQTT-SN frames that are on thewireless side before forwarding them to the GW un-altered It then decapsulates the received frames tooresending them to the client as sent by the GW [52]ere are 2 types of gateways (a) transparent gatewayand (b) aggregating gateway as shown in Figure 6

(i) Transparent gateway this launches and sustains adirect MQTT connection between the MQTT-SNclient and the MQTT server which is entirely re-served for end-to-end client-server messaging enumber of servers and that of MQTT-SN clientsconnected to the GW are the same is transparent

GW translates syntax between these protocols and iseasier to implement compared to the aggregatingGW But the MQTT server must support dierentsessions for all connected clients Some MQTTservers are congured to enable few parallel sessions

(ii) Aggregating gateway this maintains a single MQTTconnection to the server on behalf of all clients Allexchanged messages from the MQTT-SN client to theaggregating GW are directed to the GW which sortsthe information to forward to the server e aggre-gating GW decreases the number of MQTT sessionssimultaneously supported by the server in WSNs withmany battery-operated sensors and actuators (SAs)having limited storage and processing capabilities It ishowever more complex to implement compared to atransparent GWe general format of the MQTT-SNprotocol message has two or four message headeroctets and n octets of variable header that is optionaldepending on the message in question MQTT-SN has256 message types (MsgType) with those from0X1Eminus 0XFD 0X19 and 0XFF reserved for futureusage [51]

Advantages

(i) e protocol can operate under environmentsof low bandwidth high link failures and shortmessage length

MQTT-SN gateway

MQTT-SN

MQTT

MQTT-SN

MQTT-SN

MQTT-SN

MQTT-SN client

MQTT-SN clients

MQTT broker

MQTT-SN client

MQTT-SN client

MQTT-SN gateway

MQTT-SN forwarder

Figure 5 MQTT-SN architecture

20 Journal of Computer Networks and Communications

(ii) It is a pubsub protocol designed for UDP-basedWSNs to visualize communication for low-power devices

(iii) It can be optimal for low-cost battery-operatednodes having low processing power and lessstorage

(iv) It supports nodes in the sleeping mode using theobrvbarine keep-alive procedure used by battery-operated devices

35e Extensible Messaging and Presence Protocol (XMPP)e standardization of this protocol by the IETF made itpopular for usage over the Internet especially in instantmessaging and chatting platforms It is however susceptibleto spamming attacks and its failure to have worldwidesupport has made Google too withdraw its support of thestandard [17] e protocol is mainly useable in the client-server architecture but can also be applicable in client-clientXMPP communication [53] Among the application layerprotocols XMPP is the only one that supports both asyn-chronous (publishsubscribe) and synchronous (requestresponse) services while running over TCP but can still runover other protocols

Given the simplicity of its implementation and ease ofusability in XMPP instant messaging (IM) is suitable forXMPP and is implemented using the message stanzaltmessagegt ese stanzas are usable in one-to-one ormultiuser sessions or chatting or in error reporting andsending headlines Attribute type of the stanza denes themessage type e attribute type could be a chat group chatheadline normal or error An instant messaging network inXMPP can support many XMPP clients and servers with agateway congured to enable other IM network users toconnect to other XMPP users e XMPP server mainlycontrols sessions and XML streams to clients and servers inaddition to routing XML stanzas in the streams [53] eTCP ports 5269 and 5222 have been respectively registeredfor the XMPP server-to-server connections and XMPP clientconnections by the Internet Assigned Numbers Authority(IANA) XMPP enables specication of XMPP ExtensionProtocols (XEP) to enhance its usefulness [17]

XMPP security clients connect to the server usingtransport layer security (TLS) and simple authentication andsecurity layer (SASL) protocols which are at the same timeused by the servers for interdomain communications Fordata condentiality and integrity TLS will encrypt thestreams of XML for all sessions To conrm the identity of anyclient that needs to access the server the SASL protocol doesthe authentication of the XML stream For any communi-cating session with a server to start the client needs to resolvethe serverrsquos DNS hostname For protection of the server of theclient from third-party attacks the server maintains the au-thentication details secret ie IP address and access methodand only original connections of the server are required

In server-to-server communications the SASL protocolis employed for authentication and condentiality Server-to-server communications may as well be deactivated from aserver by an administrator depending on the organizationalpolicy To guard against domain-spoong attacks serverdialback can be a good solution since servers support itsecuring more the XML stanzas Nevertheless it cannot beused to authenticate secure or encrypt in-server streamsand the server identication resulting therein is not strongenough to be relied on It further cannot secure against DNSpoisoning attacks and IP session hijacking for remote do-mains [53] High-security domains can use TLS and SASL

Advantages

(i) It is popular for usage over the Internet especially ininstant messaging and chatting platforms

Disadvantages

(i) It is susceptible to spamming attacks(ii) It has failed to have worldwide support making

Google withdraw support of the standard(iii) It is not secure against DNS poisoning attacks and

IP session hijacking for remote domains

36 Advanced Message Queuing Protocol (AMQP) isprotocol can be utilized by a number of platforms formessage exchange since it is open source and written inmany languages It is a publishsubscribe model depending

MQTT-SNclients Broker

MQTT

(a)

MQTT-SNclients Broker

(b)

Figure 6 (a) Transparent gateway and (b) aggregating gateway

Journal of Computer Networks and Communications 21

on a reliable and efficient messaging queue with capacity forinstant messaging or e-mail for different applications es-pecially in IoT environments It differs from others in thesense that message types can be specified and have theirsource traceable and how to do trade-off between perfor-mance security and reliability For reliability of messagetransmission AMQPmakes use of some delivery guaranteeslike at most once at least once and exactly once It can alsohowever utilize the TCP transport layer for guaranteeingreliability -e AMQPrsquos publishsubscribe method has twocomponents namely (i) exchange queue meant for routingof messages to an appropriate order in the queue and (ii)message queue that acts as storage pending their delivery toreceivers [16] In this protocol model the broker can makerouting decisions contrary to othermessaging systems whereapps using the queue have in them embedded the decision-making logic for recipients or senders -e developer mayneed to check each of the affected apps so as to change thelogic for routing and delivery of the messages [54]

Advantages

(i) It is convenient(ii) It has capacity for interoperability between dif-

ferent vendors(iii) It can use open standards to enable connection

between business partners(iv) It can enable innovations based on AMQP foun-

dations [55](v) It can enable supporting of mission-critical ap-

plications in e-commerce(vi) It can guarantee reliable delivery of messages

Disadvantages

(i) It is not very appropriate for real-timeapplications

(ii) It may not offer automatic discovery(iii) Interoperability is not guaranteed(iv) It lacks open-source libraries for constrained

gadgets [49]

37 Data Distribution Service (DDS) A product of theObject Management Group (OMG) DDS is a pubsubprotocol meant for M2M communications with two sub-layers -ese are the data-centric pubsub sublayer re-sponsible for connecting anonymous data publishers tosubscribers and the data-local reconstruction sublayer that isoptional and integrates DDS into the application layer [56]-e DDS architecture connects system components asshown in Figure 7 Publishers and subscribers are discon-nected with respect to (i) time (ii) space (iii) flow andbehavior (iv) platforms and (v) programming languages[57] A DDS data publisher generates topics and publisher-subscriber coupling is represented as topic name data typeschema and publishers and subscribersrsquo QoS attributesDelivering of data is done by the publisher layer -e datawriter and publishers together make a decision about anynecessary alterations for subsequent forwarding to thesubscribers who need to send data to IoT applications Data

readers are responsible for delivering topics to the sub-scribers after reading through them [56]

Some of the major components of the DDS architectureinclude the following [57]

(i) Topic this is a logical path specifying publicationand subscriptionrsquos data type between DataWritersand DataReaders For a successful session to begintopic names types and DataWriters and Data-Readersrsquo QoS should correspond

(ii) Domain this is a logical setting for transmissionthat can be utilized for separating and optimizingnetwork transmission in a setup of similar appli-cations If DDS apps share a domain ID then theycan exchange data among themselves

(iii) DataWriter and DataReader these are terminalobjects for writing and reading messages re-spectively to and from a universal data space

(iv) Participant this is an object that stands for apublisher or subscriber of the DDS app in a do-main acting like a container for other objects

Advantages

(i) It is highly reliable and offers QoS support(ii) It is highly scalable(iii) It is appropriate for real-time apps(iv) It is fault tolerant(v) It has high level of interoperability(vi) It does not need brokers that may act as

bottlenecks(vii) It offers automatic discovery(viii) It has a user-defined data structure for topic(ix) It supports publisher and subscriber decoupling

Disadvantages

(i) It is originally developed for only stand-aloneLANs

(ii) A lot of memory is needed(iii) It has no open-source libraries to support con-

strained devices(iv) It is hard to design and configure compared to

other protocols [49]

38 RESTful Services REpresentational State Transfer (REST)-is is an architectural style based on a set of principles thatdefine and address networked resources It offers web servicesthat enable transmission from the HTTP device to the devicewithin the IoT architecture and enables clients to reach theserver [16]-is protocol utilizes HTTP commands to backupthe requestresponse messaging model HTTP is a popularand secure protocol in the WWW using TLSSSL [16] ARESTful application has the following characteristics

(i) State and functionality are separated into distrib-uted resources

(ii) All resources are distinctively named by the use ofunique HTTP commands (GET POST PUT orDELETE over the Internet)

22 Journal of Computer Networks and Communications

(iii) It is a clientserver stateless layered protocol thatsupports caching REST uses the same methods asHTTP for all requests or responses with POST andGET for creating and retrieving resources PUT forupdating and changing the state of resources andDELETE for removing resourcesAdvantages of REST over arbitrary web services(SOAP) in the IoT context

(i) It reduces overhead(ii) It reduces parsing complexity(iii) It is stateless(iii) It integrates more tightly with HTTP [58](iv) It is easy to implement and learnis together

with its natural t over HTTP makes it apreferred method for Web 20 applications toshowcase their data and secure M2M sessionsin the IoT

(v) Applications that support RESTful web ser-vices exhibit better performance than the re-source-constrained WSN nodes [59]

(vi) REST is important for the IoT since it supportscommercial M2M cloud platforms

(vii) Its implementation in smart phones and tabletapps is also easy since it needs only an HTTPlibrary that is accessible to all operating systemplatforms [56]

Disadvantages

(i) Although it enjoys wide usage in commercialM2M platforms it is not likely to dominatebecause it is hard to implement

(ii) Its HTTP usage makes it incompatible withconstrained M2M devices leaving its usage toend applications

39 WebSockets It is a web-based protocol operating on asingle TCP channel and providing full-duplex transmissions

It permits secure duplex communication between a clientexecuting unreliable code in a restricted environment and adistant node communicating from that code while using theorigin-based security model employed by web browsers overTCP [60] e protocol is not a requestresponse model andis not a publishsubscribe model So unlike other protocolsa web-socket session is established by the client initiating ahandshake to the server to launch the session is onceestablished a full-duplex client-to-server connection beginsasynchronously and runs until terminated by either party[56] It gives a model for browser-based apps utilizing bi-directional communication with no need of many HTTPconnections to servers

WebSocket comprises two parts ie (i) the handshakehaving a clientrsquos message and the serverrsquos handshake responseand (ii) the data transfer HTTP headers should be repeated byapplications in all client requests and server responses arisingfrom the endless polling which might increase the commu-nication overhead subject to the application [61] WebSocketcan be used to come up with scalable real-time web appli-cations since it can enable full-duplex transmission via onesocket across the Web Its security can be ensured with thehelp of TLSSSL is protocol is suitable for resource-con-strained devices and IoTapplications will not have a good userexperience with it since it operates a clientserver architectureNevertheless it can outcompete all TCP-based protocolsbecause of the following advantages

(i) It is t for real-time transmission(ii) It has improved security(iii) It minimizes transmission overhead(iv) It can provide ecient messaging systems using

WebSocket ApplicationMessaging Protocol (WAMP)[17]

310 Streaming (Simple) Text-Oriented Messaging Protocol(STOMP) is is a TCP-based protocol designed for textmessage-oriented middleware (MOM) with the use of

Topic

Datareader

Datareader

Datareader

Datawriter

Datawriter

Datawriter

Subscriber PublisherPublisher

Data bus (DDS domain)

Subscriber

Topic Topic

Figure 7 DDS architecture

Journal of Computer Networks and Communications 23

HTTP-like commands for interoperability among platformslanguages and brokers Data communication takes placebetween the client and the broker in line with commands(like CONNECT DISCONNECT ACK NACK SUB-SCRIBE UNSUBSCRIBE SEND BEGIN COMMIT orABORT) then headers in the form ltkeygt ltvaluegt (one perline) then a blank line and then body content terminatingin a null character Server-client communication takes placeover a MESSAGE RECEIPT or ERROR frame with headerand body content format being analogous MOM productssupporting STOMP include Apache ActiveMQ or FuseMessage Broker HornetQ Open Message Queue(OpenMQ) RabbitMQ and syslog-ng through its STOMPdestination plugin [62] In addition to these other appli-cation layer protocols include simple media control protocol(SMCP) lightweight local automation protocol (LLAP)simple sensor interface (SSI) lightweight M2M (LWM2M)XMPP-IOT and simple object access protocol (SOAP)

Advantages

(i) It is lightweight for constrained networks(ii) It has flexibility to choose quality of services with

the given functionality(iii) It is standardized by OASIS Technical Committee(iv) It is easy and quick to implement

Challenges and opportunities

Here security threats are imminent because of the open-endedness of the application layer [63] -ese could includethe following

(i) -ere is poor or lack of a security design for anapplicationrsquos function Coming up with a gooddesign to address this could be a good ideaopportunity

(ii) Some of the programs might have backdoorsevading otherwise secure controls to allow un-authorized access to network resources We requiremechanisms to patch up such loopholes

(iii) Applications that lack or have weak authenticationmechanisms can be easily targeted by unlicensedusers to abuse the network Such mechanisms needto be strengthened further

(iv) Excessively complicated access control rules maybecause of their misunderstanding or poor writinggive unauthorized access allowing dubious activityrather than protecting the network they purport toprotect -ese should be simple and clear to un-derstand by all users

(v) In booting workstations without disks and managingnetwork devices we usually use the TFTP protocolalthough access does not require authenticating theuser by usernamepassword -is makes an intruderable to gain easy access to configaccess info as longas he can guess the filenames -is needs to be fixed

A few steps can be used to secure the application layerTo protect network data transmitted across the network

applications need to strengthen their encryption and au-thentication mechanisms Apps too need to enable strictcontrols of privileges for data access using good mechanismsto balance between usability and effectiveness [63] Appli-cations handling sensitive data need detailed logging andaudit capability as well as testing and reviewing

Table 3 summarizes the application layer protocols dis-cussed with their weaknesses and differences among them

4 Cross-Layer QoS Strategies

41 Service-Differentiated Real-Time Communication Scheme(SDRCS) -is is an event-driven routing protocol that routesreal-time traffic by featuring a cross-layer packet-forwardingdesign in which it integrates the real-time routing functionalitywith a new priority-based MAC scheme -e protocol ap-proximates distributed packet traversal speed for traffic clas-sification and admission control on the basis of this design Itfurther localizes decision-making by prioritizing packet for-warding to maximize the speed of packets [64]

Advantages

(i) It improves bandwidth utilization bandwidthdegradation due to unschedulable packets can beavoided

(ii) It offers a lightweight packet schedulability estima-tion mechanism Here it uses received signalstrength and admission control and then earlymissed-deadline packet-dropping policies are made

(iii) It delivers high end-to-end throughput via highersource data rates alongside strict end-to-end latencyneeds

(iv) Nodes which meet real-time requirements of lesstraffic load and better channel quality are givenhighest priority when it comes to packet forwarding

-e SDRCS operates via 5 mechanisms namely (1) RSS-(Rich Site SummaryReally Simple Syndication-) basedgrouping (2) admission control (3) prioritized queuing(4) real-time MAC and (5) dynamic forwarding -isprotocol has a queuing policy that employs per-hopdeadline-based queues and the nodes schedule theirpackets using FIFO priority-based queues [1] -e prin-cipal features of this protocol are mentioned as follows

(i) It is event driven and is easily adaptable to anynetwork changes

(ii) It does not need additional hardware for locali-zation or multichannel transmission

(iii) -e cross-layer packet-forwarding strategy can beused when transmitting multimedia traffic

(iv) It is able to efficiently circumvent any voids therein

Early deadline miss (EDM) policy drops all unscheduledpackets -e SDRCS circumvents wrong packet dropsdue to EDM since it adjusts to network changes easily

42 Sensor Fuzzy-Based Image Transport (SUIT) -is pro-tocol regulates congestion using a scheme based on fuzzy

24 Journal of Computer Networks and Communications

logic It transmits packets with reduced quality to allowablelevels if there is congestion enabling it to transmit lower-quality packetsframes without being dropped-e numberof packets delivered per second thus rises and while thisenhances video streaming apps it is more appropriatelymade for those that quickly transmit non-real-time videosuch as in video surveillance SUIT can stream JPEG pic-tures other than the traditional predictive video codingmethods that are not suitable for sensor devices because oftheir constraints of energy memory and computationalspeed Progressive JPEG-PJPEG performs better via packettransmission rate and delay though it degrades the imagequality because of congestion [65] -is protocol employscross-layer methods for interlayer data exchange though itstill employs some layering making it still possible topreserve all advantages of the layered design since it hasapplication and transport layers but has noMAC or routinglayer

For congestion detection and control SUIT bases on thepercentage number of in-bound and out-bound frames perwindow number of contenders and buffer occupancy of thenext-hop node Since buffer management has an effect oneffectiveness and QoS inWMSNs it is useful in the transportlayer so the protocol prioritizes packets for better QoS SUITreorders packets by tracking the sequence numbers of thedelivered packets since they all have a source ID framenumber and sequence number that it uses to detect anymissing packets -e protocol however does not providepacket reliability [65]

Advantages

(i) -e protocol prioritizes packets for better QoS(ii) It can transmit lower-quality packetsframes

without being dropped

Disadvantages

(i) It does not provide packet reliability

43 Network Layer QoS Support Enforced by a Cross-LayerController (NLQS) -is scheme permits packet-level servicedifferentiation as a function of throughput end-to-endpacket error rate and delay [66] (see Figure 8) -is en-hances network layer QoS It comprises a cross-layer controlunit (XLCU) for configuring and controlling networkingfunctionalities at physical MAC and network layers [18]-is depends upon the unified logic which affects choices forapplication layer needs and status of functional blocks whichdo implementation of networking functions [66]

Advantages

(i) Cross-layer interactions can be controlled withoutweakening the upgradability modularity andsimplicity of designing the system

44 Cross-Layer Signaling Shortcuts (CLASS) CLASS has ahigh level of effectiveness flexibility and comprehensive-ness According to [67] it has the following features

(i) Direct signaling among remote layers leading tofaster transmission

(ii) Lightweighted internal message format standard-ized protocols are not suitable for use in internalsignaling since they are not usually lightweight forinstance transmitting against faults in the networkCLASS requires only three fields namely destina-tion address event type and event contents

(iii) -e setup of its exterior messages is standardconsidering external signaling ICMP and TCPIPare suitable for general messages and short notifi-cations respectively

(iv) Message control protocol should ensure optimizedand organized exchange of dense simultaneousmessages across layers for high efficiency andavoidance of possible conflicts -e protocol appliesto many cross-layer signaling situations

Table 3 Differences and problems with application layer protocols

Protocol Architecture Transport QoS options Security Weaknesses

CoAP Requestresponse UDP DTLS No in-built security features DTLS does not supportmulticast

MQTT Publishsubscribe TCP TLSSSL Insufficient security at the protocol level sincepayload values are not encrypted

SMQTT Publishsubscribe TCP CPKP-ABE Key revocation and group pubsub for distributedSMQTT are still a challenge

MQTT-SN Publishsubscribe UDP TLSSSL SSLTLS suffers from attacks like BEAST CRIMERC4 and Heartbleed

XMPP RequestresponsePublishsubscribe TCP X TLSSSL Susceptible to spamming attacks and lacks worldwide

supportAMQP Publishsubscribe TCP TLSSSL Inappropriate for real-time applications

DDS Publishsubscribe TCPUDP TLSSSL Memory intensive and no open-source libraries forconstrained devices

REST Requestresponse HTTP X HTTPS Hard to implement Uses HTTP so incompatiblewith constrained apps

WebSockets PublishsubscribeClientserver TCP X TLSSSL Bad user experience for IoT apps since it runs the

clientserver architectureSTOMP Clientserver TCP HTTP -e broker can act as a bottleneck

Journal of Computer Networks and Communications 25

Advantages

(i) e protocol is scalable and has a very lowpropagation delay

(ii) It is very ecient and exible since it uses directsignaling between any two layers

(iii) It is very comprehensive and eective

45 Cross-Layer and Multipath-Based Video Transmission(CMVT) CMVT is a hybrid of application and networklayers At the application layer it carries out the encoding ofvideo streams to video data frames (I- P- and B-frames) bymeans of the MPEG-4 encoding format At the networklayer route discovery and data transfer take place Twoalgorithms greedy forwarding and rollback are used todiscover a number of paths from the source to the sink nodevia the route discovery method [20] A specied node icalculates the evaluation of its adjacent node j using thefollowing equation

fij (1 minus α)d2(j D) minus d2min(i)d2max(i) minus d

2min(i)

+ αeinit(j) minus eres(j)

einit(j)

(19)

where fij evaluation value for the node i to j d2(j D) isthe distance from the node j to the destination node D

d2min(i) and d2max(i) are the minimum and maximum dis-

tances of neighbors of the node i to D respectively einit(j) isthe initial energy of the node j eres(j) is the current residualenergy of the node j and α is the energy coecient given asfollows

α emax(i) minus emin(i)

emax(i) (20)

where emax(i) and emin(i) are respectively the maximumand minimum energy left of all node irsquos neighbors

e network layer further transmits video streams withCMVT doing status evaluation choosing an appropriatecommunication path for any packets and the QoS guaranteelevel for the path i is calculated as

fi (1 minus ω)hisum hi

+ ωnisum ni

(21)

where fi is the evaluation value for the path i hi are the hopsfor the path i ni is the summation of packets sent via isum ni isthe summation of packets sent by sources and ω is theenergy consumption factor [18]

Advantages

(i) e CMVT protocol is very superior in mediatransmission particularly large-scale WMSNs

To network peers

Flow requirements

-

Atimes

Admissioncontrol

Capacityassignment

Next-hopselection

Coding rate

Data rate

QoScontracts

Cross-layer controlunit (XLCU)

Localization subsystem

Synchronizationsubsystem

Networkcoordinationunit (NCU)

UWB modulator

Channel coder

QoS scheduler

Georouting

Source coder

Rate control

Figure 8 Cross-layer controller architecture

26 Journal of Computer Networks and Communications

46 Cross-Layer Cooperative MAC (CoopMAC)CoopMAC [68] is based on the IEEE 80211 DCF approach-ere is a reduction in interference of neighboring cellsleading to uniform coverage in densely deployed networks[68] A connection is established by two-way authenticationbetween the sender and the receiver When a Clear-to-Send(CTS) packet and short interframe space (SIFS) are obtainedby the sender packets are transmitted right to the receiverwith no valuable cooperation If available the two nodes findout whether they can exchange data -is is done by the useof a helper identification signal to establish how feasiblecooperative communication is It is established in thepresence of a signal and if it is not available direct com-munication is activated For distributed systems helper-initiated cooperation is preferable because of the presence ofthe RTSCTS packets [69]

Advantages

(i) CoopMAC has an advantage over others due to thespatial diversity amongst the three nodes [70]

(ii) It has the potential to achieve significantthroughput and delay with no complexity in thedesign of the system

47 Cooperative MAC Protocol for Multihop Networks (M-CMAC) Just as CoopMAC M-CMAC is made in such away that high data rate stations help low data rate stations toforward traffic for broadcasting [71] -ere are also helperschosen in such a way that 2 fast-hop transmissions replace aslow hop transmission -e helper via whom there existsleast delay from the sender to the receiver is considered bestbecause of theminimum two-hop transmission rate and so isthe chosen neighbor to the duo Here the working as-sumption is that all nodes have their position coordinatesknown and thus the Euclidean distances between all pairscan be calculated and converted to the data rate for that linkAll nodes have cooperative tables (CT) of potential helperswith destination and helper MAC address Euclidean dis-tance and total distance through the helper Any sender withdata to transmit checks out for the existence of any helper inthe CT for the destination and if available forwards RTS forchannel reservation for single-hop duration M-CMAC hasgot an RTS format having a shape with five fields similar tothe one given as follows

Framecontrol Duration Source

addressDestination

addressHelperaddress

-e destination nodersquos address is kept by the source in thepart for the helper address that is in turn saved as the desti-nation address Nodes inspect helper and destination addressfields on receipt of RTS whereby the node acts as the othernodersquos helper in case the helper address field is not the sameand the helper transmits CTS back to the source if at all it wantsto forward data On delivery of the CTS packet the sourcetransmits to the helper which in turn sends the packets to thedestination that also sends ACK to the helper on receipt of thepacket In M-CMAC there exists a higher level of channelreuse (parallel transmissions) due to the increased number of

nodes therein that subsequently raises availability of helpers fordata forwarding -is results in increased throughput incomparison with CoopMAC and IEEE 80211 DCF [71]

Advantages

(i) It promises a higher throughput compared toCoopMAC and IEEE 80211 DCF

48 Cluster-Based Cooperative Routing (CBCR) Protocol-e CBCR protocol has a multihop data-forwarding func-tion realized at the link layer with cooperative links that useM-CMAC -is protocol encompasses two stages namelythe routing relay selection phase and data forwarding phase

(i) Routing relay selection phase all nodes announcetheir presence to their neighbors by broadcastingperiodical beacon messages that carry the MAC ad-dress of the node Each of these constructs a relaytable that contains all neighboring nodes with whichit is able to communicate -e node further broad-casts its neighbor list in case of any change to itsentries since the previous broadcast -e MAC ad-dresses of nodes next to the node X are contained inits relay tablersquos column one and the neighbor nodersquosrow has MAC addresses for neighbors of the adjacentnode Based on its relay table each node in-dependently chooses routing relays Choosing a nodeas a relay node depends on the number of nodes itconnectsmdashit should be the highest number [71]

(ii) Data forwarding phase a node with packets totransmit needs to first verify if the receiver is in thesame cluster and if that receiver has a helper to whichthe packet is thence forwarded but in whose absencethe packet is delivered right to the receiver In casethe intended receiver is in a different cluster therelay table will be checked to see whether that re-ceiver is reachable via other routing relays If it isreachable packets will be sent to the routing relaydirectly or via the helper if at all the relay has one Incase of destination unreachable via relays packetsare multicast to all relays by the node [71]Advantages

(i) It enables multicasting

49 MAC-PHY Cross-Layer Protocol A cooperative cross-layer standard for cooperation at the physical layer in next-generation WMSNs is developed in [70] It provides acomplete MAC layer algorithm that gives a supportive shellfor the PHY-MAC layer Like CoopMAC its scheme de-pends on the middle node for communication betweennodes-eMAC layer protocol has been changed to regulateinformation exchanges at the physical layer -e receivernode receives duplicate packets from source and helpernodes to decode the data [70]

410 MAC-Centric Approach -is cross-layer protocoltargets multimedia applications by using the MPEG-4

Journal of Computer Networks and Communications 27

scheme [21] It is characterized by 4 access categories(AC3ndashAC0) according to their priority in transmission It ismeant for supporting different QoS needs in upcomingvideo applications and enabling MAC layer differentiationfor H264 partitioning [18] Because of low bandwidth delayand other QoS challenges that cause inefficiency in trans-mittingmultimedia data inWMSNs a number of algorithmsdepending on IEEE 80211e have been proposed to supporttransmission of quality videos [72] Selecting an AC dependson the measures for QoS such as the loss rate and delay-us AC3 having the highest priority is mapped to theparameter set concept since the stream is sensitive to bit lossespecially in video transmission [21]

411 Adaptive Cross-Layer Forward Error Correction(ACFEC) In the ACFEC [73] model data packets are ex-changed among nodes through the access point (AP) whichoperates the infrastructure mode in which it adds FEC to videodata [72] -ese data are dealt by encapsulation using astreaming server to the receiver through thewireless AP as RTPpackets An adaptive FEC controller senses the category ofpackets out of the RTP header and recovers the header ofpackets from UDP -e encoder generates some error-cor-recting packets whose number is determined by the sourcepacket number of the block Multimedia transmissions aremonitored by the controller using MAC failure data and itscounter is increased by one if there is a failed transmission-econtroller uses the failure counter to change the packet numberproduced after transmitting a block [18] In case there are lostpackets it changes the redundancy rates and produces extrapackets to replace the lost ones and satisfy the requirements ofthe receiving node [72] -e FEC packet number is enhancedor reduced to satisfy the receiverrsquos requirements and stoppacket losses-is is done through accurate detection of packetlosses and adjustment of redundancy rates If all video datapackets are well received there will not be generation of FECpackets [73]

Advantages

(i) It promises a good QoS through packet loss re-duction and redundancy rate adjustment

412 Balanced Cross-Layer Fuzzy Logic (BCFL) DesignRouting Algorithm in WSNs A new fuzzy logic-basedrouting algorithm (BCFL) [74] was designed using disper-sion of the cross-layer parameter as the fuzzy logic inferencesystem input Each cross-layer parameter has a dynamicweight depending on the value of dispersion -e designcomes with some innovations as follows (i) for fuzzy logicinference system input the absolute parameter value issubstituted by parameter dispersion thereby significantlyreducing algorithmic complexity (ii) dispersion does notchange with the order of magnitude according to the dis-persion formula and (iii) the weight of the parameter de-pends on the size of its dispersion and the two are inverselyproportional to each other -is enables BCFL to have somenobility unlike other algorithms as per the following dis-tinguishing properties making it advantageous

(i) It has got simple if-then rules which remain con-stant even when the constraints increase

(ii) It is capable of dealing with many constraints withno increment in complexity

(iii) It has capacity to yield a more balanced solutionthan other algorithms

(iv) It is easily adaptable to changes in network con-ditions and topology even when the changes arefrequent like in underwater WSNs -e algorithm isuseable in choosing the CH in cluster-based routingprotocols [74]

413 Minimum Hop Disjoint Multipath Routing Algorithmwith Time Slice Load-Balancing Congestion Control SchemeMHDMwTS is a two-phased routing protocol comprisingpath build-up and path acknowledgment phases It consists ofmultiple sources each of which has three build-up disjointpaths namely primary alternate and back-up paths Onactivation the source node requests to build up a route to thenearest hop neighbor which is the path build-up phase [19] Inthis phase step one has source activation in which therequested node for path building adds its number and time-stamp and sends to the least hop-count neighboring node-isgoes on until the least time latency sink having the informationneeded to construct the primary route is reached [75] In steptwo there is path extraction when the new package fromanother path arrives whereby the extracted path is comparedwith the primary path If the node is shared the package will berejected or an alternative path will be searched for to get abackup through comparison of the preceding two paths Phasetwo is path acknowledgment where we have the third step inwhich the sink returns the ACK packet to the sender with pathinformation having nodes and their time information after ithas computed it using the timestamp [1819]

Advantages

(i) -e protocol minimizes end-to-end latency(ii) It enhances congestion control

414 Cross-Layer Optimal Design (CLOD) Authors in [76]come up with a CLOD for scheduling at the data link layerrouting at the network layer and controlling congestion atthe transport layer with an assumption of fixed link capacity-rough congestion control energy efficiency is improvedTransport layer congestion at the nodes is minimized bycompressed sensing (CS) in which transmitted bits are re-duced whereas optimally allocating resources decreasescongestion on the links at the data link layer CLODpromises minimized computational complexity and betterperformance in light traffic scenarios [77] Generating andstoring CLOD is more efficient compared to Gaussianrandom matrices It prolongs network lifetime and savesenergy

Advantages

(i) It does congestion control subsequently increasingenergy efficiency

28 Journal of Computer Networks and Communications

(ii) It reduces computational complexity(iii) It improves performance in light traffic scenarios

415 Challenges and Opportunities of the Cross-Layer Design(CLD)

(i) Different layers in a real network assume differentfunctions andor services A layer only communi-cates with its neighboring layers -e layered modeldegrades system performance because of many

features in wireless transmission QoS parameters inthe layers of WMSNs are shared by routing pro-tocols to optimize the performance Neverthelessthere is a need to design cross-layer models forincreased efficiency in routing [1]

(ii) -e physical layer plays a highly vital role in CLDRate adaptation and channel allocation take place atthe physical layer via signal processing to enhanceQoS End-to-end performance in wireless media isaffected by the changes therein given the effect on

Table 4 A summary of some of the reviewed cross-layer models

Protocol Layers Aims Comments QoS parameters

SDRCS MACPHY Routing real-time traffic Transmits multimedia traffic viacross-layer packet forwarding

-roughputlatency

SUIT Applicationtransport Transmission of non-real-time video Uses fuzzy logic to regulate

congestionTransmission rate

delay

CLASS Any twoDesign serves as a framework fordifferent implementations ofdifferent application scenarios

Scalable very efficient flexible andhas very low propagation delay

Propagation delayjitter

NLQS PHYMACnetwork

Permits packet-level servicedifferentiation as a function of

throughput packet error rate anddelay

Network layer QoS is enhanced -roughput delaypacket error rate

CMVT Applicationnetwork

Encodes video streams to video dataframes via MPEG-4 encoding anddoes route discovery and data

transfer

Greedy forwarding and rollback areused to find source-to-sink paths Energy

CoopMAC PHYMAC Offers spatial diversity among thethree nodes

Helper ID signal finds out howfeasible cooperative communication

is

-roughput ratedelay

M-CMAC PHYMAC Increases end-to-end throughputand packet delivery ratio

Euclidean distances between nodesare calculated and converted to the

data rate for a link

-roughputpacket delivery

ratio

CBCR PHYMACMinimizes control overhead andtime consumed in establishing thecooperative paths than M-CMAC

Energy consumption is moreuniformly distributed in a network

enhancing network lifetime

-roughputpacket deliveryratio energy

MAC-PHY MACPHYMeant to enable PHY layer

cooperation and maximize gains ofcooperation at the MAC layer

Leverages both spatial diversity andcoding gain -roughput delay

MAC-centric MACAPPL

Meant to support QoS needs in newvideo apps and enable MAC layer

differentiation for H264partitioning

Targets multimedia applications byusing the MPEG-4 scheme

Delay packet lossrate

ACFEC MACnetwork(UDP)

Meant to enhance the quality ofvideo streaming over 80211WLANs

and overcome packet losses

Adjusts redundancy rates toovercome channel fluctuations and

detect and reduce packet lossPacket loss rate

MHDMwTS mdash

Meant to provide reliable datatransfer with the multipath routingand load-balancing congestioncontrol method in WMSNs

More reliable than basic routingschemes for transport multimedia

data

Latency packagetransmit rate

BCFL mdash

Introduces dispersion into fuzzylogic-based routing and sets every

cross-layer parameter with adynamic weight

Can be used to select a CH in cluster-based routing protocols andproposes a dispersion formula

Node utilitydispersion

CLODDatalinknetworktransport

Prolongs network lifetime andachieves congestion control anddesigned for lightly loaded WSNs

Assumes fixed link capacity andintegrates compressed sensing

technology

-roughputaverage energy CS

error ratio

Journal of Computer Networks and Communications 29

the purposefulness of the protocols at the networklayer CLD offers solutions to conserving powerminimizing energy and controlling both the flowand congestion in the network as well as fault tol-erance so it is an opportunity for designers con-sidering other layers We further desire to developCLDs that shall make an impact on network oper-ation and get close attention [18]

(iii) CLD interface standardization the architecturemust deliver module functionality although thereexist some queries amongst modules about possibleinterfaces predicted by the necessity to share in-formation among remote protocol layers Technicalchallenges include developing designing andstandardizing of cross-layer interfaces and algo-rithms that satisfy the cross-layer optimization re-quirements among protocol layers

Table 4 summarizes these cross-layer models with a fewcomments

5 Future Research Direction

51 At theMACLayer In WMSNs achieving duty cycling istricky since video traffic is volatile Listening to adjacentnodes wastes energy We need to switch states from awake tosleep -is requires further research

Self-organizing networks and green communicationalthough most CRWSN energy-saving algorithms aredesigned to reduce node transmission power energyconsumption is inevitable during the operational mode ofBSs because of internal processing Scholars are trying withpolicy-makers to encourage a shift to greenWSNs to reduceoperational costs and carbon footprints With self-orga-nizing CRWSNs for automatic switch-off when idle energyis saved which calls for cognitive switching algorithms Sointegrating cognitive switching with radio resource allo-cation in CRWSNs is a promising research area [78]

52 At the Application Layer Considering a WMSN ar-chitecture end users access the network via the BS to andfrom which they send their sensed data We desire to getguaranteed QoS and energy efficiency by designing a newarchitecture in the form of hardware and software whichneeds to be investigated further for a fix of the same [1]

QoS and energy efficiency are very crucial in WMSNsand mostly real-time apps that require guaranteed band-width and throughput in their network lifetime Mostprotocols ignore base station (BS) mobility and WMSNnodes Traffic management telemedicine and battlefieldsurveillance apps require mobile nodes or BSs so designingdynamic routing protocols to be adaptable in these cir-cumstances is necessary

53 Cross-Layer Security of data transmitted on WMSNseg in military surveillance and e-commerce is key withQoS and energy efficiency WMSNs are vulnerable to attacks

like worms and sinkhole Moreover their computationpower is not that high complicating implementation ofstrong secure protocols -is needs further attention

For fairness and priority issues in CRWSNs nodes areassigned diverse priorities depending on importance andurgency of their data For better network performance thereis a need to develop fair resource allocation mechanismswith priority-based fairness in sensors-ere is still potentialin this research area [78]

We must develop multiobjective adaptive protocols tooptimize QoS metrics involved in routing for best trans-mission results to get a trade-off between different opti-mization metrics We need to design standardized cross-layer algorithms which are able to meet cross-layer opti-mization standards

6 Conclusion

We have reviewed the different QoS strategies for WSNs inthe context of IoT from the MAC layer and application layeras well as the cross-layer paradigm For the MAC layer wehave reviewed protocols for WSNs such as the hybrid of thecontention-free and contention-based MAC protocols andthose for WMSNs We further reviewed a number of ap-plication layer protocols including many machine-to-ma-chine requestresponse and publishsubscribe protocols Forsystem optimization cross-layer QoS strategies are veryimportant in wireless communication We have reviewed anumber of cross-layer strategies For all categories reviewedchallenges and opportunities are discussed Finally somepossible future directions are discussed for research andapplication showing a promising potential for future re-search in this relatively new area especially as more WMSNapplications emerge recently

Conflicts of Interest

-e authors hereby declare no conflicts of interest regardingthe publication of this work

Acknowledgments

-is research was funded by the National Natural ScienceFoundation of China (Nos 61602398 61672447 and61711540306)

References

[1] V Bhandary A Malik and S Kumar ldquoRouting in wirelessmultimedia sensor networks a survey of existing protocolsand open research issuesrdquo Journal of Engineering vol 2016Article ID 9608757 27 pages 2016

[2] I F Akyildiz T Melodia and K R Chowdhury ldquoWirelessmultimedia sensor networks applications and testbedsrdquoProceedings of the IEEE vol 96 no 10 pp 1588ndash1605 2008

[3] M Iqbal M Naeem A Ahmed M Awais A Anpalagan andA Ahmad ldquoSwarm intelligence based resource managementfor cooperative cognitive radio network in smart hospitalsrdquoWireless Personal Communications vol 98 no 1 pp 571ndash5922018

30 Journal of Computer Networks and Communications

[4] S A Alvi B Afzal G A Shah L Atzori and W MahmoodldquoInternet of multimedia things vision and challengesrdquo AdHoc Networks vol 33 pp 87ndash111 2015

[5] Q F Hassan A U R Khan and S A Madani Internet ofings Challenges Advances and Applications CRC PressTaylor amp Francis Group A Chapman amp Hall Group BocaRaton FL USA 2018

[6] Q Wang Y Zhao W Wang et al ldquoMultimedia IoT systemsand applicationsrdquo in Proceedings of the 2017 Global Internet ofings Summit (GIoTS) vol 2 Geneva Switzerland June 2017

[7] A Aslam and E Curry ldquoTowards a generalized approach fordeep neural network based event processing for the internet ofmultimedia thingsrdquo IEEE Access vol 6 pp 25573ndash25587 2018

[8] M Jridi T Chapel V Dorez L Bougeant and A Le BotlanldquoSoC-based edge computing gateway in the context of theinternet of multimedia things experimental platformrdquoJournal of Low Power Electronics and Applications vol 8no 1 p 1 2018

[9] H Noura A Chehab L Sleem andM N Rapha ldquoOne roundcipher algorithm for multimedia IoT devicesrdquo MultimediaTools and Applications vol 77 no 14 pp 18383ndash18413 2018

[10] T Balan D Robu and F Sandu ldquoMultihoming for mobileinternet of multimedia thingsrdquo Mobile Information Systemsvol 2017 Article ID 6965028 16 pages 2017

[11] H Shen and G Bai ldquoRouting in wireless multimedia sensornetworks a survey and challenges aheadrdquo Journal of Networkand Computer Applications vol 71 pp 30ndash49 2016

[12] A Ahmad M H Rehmani H Tembine O A Mohammedand A Jamalipour ldquoIEEE access special section Editorial optimization for emerging wireless Networks IoT 5G andsmart grid communication networksrdquo IEEE Access vol 5pp 2096ndash2100 2017

[13] B A Muzakkari M A Mohamed M F A Kadir andZ Mohamad ldquoRecent advances in energy efficient-QoS awareMAC protocols for wireless sensor networksrdquo InternationalJournal of Advanced Computer Research vol 8 no 38pp 212ndash228 2018

[14] M A Yigitel O D Incel and C Ersoy ldquoQoS-aware MACprotocols for wireless sensor networks a surveyrdquo ComputerNetworks vol 55 no 8 pp 1982ndash2004 2011

[15] A M Abbas and O Kure ldquoQuality of Service in mobile ad hocnetworks a surveyrdquo International Journal of Ad Hoc andUbiquitous Computing vol 6 no 2 p 75 2010

[16] M Q Shatnawi ldquoApplication layer protocols for the internetof things a surveyrdquo in Proceedings of the InternationalConference on Engineering amp MIS (ICEMIS) Agadir Mo-rocco September 2016

[17] V Karagiannis P Chatzimisios F Vazquez-gallego andJ Alonso-zarate ldquoA survey on application layer protocols forthe internet of things research motivationrdquo InternationalJournal of Advanced Research in Computer Science vol 8no 3 pp 1ndash10 2015

[18] M S Bernard T Pei Z Li and K Li ldquoQoS strategies forwireless multimedia sensor networks in the context of IoTrdquoLecture Notes of the Institute for Computer Sciences SocialInformatics and Telecommunications Engineering vol 4pp 228ndash253 2019

[19] A AlAmri and M Abdullah ldquoCross-layer quality of serviceprotocols for wireless multimedia sensor networksrdquo in Pro-ceedings of the International Conference on CommunicationManagement and Information Technology pp 649ndash658Cosenza Italy April 2017

[20] J Guo L Sun and R Wang ldquoA cross-layer and multipathbased video transmission scheme for wireless multimedia

sensor networksrdquo Journal of Networks vol 7 no 9pp 1334ndash1340 2012

[21] A Ksentini A Naimi and M Gueroui ldquoToward an im-provement of H264 video transmission over IEEE 80211ethrough a cross-layer architecturerdquo IEEE CommunicationsMagazine vol 44 no 1 pp 107ndash114 2006

[22] W Ye J Heidemann and D Estrin ldquoAn energy-efficientMAC protocol for wireless sensor networksrdquo in Proceedings ofthe 21st International Annual Joint Conference of the IEEEComputer and Communications Societies (INFOCOM 2002)New York NY USA June 2002

[23] J Polastre J Hill and D Culler ldquoVersatile low power mediaaccess for wireless sensor networks categories and subjectdescriptorsrdquo in Proceedings of the Second InternationalConference on Embedded Networked Sensor Systems Sen-Sysrsquo04 pp 95ndash107 Baltimore MA USA November 2004

[24] T Danmanee K N Nakorn and K Rojviboonchai ldquoCU-MAC a duty-cycle MAC protocol for internet of things inwireless sensor networksrdquo ECTI Transactions on ElectricalEngineering Electronics and Communications vol 16 no 2pp 30ndash43 2018

[25] T Van Dam and K Langendoen ldquoAn adaptive energy-efficientMAC protocol for wireless sensor networksrdquo in Proceedings ofthe 1st International Conference on Embedded NetworkedSensor Systems Los Angeles CA USA November 2003

[26] P Xie and J Cui ldquoR-MAC an energy-efficient MAC protocolfor underwater sensor networksrdquo in International Conferenceon Wireless Algorithms Systems and Applications (WASA2007) Chicago IL USA August 2007

[27] A Rahim N Javaid M Aslam U Qasim and Z A KhanldquoAdaptive-reliable medium access control protocol forwireless body area networksrdquo in Proceedings of the AnnualIEEE Communications Society Conference on Sensor Meshand Ad Hoc Communications and Networks (SECON) SeoulKorea June 2012

[28] H S Kwak S Ullah D H Kwak and C H Lee ldquoA powerefficient MAC protocol for wireless body area networksrdquoJournal of the Korean Institute of Intelligent Systems vol 8no 6 pp 131ndash140 2009

[29] C Omeni A J Burdett and M Park ldquoEnergy efficientmedium access protocol for wireless medical body area sensornetworksrdquo in Proceedings of the 4th IEEEEMBS InternationalSummer School and Symposium on Medical Devices andBiosensors pp 29ndash32 Cambridge UK August 2007

[30] L Sitanayah C J Sreenan and K N Brown ldquoER-MAC ahybrid MAC protocol for emergency response wireless sensornetworksrdquo in Proceedings of the Fourth International Con-ference on Sensor Technologies and Applications pp 244ndash249Mestre Italy July 2010

[31] Y Sun and D B Johnson ldquoRI-MAC a receiver-initiatedasynchronous duty cycle MAC protocol for dynamic trafficloads in wireless sensor networksrdquo in Proceedings of the 6thACM conference on Embedded network sensorsystemsmdashSenSysrsquo08 pp 1ndash14 Raleigh NC USA November2008

[32] M Buettner G V Yee E Anderson and R Han ldquoX-MAC ashort preamble MAC protocol for duty-cycled wireless sensornetworksrdquo in Proceedings of the 4th International Conferenceon Embedded Networked Sensor Systems SenSys 2006pp 307ndash320 Boulder Colorado USA October 2006

[33] S Boulfekhar and M Benmohammed ldquoSynchronous receiverinitiated MAC protocol for long-lived sensor networksrdquoComputers amp Electrical Engineering vol 40 no 2 pp 504ndash516 2014

Journal of Computer Networks and Communications 31

[34] S Mehta and K S Kwak ldquoH-MAC a hybrid Mac protocol forwireless sensor networksrdquo International Journal of ComputerNetworks amp Communications vol 2 no 2 pp 108ndash117 2010

[35] I Rhee A Warrier M Aia J Min and M L Sichitiu ldquoZ-MAC a hybrid MAC for wireless sensor networksrdquo IEEEACM Transactions on Networking vol 16 no 3 pp 511ndash5242008

[36] S Arshad A Al-sadi and A Barnawi ldquoZ-MAC performanceevaluation and enhancementsrdquo Procedia Computer Sciencevol 21 pp 485ndash490 2013

[37] Y Liu I Elhanany and H Qi ldquoAn energy-efficient QoS-aware media access control protocol for wireless sensornetworksrdquo in Proceedings of the IEEE International Confer-ence on Mobile ad hoc and Sensor Systems pp 9ndash11 Wash-ington DC USA December 2005

[38] J Decotignie ldquoWiseMAC an ultra low power MAC protocolfor the downlink of infrastructure wireless sensor networksrdquoin Proceedings of the Ninth International Symposium onComputers And Communications pp 244ndash251 AlexandriaEgypt June 2004

[39] R A Rashid ldquoDevelopment of energy aware TDMA-basedMAC protocol for wireless sensor network systemrdquo EuropeanJournal of Scientific Research vol 30 no 4 pp 571ndash578 2009

[40] H P Van Hoesel ldquoA lightweight medium access protocol(LMAC) for wireless sensor networksrdquo in Proceedings of theInternational Workshop on Networked Sensing Systems Bal-timore MD USA November 2004

[41] C Li P Wang H Chen andM Guizani ldquoA cluster based on-demand multi-channel MAC protocol for wireless multi-media sensor networksrdquo in Proceedings of the IEEE In-ternational Conference on Communications pp 2371ndash2376Beijing China May 2008

[42] J L Sobrinho and A S Krishnakumar ldquoQuality-of-service inad hoc carrier sense multiple access wireless networksrdquo IEEEJournal on Selected Areas in Communications vol 17 no 8pp 1353ndash1368 1999

[43] J P John ldquoProviding MAC QoS for multimedia traffic in80211e based multi-hop ad hoc wireless networksrdquo ComputerNetworks vol 51 no 1 pp 153ndash176 2007

[44] S Lofty and N Padmavati ldquoA survey on mobility basedprotocols in WSNsrdquo in Proceedings of the InternationalConference Communication and Manufacturing pp 12ndash15Hangzhou China July 2014

[45] S Deng L Shen and J Li ldquoMobility-based clustering pro-tocol for wireless sensor networks with mobile nodesrdquo IETWireless Sensor Systems vol 1 no 1 pp 39ndash47 2011

[46] M Khalil A Khalid F U Khan and A Shabbir ldquoA review ofrouting protocol selection for wireless sensor networks insmart citiesrdquo in Proceedings of the 24th Asia-Pacific Confer-ence on Communications (APCC) Ningbo China November2014

[47] O Message Q Telemetry and T Mqtt Committee Specifi-cation Draft 02 April 2014

[48] M Collina M Bartolucci A Vanelli-coralli andG E Corazza ldquoInternet of things application layer protocolanalysis over error and delay prone linksrdquo in Proceedings ofthe 7th Advanced Satellite Multimedia Systems Conference andthe 13th Signal Processing for Space CommunicationsWorkshop Livorno Italy September 2014

[49] A Talaminos-barroso M A Estudillo-valderrama L M RoaJ Reina-tosina and F Ortega-ruiz ldquoA machine-to-machineprotocol benchmark for e health applicationsmdashuse case re-spiratory rehabilitationrdquo Computer Methods and Programs inBiomedicine vol 129 pp 1ndash11 2016

[50] V Goyal O Pandey A Sahai and B Waters ldquoAttribute-based encryption for fine-grained access control of encrypteddatardquo in Proceedings of the 13th ACM Conference on Com-puter and Communications Security pp 89ndash98 AlexandriaVA USA October 2006

[51] M Singh M A Rajan V L Shivraj and P BalamuralidharldquoSecure MQTT for internet of things (IoT)rdquo in Proceedings ofthe Fifth International Conference on Communication Systemsand Network Technologies Gwalior India April 2015

[52] A Stanford-Clark and H L Truong MQTT for Sensor Net-works (MQTT-SN) Protocol Specification httpmqttorgnewwp-contentuploads200906MQTT-SN_spec_v12pdf 2013

[53] RFC ldquoExtensible messaging and presence protocol (XMPP)rdquoin Proceedings of the IETF pp 1ndash9 San Diego CA USAAugust 2005

[54] S Vinoski and I Technologies ldquoQueuing protocolrdquo IEEEInternet Computing vol 10 no 6 pp 87ndash89 2006

[55] J L Fernandes I C Lopes J J P C Rodrigues and S UllahldquoPerformance evaluation of restful web services and AMQPprotocolrdquo in Proceedings of the Fifth International Conferenceon Ubiquitous and Future Networks (ICUFN) pp 2ndash7 DaNang Vietnam July 2013

[56] M Asim ldquoA survey on application layer protocols for internetof things (IoT)rdquo International Journal of Advanced Researchin Computer Science vol 8 no 3 pp 996ndash1000 2017

[57] K An A Gokhale D Schmidt S Tambe P Pazandak andG Pardo-castellote ldquoContent-based filtering discovery pro-tocol (CFDP) scalable and efficient OMG DDS discoveryprotocolrdquo in Proceedings of the 8th ACM InternationalConference on Distributed Event-Based Systems New YorkNY USA July 2014

[58] E Web What is a web service pp 52ndash57 2010[59] M Laine RESTful Web Services for the Internet of ings

pp 2ndash4 2012 httpmediaTkk[60] IETF -e WebSocket protocol pp 1ndash71 2011[61] V Pimentel ldquoCommunicating and displaying real-time data

with websocketrdquo IEEE Internet Computing vol 16 no 4pp 45ndash53 2012

[62] Wikipedia 2019 httpsenwikipediaorgwikiStreaming_Text_Oriented_Messaging_Protocol2019

[63] K Raghavendra and S Nireshwalya ldquoApplication layer se-curity issues and its solutionsrdquo IJCSET vol 2 no 6pp 1266ndash1269 2012

[64] Y Xue B Ramamurthy andM C Vuran ldquoSDRCS a service-differentiated real-time communication scheme for eventsensing in wireless sensor networksrdquo Computer Networksvol 55 no 15 pp 3287ndash3302 2011

[65] C Sonmez S Isik M Y Donmez O D Incel and C ErsoyldquoSUIT a cross layer image transport protocol with fuzzy logicbased congestion control for wireless multimedia sensornetworksrdquo in Proceedings of the 5th International Conferenceon New Technologies Mobility and Security (NTMS) pp 1ndash6IEEE Istanbul Turkey May 2012

[66] T Melodia and I Akyildiz ldquoCross-layer QoS-aware com-munication for ultra wide band wireless multimedia sensornetworksrdquo IEEE Journal on Selected Areas in Communica-tions vol 28 no 5 pp 653ndash663 2010

[67] Q Wang and M A Abu-Rgheff ldquoCross-layer signalling fornext-generation wireless systemsrdquo in Proceedings of the 2003IEEE Wireless Communications and Networking 2003WCNC 2003 vol 2 pp 1084ndash1089 IEEE New Orleans LAUSA March 2003

[68] P Liu Z Tao S Narayanan T Korakis and S PanwarldquoCoopMAC a cooperative MAC for wireless LANsrdquo IEEE

32 Journal of Computer Networks and Communications

Journal on Selected Areas in Communications vol 25 no 2pp 340ndash354 2007

[69] H Shan H T Cheng and W Zhuang ldquoCross-layer co-operative MAC protocol in distributed wireless networksrdquoIEEE Transactions onWireless Communications vol 10 no 8pp 2603ndash2615 2011

[70] F Liu T Korakis Z Tao and S Panwar ldquoAMAC-PHY cross-layer protocol for wireless Ad-Hoc networksrdquo in Proceedingsof the 2008 IEEE Wireless Communications and NetworkingConference pp 1792ndash1797 IEEE Las Vegas NV USA March2008

[71] L Jacob and H R Shamna ldquoEfficient cooperative MAC androuting in wireless networksrdquo Transactions on Networks andCommunications vol 3 no 5 2015

[72] S Rao and K Shama ldquoCross layer protocols for multimediatransmission in wireless networksrdquo International Journal ofComputer Science amp Engineering Survey vol 3 no 3pp 15ndash28 2012

[73] L Han S Park S Kang and H Peter ldquoAn adaptive cross-layer FEC mechanism for video transmission over 80211WLANsrdquo in Proceedings of the 2009 International Conferenceon Internet pp 209ndash215 Las Vegas NV USA December2009

[74] N Li J Martınez and V H Dıaz ldquo-e balanced cross-layerdesign routing algorithm in wireless sensor networks usingfuzzy logicrdquo Sensors vol 15 no 8 pp 19541ndash19559 2015

[75] G Sun J Qi Z Zang and Q Xu ldquoA reliable multipathrouting algorithm with related congestion control scheme inwireless multimedia sensor networksrdquo in Proceedings of the2011 3rd International Conference on Computer Research andDevelopment vol 4 pp 229ndash233 IEEE Shanghai ChinaMarch 2011

[76] M Li Y Jing and C Li ldquoA robust and efficient cross-layeroptimal design in wireless sensor networksrdquoWireless PersonalCommunications vol 72 no 4 pp 1889ndash1902 2013

[77] J Yan M Zhou and Z Ding ldquoRecent advances in energy-efficient routing protocols for wireless sensor networks areviewrdquo IEEE Access vol 4 pp 5673ndash5686 2016

[78] A Ahmad S Ahmad M H Rehmani and N U Hassan ldquoAsurvey on radio resource allocation in cognitive radio sensornetworksrdquo IEEE Communications Surveys amp Tutorials vol 17no 2 pp 888ndash917 2015

Journal of Computer Networks and Communications 33

International Journal of

AerospaceEngineeringHindawiwwwhindawicom Volume 2018

RoboticsJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Active and Passive Electronic Components

VLSI Design

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Shock and Vibration

Hindawiwwwhindawicom Volume 2018

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawiwwwhindawicom

Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Control Scienceand Engineering

Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Journal ofEngineeringVolume 2018

SensorsJournal of

Hindawiwwwhindawicom Volume 2018

International Journal of

RotatingMachinery

Hindawiwwwhindawicom Volume 2018

Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Navigation and Observation

International Journal of

Hindawi

wwwhindawicom Volume 2018

Advances in

Multimedia

Submit your manuscripts atwwwhindawicom

Page 16: ReviewArticle - downloads.hindawi.com · ReviewArticle QoSStrategiesforWirelessMultimediaSensorNetworksinthe ContextofIoTattheMACLayer,ApplicationLayer,and Cross-LayerAlgorithms MuwongeSsajjabbiBernard

(iv) Implementation of real-time communication withvarying needs of bandwidth good for multimediatraffic

Amechanism for dynamically assigning packet prioritiesbased on their deadlines and the traversed hops is proposedin [43] -e ReAllocative Priority (ReAP) mechanism ismeant to provide QoS at the MAC layer to video traffic -isscheme introduces adaptive TXOP by modifying the TXOPdynamically based on the queue length [15] Other protocolsin this category include dynamic duty cycle and adaptivecontention window-based MAC protocol frame sharing(FRASH) MAC protocol and real-time independentchannels (RICH) MAC protocol

Advantages

(i) It can implement real-time communication withvarying needs of bandwidth

23 Challenges and Opportunities(a) One of the challenges in satellite transmission is

cross-talk leading to interference in different fre-quency bands It is necessary to increase the rise andfall times as well as protect sensitive nodes as wecircumvent floating ones if we are to regulate cross-talk as in satellite communication therefore Wefurther need to investigate the possibility of usingdifferential signaling -is constitutes an opportu-nity for further research

(b) Another challenge is that the hidden and exposedterminal problems are still noticeable despite the useof RTSCTS solutions in MAC since collisions arestill possible from transmissions of separate nodesAt the MAC layer ACKs might block efficientchannel usage by prohibiting exposed terminalsfrom reusing the channel and may be eliminatedleaving no proof of sender data packet receptionUsing a directional antenna is one of the proposedtechniques of high-speed WSNs like IEEE 80211-is constitutes an opportunity for further study intothese issues for better solutions though

(c) Another challenge is that since nodes in contention-based MAC protocols are mobile there are manycollisions by the packets and schedule in-consistencies in schedule-based protocols in the two-hop neighborhood information when nodes areentering or leaving [44] Deng et al [45] propose amobility-based clustering (MBC) protocol for mo-bile nodes to enhance the packet delivery rate -isprotocol in comparison with CBR has 25 chancesof decreasing collision-induced packet losses and50 chances in comparison with low-energyadaptive clustering hierarchy (LEACH)

(d) -e process of analyzing and evaluating routingperformance for WMSNs is quite complex becauseof the bursty multimedia traffic Because of the wayencoding scheme frames are structured as well as thechanges in between the scenes that occur naturally

the compressed video frames display some substantialburstiness on several time scales -ere will also befrequent route recalculation that is as well energy in-tensive for nodes frequently capturing and transmittingenergy -e solution here could as well possibly be theuse of the MBC protocol Route recalculation tooshould be dynamic to cater for the mobile nodes whichmight be in and out of the range at different instants oftime Recently the use of smart dumpsters for effectivewaste management in smart cities has also been pro-posed as a possible solution [46]

(e) For WSNs their routing protocols do not look atmultimedia applications which require enormousbandwidth and processing energy and should betransmitted in real time with utmost fidelity Extrachallenges in routing are imposed in WMSNs indeployment of heterogeneous sensors since therecould be audio video and still pictures in addition toscalar data during transmission Heterogeneous datahandled also have different QoS requirements andwith many different business needs different ser-vices with varying requirements give rise to a bigchallenge for the routing design -ere is a need todesign protocols that can support WMSNs such asEQoSA [14] for this purpose

Table 2 summarizes the reviewed MAC protocols forWSNs and the ones designed for suitability in WMSNs

3 QoS Strategies at the Application Layer

We discuss many protocols used to solve different needs ofcommunication between machines

31 ConstrainedApplication Protocol (CoAP) It is a requestresponse protocol designed by the Internet Engineering TaskForce (IETF) for constrained devices such as those havinginsufficient RAM or CPU and WPANs with low-powerconstraints resulting in poor transfer of packets andenormous overhead -e protocol was designed for ma-chine-to-machine (M2M) apps and system automation toreduce bandwidth requirements increase packet transferdecrease the overhead and make work simpler leading tolightweight implementation HTTP commands like GETPOST PUT and DELETE are used in a client-server ar-chitecture Using the publishsubscribe architecture CoAPcan support multiple users and hence yield better perfor-mance via multicasting enabling asynchronous messageexchange -is protocol can support both multicasting andunicasting via UDP and hence is a reliable and simpleprotocol -is is done via the messaging sublayer (for re-liability based on UDP) and the requestresponse sublayer(for communication and interaction) [16]

It uses the following messages

(i) Confirmable message this guarantees reliabilityusing ACK messages sent by the receiver node-ese could be synchronous or asynchronous ifthere is a need for more computational time

16 Journal of Computer Networks and Communications

(ii) Nonconfirmable message here no ACK messagesare sent back

(iii) Acknowledgment (ACK) message this confirmsreceipt of a confirmable message

(iv) Reset message this confirms receipt of an un-processed message in case any critical parts nec-essary for interpreting the message are missing

(v) Piggybacked response the receiver directly re-sponds while getting the ACK message

(vi) Separate response the reply by the receiver is dis-crete from the ACK message

CoAP has no in-built security features It is secured bythe datagram transport layer security (DTLS) that runs ontop of UDP though it was not designed for the IoT -ismakes its suitability debatable since DTLS offers no supportfor multicasting It also needs more packets for handshakingthereby increasing the traffic volume and computationalresources and so shortening the lifespan of mobile devicessince they use batteries [17]

Advantages

(i) It can support multiple users and hence yieldbetter performance via multicasting

Table 2 Comparison of MAC layer protocols used in QoS provisioning

Protocol Features QoS parameter Comments Access Priorityassignment Energy eff Designed for

S-MAC[22] Collision avoidance Energy latency Requires periodic

synchronization CSMA Hybrid Medium WSNs

B-MAC[23] Collision avoidance Energy delay

throughput Flexible interface CSMA Dynamic High WSNs

T-MAC[25] Collision avoidance Energy Complex nonscalable CSMA Dynamic High WSNs

C-MAC[13]

Dynamic channelallocation Energy Multichannel MAC CSMA Dynamic Medium WSNs

CU-MAC[24]

Multichannelmechanism

-roughputlatency High packet delivery ratio TDMA Dynamic mdash WSNs

A-MAC[39]

Per flow serviceguarantees Delay m-ary tree model

dynamic prioritiesTDMACSMA Dynamic Medium WSNs

ER-MAC[30]

Emergencyresponse

Energy delaydelivery ratio

Adapts to traffic andtopology changes

TDMACSMA Hybrid Higher than Z-

MAC WSNs

RI-MAC[31] Collision detection Delay packet

delivery ratioHigh packet delivery ratio

and throughput CSMA Dynamic Medium WSNs

SRI-MAC[33] Collision avoidance Energy delay Good data transfer and

energy consumption TDMA Dynamic High WSNs

H-MAC[34] Slotted sleep time -roughput

latencyData transmission only in

sleep timeCSMAAloha Dynamic Medium WSNs

Z-MAC[36] Collision avoidance Energy delay

throughputVery good channel usage

and error syncCSMATDMA Dynamic Medium WSNs

Q-MAC[37]

Intrainternodescheduling Energy High QoS improved

energy efficiency CSMA Hybrid High WSNs

X-MAC[32]

Overhearingavoidance

Energy delaythroughput

Reduces excessivepreamble by strobing CSMA Adaptive

static Medium WSNs

L-MAC[40] Collision avoidance Network

lifetime energyReduces the number oftransceiver switches TDMA Static Medium WSNs

WiseMAC[38]

No downlinkchannel collision Power delay Significantly low-power

consumptionAlohaCSMA Static Higher than

ZigbeeInfrastructure

WSNsAR-MAC[27] Collision avoidance Energy delay Uses data packets and

control packets TDMA Static Higher thanIEEE 802154 WBANs

BB-MAC[42]

Real-time trafficprioritization Delay Employs round robin in

real-time nodesCSMATDMA Dynamic High WMSNs

COM-MAC [41] Collision avoidance Packet delay

throughputBalances reliability and

retransmissionTDMAFDMA Dynamic High WMSNs

EQoSA[14]

Busty traffic loadaccommodation Energy Provides QoS support for

video and imagesTDMACDMA mdash High WMSNs

QE-MAC[42] Collision avoidance Energy jitter

latencyRuns central polling-like

channel access CSMA Dynamic High WMSNs

R-MAC[26] Collision avoidance Energy delay Reservation based mdash Static High Underwater

SNs

Journal of Computer Networks and Communications 17

(ii) It enables asynchronous message exchange(iii) It is a reliable and simple protocol

32 Message Queue Telemetry Transport (MQTT)Developed by IBM and later improved by OASIS with anintention of decreasing bandwidth MQTT is a simple open-source publishsubscribe protocol analogous to the client-server version It is appropriate for lightweight IoT appli-cations and M2M transmissions in constrained situationslike limited power computation capability and memory orbandwidth -e network bandwidth and the computationalresource usage are decreasing since clients are not updatedmaking publishsubscribe protocols outperform requestresponse protocols via IoT requirements [16] -is protocolcan run over TCPIP and so has the following advantages

(i) Delivering of packets with reliable guarantee(ii) Enabling multicasting(iii) Remote device session establishment(iv) Reduction of transport overhead and protocol ex-

change to minimize network traffic(v) Provision of the notification mechanism when

anomalies occur [47]

MQTTrsquos design targets economical use of bandwidthand battery so it has a major application in FacebookMessenger MQTT ensures QoS and reliability of messagedelivery by providing the following three QoS levels

(1) Fire and forget a message is sent once depending onthe networkrsquos best effort and there is no need forACK (lowest QoS level)

(2) Delivered at least once a message is sent at least onceand an ACK message is needed to avoid duplicates

(3) Delivered exactly once by a four-way handshake itensures a message is sent exactly one time (highestQoS level)-e architecture it has three key components (i)publishersmdashlightweight nodes sending data to thebroker before returning to sleep (ii) broker whichorganizes sensory data in topics and sends them tointerested subscribers and (iii) subscribersmdashIoTapps interested in data sent by sensors through thebroker as shown in Figure 3

MQTT is a simple lightweight protocol that is betterthan CoAP via low and high packet loss probability interms of throughput and exhibits lower latency in lowsampling rates [48] It supports a number of imple-mentations for embedded devices and mobile applicationsand offers web-socket support too Nevertheless it hassome shortcomings

(i) It is only employed in simple data types(ii) It exhibits high latency in high sampling rates(iii) It is unsuitable for real-time apps in such cases(iv) It does not provide automatic discovery andor

sufficient security at the protocol level [49]

33 Secure Message Queue Telemetry Transport (SMQTT)-is is an extension of MQTT that is based on lightweightattribute-based encryption using elliptic curve cryptogra-phy -e protocol has three entities as shown in Figure 4(i) publisher that publishes information under a certaintopic (ii) subscriber that obtains the same topic in-formation via a broker and (iii) public key generator(PKG) or broker that acts as the trusted third party -eprotocol has four phases namely (a) set-up phase (b)encrypt phase (c) publish phase and (d) decrypt phase

(a) Set-up phase publisher and subscriber provide aunique ID with attributes for registration with thePKG -is generates a secret master key set andpublic parameters that it publishes with set U A1A2 An For ciphertext policy-attribute-basedencryption (CP-ABE) the PKG gives a key set todevices with attributes [50]

(b) Encrypt phase based on the access tree the publishercreates an access policy An access tree is sent by thepublisher to the PKG for generating the key-gen-eration policy in the key policy- (KP-) based ABEalgorithm -e publishers know beforehand thetopics and subscribers to access as well as the rulesfor access Subscribers are issued with private keysfor all communication sessions Publishers in theCP-ABE algorithm come up with access trees andrules where the payload is encrypted by the pub-lisher who also issues more info needed for de-cryption and the rules With the help of public keycryptography data are encrypted giving rise to ci-phertext on the basis of KPCP-ABE and the receivergets SUBACK from the broker [51]

(c) Publish phase ciphertext is embedded in theSPublish command as payload and topic name isdefined in the header by the publisher -e brokerreceives the SPublish packet and replies withPUBACK prompting the publisher to respond withthe PUBREL packet -e broker then broadcasts totopic subscribers and deletes the data forwarding thePUBCOMP packet to the sender node

(d) Decrypt phase the encrypted information isdecrypted by the subscriber with the help of a secretkey Considering KP-ABE on verification that itsatisfies the access policy the subscriber invokes thePKG for a corresponding key which it sends afterverifying the request With CP-ABE subscribers usethe private key for decryption after authenticationthereby enabling offline interaction [51]Advantages

(i) -e information being encrypted makes it moresecure

34 MQTT for Sensor Networks (MQTT-SN) -is is a ver-sion of MQTT designed to operate under environments oflow bandwidth high link failures and short message lengthIt is a publishsubscribe (pubsub) protocol designed for

18 Journal of Computer Networks and Communications

UDP-based WSNs to visualize communication for low-power devices [5152] It comprises a publisher device actingas the MQTT-SN client to forward communications to agateway that also changes the MQTT-SN message to theMQTTmessage before delivering it to the broker that sendsit to MQTT-SN subscribers [51]

It has a gateway for changing protocols from MQTT toMQTT-SN (Figure 5) [52] -is protocol supports nodes in

the sleeping mode using an offline keep-alive procedure thatcan be used by battery-operated devices to get into thesleeping mode To support sleeping clients MQTT-SN has anew offline keep-alive procedure with which battery-oper-ated devices can go to the sleeping state in which theirmessages are buffered at the server until they are awake InMQTT-SN clients with no preconfigured gateway addresscan figure out the real address of their operational gateway

Publisher

Subscriber

Subscriber

Broker

Publisher

4G

Figure 3 Architecture of the MQTT protocol

Register by mentioning URI Register by mentioning URI

Access tree

Set-up

Publisher ReceiverPKG(broker)

Key management Key management

Encrypt data using public parametersand random key This is done using

CP-ABE Publish it as

Subscribe (topics)SUBACK

S-Publish (TopicEncrypt (data))S-Publish (TopicEncrypt (data))

Encryptand publish

PUBACK

PUBACK

PUBRELPUBRELPUBCOMP

PUBCOMP

Figure 4 SMQTT protocol

Journal of Computer Networks and Communications 19

within the network through a discovery procedure Simul-taneously we may have several gateways in a single wirelessnetwork that can be cooperative via load sharing

MQTT-SN architecture it is composed of MQTT-SN clients gateways and forwarders (Figure 5)rough the MQTT-SN gateway (GW) the MQTT-SN protocol is used by the MQTT-SN clients toconnect to the MQTT server Integration betweenthe MQTT-SN GW and the MQTT-SN server mightbe possible or not Translating between the MQTT-SN and the MQTT for a stand-alone GW is doneusing the MQTT-SN protocol If a GW is detachedfrom the network it can be accessed by the MQTT-SN clients through a forwarder is just encapsu-lates the received MQTT-SN frames that are on thewireless side before forwarding them to the GW un-altered It then decapsulates the received frames tooresending them to the client as sent by the GW [52]ere are 2 types of gateways (a) transparent gatewayand (b) aggregating gateway as shown in Figure 6

(i) Transparent gateway this launches and sustains adirect MQTT connection between the MQTT-SNclient and the MQTT server which is entirely re-served for end-to-end client-server messaging enumber of servers and that of MQTT-SN clientsconnected to the GW are the same is transparent

GW translates syntax between these protocols and iseasier to implement compared to the aggregatingGW But the MQTT server must support dierentsessions for all connected clients Some MQTTservers are congured to enable few parallel sessions

(ii) Aggregating gateway this maintains a single MQTTconnection to the server on behalf of all clients Allexchanged messages from the MQTT-SN client to theaggregating GW are directed to the GW which sortsthe information to forward to the server e aggre-gating GW decreases the number of MQTT sessionssimultaneously supported by the server in WSNs withmany battery-operated sensors and actuators (SAs)having limited storage and processing capabilities It ishowever more complex to implement compared to atransparent GWe general format of the MQTT-SNprotocol message has two or four message headeroctets and n octets of variable header that is optionaldepending on the message in question MQTT-SN has256 message types (MsgType) with those from0X1Eminus 0XFD 0X19 and 0XFF reserved for futureusage [51]

Advantages

(i) e protocol can operate under environmentsof low bandwidth high link failures and shortmessage length

MQTT-SN gateway

MQTT-SN

MQTT

MQTT-SN

MQTT-SN

MQTT-SN

MQTT-SN client

MQTT-SN clients

MQTT broker

MQTT-SN client

MQTT-SN client

MQTT-SN gateway

MQTT-SN forwarder

Figure 5 MQTT-SN architecture

20 Journal of Computer Networks and Communications

(ii) It is a pubsub protocol designed for UDP-basedWSNs to visualize communication for low-power devices

(iii) It can be optimal for low-cost battery-operatednodes having low processing power and lessstorage

(iv) It supports nodes in the sleeping mode using theobrvbarine keep-alive procedure used by battery-operated devices

35e Extensible Messaging and Presence Protocol (XMPP)e standardization of this protocol by the IETF made itpopular for usage over the Internet especially in instantmessaging and chatting platforms It is however susceptibleto spamming attacks and its failure to have worldwidesupport has made Google too withdraw its support of thestandard [17] e protocol is mainly useable in the client-server architecture but can also be applicable in client-clientXMPP communication [53] Among the application layerprotocols XMPP is the only one that supports both asyn-chronous (publishsubscribe) and synchronous (requestresponse) services while running over TCP but can still runover other protocols

Given the simplicity of its implementation and ease ofusability in XMPP instant messaging (IM) is suitable forXMPP and is implemented using the message stanzaltmessagegt ese stanzas are usable in one-to-one ormultiuser sessions or chatting or in error reporting andsending headlines Attribute type of the stanza denes themessage type e attribute type could be a chat group chatheadline normal or error An instant messaging network inXMPP can support many XMPP clients and servers with agateway congured to enable other IM network users toconnect to other XMPP users e XMPP server mainlycontrols sessions and XML streams to clients and servers inaddition to routing XML stanzas in the streams [53] eTCP ports 5269 and 5222 have been respectively registeredfor the XMPP server-to-server connections and XMPP clientconnections by the Internet Assigned Numbers Authority(IANA) XMPP enables specication of XMPP ExtensionProtocols (XEP) to enhance its usefulness [17]

XMPP security clients connect to the server usingtransport layer security (TLS) and simple authentication andsecurity layer (SASL) protocols which are at the same timeused by the servers for interdomain communications Fordata condentiality and integrity TLS will encrypt thestreams of XML for all sessions To conrm the identity of anyclient that needs to access the server the SASL protocol doesthe authentication of the XML stream For any communi-cating session with a server to start the client needs to resolvethe serverrsquos DNS hostname For protection of the server of theclient from third-party attacks the server maintains the au-thentication details secret ie IP address and access methodand only original connections of the server are required

In server-to-server communications the SASL protocolis employed for authentication and condentiality Server-to-server communications may as well be deactivated from aserver by an administrator depending on the organizationalpolicy To guard against domain-spoong attacks serverdialback can be a good solution since servers support itsecuring more the XML stanzas Nevertheless it cannot beused to authenticate secure or encrypt in-server streamsand the server identication resulting therein is not strongenough to be relied on It further cannot secure against DNSpoisoning attacks and IP session hijacking for remote do-mains [53] High-security domains can use TLS and SASL

Advantages

(i) It is popular for usage over the Internet especially ininstant messaging and chatting platforms

Disadvantages

(i) It is susceptible to spamming attacks(ii) It has failed to have worldwide support making

Google withdraw support of the standard(iii) It is not secure against DNS poisoning attacks and

IP session hijacking for remote domains

36 Advanced Message Queuing Protocol (AMQP) isprotocol can be utilized by a number of platforms formessage exchange since it is open source and written inmany languages It is a publishsubscribe model depending

MQTT-SNclients Broker

MQTT

(a)

MQTT-SNclients Broker

(b)

Figure 6 (a) Transparent gateway and (b) aggregating gateway

Journal of Computer Networks and Communications 21

on a reliable and efficient messaging queue with capacity forinstant messaging or e-mail for different applications es-pecially in IoT environments It differs from others in thesense that message types can be specified and have theirsource traceable and how to do trade-off between perfor-mance security and reliability For reliability of messagetransmission AMQPmakes use of some delivery guaranteeslike at most once at least once and exactly once It can alsohowever utilize the TCP transport layer for guaranteeingreliability -e AMQPrsquos publishsubscribe method has twocomponents namely (i) exchange queue meant for routingof messages to an appropriate order in the queue and (ii)message queue that acts as storage pending their delivery toreceivers [16] In this protocol model the broker can makerouting decisions contrary to othermessaging systems whereapps using the queue have in them embedded the decision-making logic for recipients or senders -e developer mayneed to check each of the affected apps so as to change thelogic for routing and delivery of the messages [54]

Advantages

(i) It is convenient(ii) It has capacity for interoperability between dif-

ferent vendors(iii) It can use open standards to enable connection

between business partners(iv) It can enable innovations based on AMQP foun-

dations [55](v) It can enable supporting of mission-critical ap-

plications in e-commerce(vi) It can guarantee reliable delivery of messages

Disadvantages

(i) It is not very appropriate for real-timeapplications

(ii) It may not offer automatic discovery(iii) Interoperability is not guaranteed(iv) It lacks open-source libraries for constrained

gadgets [49]

37 Data Distribution Service (DDS) A product of theObject Management Group (OMG) DDS is a pubsubprotocol meant for M2M communications with two sub-layers -ese are the data-centric pubsub sublayer re-sponsible for connecting anonymous data publishers tosubscribers and the data-local reconstruction sublayer that isoptional and integrates DDS into the application layer [56]-e DDS architecture connects system components asshown in Figure 7 Publishers and subscribers are discon-nected with respect to (i) time (ii) space (iii) flow andbehavior (iv) platforms and (v) programming languages[57] A DDS data publisher generates topics and publisher-subscriber coupling is represented as topic name data typeschema and publishers and subscribersrsquo QoS attributesDelivering of data is done by the publisher layer -e datawriter and publishers together make a decision about anynecessary alterations for subsequent forwarding to thesubscribers who need to send data to IoT applications Data

readers are responsible for delivering topics to the sub-scribers after reading through them [56]

Some of the major components of the DDS architectureinclude the following [57]

(i) Topic this is a logical path specifying publicationand subscriptionrsquos data type between DataWritersand DataReaders For a successful session to begintopic names types and DataWriters and Data-Readersrsquo QoS should correspond

(ii) Domain this is a logical setting for transmissionthat can be utilized for separating and optimizingnetwork transmission in a setup of similar appli-cations If DDS apps share a domain ID then theycan exchange data among themselves

(iii) DataWriter and DataReader these are terminalobjects for writing and reading messages re-spectively to and from a universal data space

(iv) Participant this is an object that stands for apublisher or subscriber of the DDS app in a do-main acting like a container for other objects

Advantages

(i) It is highly reliable and offers QoS support(ii) It is highly scalable(iii) It is appropriate for real-time apps(iv) It is fault tolerant(v) It has high level of interoperability(vi) It does not need brokers that may act as

bottlenecks(vii) It offers automatic discovery(viii) It has a user-defined data structure for topic(ix) It supports publisher and subscriber decoupling

Disadvantages

(i) It is originally developed for only stand-aloneLANs

(ii) A lot of memory is needed(iii) It has no open-source libraries to support con-

strained devices(iv) It is hard to design and configure compared to

other protocols [49]

38 RESTful Services REpresentational State Transfer (REST)-is is an architectural style based on a set of principles thatdefine and address networked resources It offers web servicesthat enable transmission from the HTTP device to the devicewithin the IoT architecture and enables clients to reach theserver [16]-is protocol utilizes HTTP commands to backupthe requestresponse messaging model HTTP is a popularand secure protocol in the WWW using TLSSSL [16] ARESTful application has the following characteristics

(i) State and functionality are separated into distrib-uted resources

(ii) All resources are distinctively named by the use ofunique HTTP commands (GET POST PUT orDELETE over the Internet)

22 Journal of Computer Networks and Communications

(iii) It is a clientserver stateless layered protocol thatsupports caching REST uses the same methods asHTTP for all requests or responses with POST andGET for creating and retrieving resources PUT forupdating and changing the state of resources andDELETE for removing resourcesAdvantages of REST over arbitrary web services(SOAP) in the IoT context

(i) It reduces overhead(ii) It reduces parsing complexity(iii) It is stateless(iii) It integrates more tightly with HTTP [58](iv) It is easy to implement and learnis together

with its natural t over HTTP makes it apreferred method for Web 20 applications toshowcase their data and secure M2M sessionsin the IoT

(v) Applications that support RESTful web ser-vices exhibit better performance than the re-source-constrained WSN nodes [59]

(vi) REST is important for the IoT since it supportscommercial M2M cloud platforms

(vii) Its implementation in smart phones and tabletapps is also easy since it needs only an HTTPlibrary that is accessible to all operating systemplatforms [56]

Disadvantages

(i) Although it enjoys wide usage in commercialM2M platforms it is not likely to dominatebecause it is hard to implement

(ii) Its HTTP usage makes it incompatible withconstrained M2M devices leaving its usage toend applications

39 WebSockets It is a web-based protocol operating on asingle TCP channel and providing full-duplex transmissions

It permits secure duplex communication between a clientexecuting unreliable code in a restricted environment and adistant node communicating from that code while using theorigin-based security model employed by web browsers overTCP [60] e protocol is not a requestresponse model andis not a publishsubscribe model So unlike other protocolsa web-socket session is established by the client initiating ahandshake to the server to launch the session is onceestablished a full-duplex client-to-server connection beginsasynchronously and runs until terminated by either party[56] It gives a model for browser-based apps utilizing bi-directional communication with no need of many HTTPconnections to servers

WebSocket comprises two parts ie (i) the handshakehaving a clientrsquos message and the serverrsquos handshake responseand (ii) the data transfer HTTP headers should be repeated byapplications in all client requests and server responses arisingfrom the endless polling which might increase the commu-nication overhead subject to the application [61] WebSocketcan be used to come up with scalable real-time web appli-cations since it can enable full-duplex transmission via onesocket across the Web Its security can be ensured with thehelp of TLSSSL is protocol is suitable for resource-con-strained devices and IoTapplications will not have a good userexperience with it since it operates a clientserver architectureNevertheless it can outcompete all TCP-based protocolsbecause of the following advantages

(i) It is t for real-time transmission(ii) It has improved security(iii) It minimizes transmission overhead(iv) It can provide ecient messaging systems using

WebSocket ApplicationMessaging Protocol (WAMP)[17]

310 Streaming (Simple) Text-Oriented Messaging Protocol(STOMP) is is a TCP-based protocol designed for textmessage-oriented middleware (MOM) with the use of

Topic

Datareader

Datareader

Datareader

Datawriter

Datawriter

Datawriter

Subscriber PublisherPublisher

Data bus (DDS domain)

Subscriber

Topic Topic

Figure 7 DDS architecture

Journal of Computer Networks and Communications 23

HTTP-like commands for interoperability among platformslanguages and brokers Data communication takes placebetween the client and the broker in line with commands(like CONNECT DISCONNECT ACK NACK SUB-SCRIBE UNSUBSCRIBE SEND BEGIN COMMIT orABORT) then headers in the form ltkeygt ltvaluegt (one perline) then a blank line and then body content terminatingin a null character Server-client communication takes placeover a MESSAGE RECEIPT or ERROR frame with headerand body content format being analogous MOM productssupporting STOMP include Apache ActiveMQ or FuseMessage Broker HornetQ Open Message Queue(OpenMQ) RabbitMQ and syslog-ng through its STOMPdestination plugin [62] In addition to these other appli-cation layer protocols include simple media control protocol(SMCP) lightweight local automation protocol (LLAP)simple sensor interface (SSI) lightweight M2M (LWM2M)XMPP-IOT and simple object access protocol (SOAP)

Advantages

(i) It is lightweight for constrained networks(ii) It has flexibility to choose quality of services with

the given functionality(iii) It is standardized by OASIS Technical Committee(iv) It is easy and quick to implement

Challenges and opportunities

Here security threats are imminent because of the open-endedness of the application layer [63] -ese could includethe following

(i) -ere is poor or lack of a security design for anapplicationrsquos function Coming up with a gooddesign to address this could be a good ideaopportunity

(ii) Some of the programs might have backdoorsevading otherwise secure controls to allow un-authorized access to network resources We requiremechanisms to patch up such loopholes

(iii) Applications that lack or have weak authenticationmechanisms can be easily targeted by unlicensedusers to abuse the network Such mechanisms needto be strengthened further

(iv) Excessively complicated access control rules maybecause of their misunderstanding or poor writinggive unauthorized access allowing dubious activityrather than protecting the network they purport toprotect -ese should be simple and clear to un-derstand by all users

(v) In booting workstations without disks and managingnetwork devices we usually use the TFTP protocolalthough access does not require authenticating theuser by usernamepassword -is makes an intruderable to gain easy access to configaccess info as longas he can guess the filenames -is needs to be fixed

A few steps can be used to secure the application layerTo protect network data transmitted across the network

applications need to strengthen their encryption and au-thentication mechanisms Apps too need to enable strictcontrols of privileges for data access using good mechanismsto balance between usability and effectiveness [63] Appli-cations handling sensitive data need detailed logging andaudit capability as well as testing and reviewing

Table 3 summarizes the application layer protocols dis-cussed with their weaknesses and differences among them

4 Cross-Layer QoS Strategies

41 Service-Differentiated Real-Time Communication Scheme(SDRCS) -is is an event-driven routing protocol that routesreal-time traffic by featuring a cross-layer packet-forwardingdesign in which it integrates the real-time routing functionalitywith a new priority-based MAC scheme -e protocol ap-proximates distributed packet traversal speed for traffic clas-sification and admission control on the basis of this design Itfurther localizes decision-making by prioritizing packet for-warding to maximize the speed of packets [64]

Advantages

(i) It improves bandwidth utilization bandwidthdegradation due to unschedulable packets can beavoided

(ii) It offers a lightweight packet schedulability estima-tion mechanism Here it uses received signalstrength and admission control and then earlymissed-deadline packet-dropping policies are made

(iii) It delivers high end-to-end throughput via highersource data rates alongside strict end-to-end latencyneeds

(iv) Nodes which meet real-time requirements of lesstraffic load and better channel quality are givenhighest priority when it comes to packet forwarding

-e SDRCS operates via 5 mechanisms namely (1) RSS-(Rich Site SummaryReally Simple Syndication-) basedgrouping (2) admission control (3) prioritized queuing(4) real-time MAC and (5) dynamic forwarding -isprotocol has a queuing policy that employs per-hopdeadline-based queues and the nodes schedule theirpackets using FIFO priority-based queues [1] -e prin-cipal features of this protocol are mentioned as follows

(i) It is event driven and is easily adaptable to anynetwork changes

(ii) It does not need additional hardware for locali-zation or multichannel transmission

(iii) -e cross-layer packet-forwarding strategy can beused when transmitting multimedia traffic

(iv) It is able to efficiently circumvent any voids therein

Early deadline miss (EDM) policy drops all unscheduledpackets -e SDRCS circumvents wrong packet dropsdue to EDM since it adjusts to network changes easily

42 Sensor Fuzzy-Based Image Transport (SUIT) -is pro-tocol regulates congestion using a scheme based on fuzzy

24 Journal of Computer Networks and Communications

logic It transmits packets with reduced quality to allowablelevels if there is congestion enabling it to transmit lower-quality packetsframes without being dropped-e numberof packets delivered per second thus rises and while thisenhances video streaming apps it is more appropriatelymade for those that quickly transmit non-real-time videosuch as in video surveillance SUIT can stream JPEG pic-tures other than the traditional predictive video codingmethods that are not suitable for sensor devices because oftheir constraints of energy memory and computationalspeed Progressive JPEG-PJPEG performs better via packettransmission rate and delay though it degrades the imagequality because of congestion [65] -is protocol employscross-layer methods for interlayer data exchange though itstill employs some layering making it still possible topreserve all advantages of the layered design since it hasapplication and transport layers but has noMAC or routinglayer

For congestion detection and control SUIT bases on thepercentage number of in-bound and out-bound frames perwindow number of contenders and buffer occupancy of thenext-hop node Since buffer management has an effect oneffectiveness and QoS inWMSNs it is useful in the transportlayer so the protocol prioritizes packets for better QoS SUITreorders packets by tracking the sequence numbers of thedelivered packets since they all have a source ID framenumber and sequence number that it uses to detect anymissing packets -e protocol however does not providepacket reliability [65]

Advantages

(i) -e protocol prioritizes packets for better QoS(ii) It can transmit lower-quality packetsframes

without being dropped

Disadvantages

(i) It does not provide packet reliability

43 Network Layer QoS Support Enforced by a Cross-LayerController (NLQS) -is scheme permits packet-level servicedifferentiation as a function of throughput end-to-endpacket error rate and delay [66] (see Figure 8) -is en-hances network layer QoS It comprises a cross-layer controlunit (XLCU) for configuring and controlling networkingfunctionalities at physical MAC and network layers [18]-is depends upon the unified logic which affects choices forapplication layer needs and status of functional blocks whichdo implementation of networking functions [66]

Advantages

(i) Cross-layer interactions can be controlled withoutweakening the upgradability modularity andsimplicity of designing the system

44 Cross-Layer Signaling Shortcuts (CLASS) CLASS has ahigh level of effectiveness flexibility and comprehensive-ness According to [67] it has the following features

(i) Direct signaling among remote layers leading tofaster transmission

(ii) Lightweighted internal message format standard-ized protocols are not suitable for use in internalsignaling since they are not usually lightweight forinstance transmitting against faults in the networkCLASS requires only three fields namely destina-tion address event type and event contents

(iii) -e setup of its exterior messages is standardconsidering external signaling ICMP and TCPIPare suitable for general messages and short notifi-cations respectively

(iv) Message control protocol should ensure optimizedand organized exchange of dense simultaneousmessages across layers for high efficiency andavoidance of possible conflicts -e protocol appliesto many cross-layer signaling situations

Table 3 Differences and problems with application layer protocols

Protocol Architecture Transport QoS options Security Weaknesses

CoAP Requestresponse UDP DTLS No in-built security features DTLS does not supportmulticast

MQTT Publishsubscribe TCP TLSSSL Insufficient security at the protocol level sincepayload values are not encrypted

SMQTT Publishsubscribe TCP CPKP-ABE Key revocation and group pubsub for distributedSMQTT are still a challenge

MQTT-SN Publishsubscribe UDP TLSSSL SSLTLS suffers from attacks like BEAST CRIMERC4 and Heartbleed

XMPP RequestresponsePublishsubscribe TCP X TLSSSL Susceptible to spamming attacks and lacks worldwide

supportAMQP Publishsubscribe TCP TLSSSL Inappropriate for real-time applications

DDS Publishsubscribe TCPUDP TLSSSL Memory intensive and no open-source libraries forconstrained devices

REST Requestresponse HTTP X HTTPS Hard to implement Uses HTTP so incompatiblewith constrained apps

WebSockets PublishsubscribeClientserver TCP X TLSSSL Bad user experience for IoT apps since it runs the

clientserver architectureSTOMP Clientserver TCP HTTP -e broker can act as a bottleneck

Journal of Computer Networks and Communications 25

Advantages

(i) e protocol is scalable and has a very lowpropagation delay

(ii) It is very ecient and exible since it uses directsignaling between any two layers

(iii) It is very comprehensive and eective

45 Cross-Layer and Multipath-Based Video Transmission(CMVT) CMVT is a hybrid of application and networklayers At the application layer it carries out the encoding ofvideo streams to video data frames (I- P- and B-frames) bymeans of the MPEG-4 encoding format At the networklayer route discovery and data transfer take place Twoalgorithms greedy forwarding and rollback are used todiscover a number of paths from the source to the sink nodevia the route discovery method [20] A specied node icalculates the evaluation of its adjacent node j using thefollowing equation

fij (1 minus α)d2(j D) minus d2min(i)d2max(i) minus d

2min(i)

+ αeinit(j) minus eres(j)

einit(j)

(19)

where fij evaluation value for the node i to j d2(j D) isthe distance from the node j to the destination node D

d2min(i) and d2max(i) are the minimum and maximum dis-

tances of neighbors of the node i to D respectively einit(j) isthe initial energy of the node j eres(j) is the current residualenergy of the node j and α is the energy coecient given asfollows

α emax(i) minus emin(i)

emax(i) (20)

where emax(i) and emin(i) are respectively the maximumand minimum energy left of all node irsquos neighbors

e network layer further transmits video streams withCMVT doing status evaluation choosing an appropriatecommunication path for any packets and the QoS guaranteelevel for the path i is calculated as

fi (1 minus ω)hisum hi

+ ωnisum ni

(21)

where fi is the evaluation value for the path i hi are the hopsfor the path i ni is the summation of packets sent via isum ni isthe summation of packets sent by sources and ω is theenergy consumption factor [18]

Advantages

(i) e CMVT protocol is very superior in mediatransmission particularly large-scale WMSNs

To network peers

Flow requirements

-

Atimes

Admissioncontrol

Capacityassignment

Next-hopselection

Coding rate

Data rate

QoScontracts

Cross-layer controlunit (XLCU)

Localization subsystem

Synchronizationsubsystem

Networkcoordinationunit (NCU)

UWB modulator

Channel coder

QoS scheduler

Georouting

Source coder

Rate control

Figure 8 Cross-layer controller architecture

26 Journal of Computer Networks and Communications

46 Cross-Layer Cooperative MAC (CoopMAC)CoopMAC [68] is based on the IEEE 80211 DCF approach-ere is a reduction in interference of neighboring cellsleading to uniform coverage in densely deployed networks[68] A connection is established by two-way authenticationbetween the sender and the receiver When a Clear-to-Send(CTS) packet and short interframe space (SIFS) are obtainedby the sender packets are transmitted right to the receiverwith no valuable cooperation If available the two nodes findout whether they can exchange data -is is done by the useof a helper identification signal to establish how feasiblecooperative communication is It is established in thepresence of a signal and if it is not available direct com-munication is activated For distributed systems helper-initiated cooperation is preferable because of the presence ofthe RTSCTS packets [69]

Advantages

(i) CoopMAC has an advantage over others due to thespatial diversity amongst the three nodes [70]

(ii) It has the potential to achieve significantthroughput and delay with no complexity in thedesign of the system

47 Cooperative MAC Protocol for Multihop Networks (M-CMAC) Just as CoopMAC M-CMAC is made in such away that high data rate stations help low data rate stations toforward traffic for broadcasting [71] -ere are also helperschosen in such a way that 2 fast-hop transmissions replace aslow hop transmission -e helper via whom there existsleast delay from the sender to the receiver is considered bestbecause of theminimum two-hop transmission rate and so isthe chosen neighbor to the duo Here the working as-sumption is that all nodes have their position coordinatesknown and thus the Euclidean distances between all pairscan be calculated and converted to the data rate for that linkAll nodes have cooperative tables (CT) of potential helperswith destination and helper MAC address Euclidean dis-tance and total distance through the helper Any sender withdata to transmit checks out for the existence of any helper inthe CT for the destination and if available forwards RTS forchannel reservation for single-hop duration M-CMAC hasgot an RTS format having a shape with five fields similar tothe one given as follows

Framecontrol Duration Source

addressDestination

addressHelperaddress

-e destination nodersquos address is kept by the source in thepart for the helper address that is in turn saved as the desti-nation address Nodes inspect helper and destination addressfields on receipt of RTS whereby the node acts as the othernodersquos helper in case the helper address field is not the sameand the helper transmits CTS back to the source if at all it wantsto forward data On delivery of the CTS packet the sourcetransmits to the helper which in turn sends the packets to thedestination that also sends ACK to the helper on receipt of thepacket In M-CMAC there exists a higher level of channelreuse (parallel transmissions) due to the increased number of

nodes therein that subsequently raises availability of helpers fordata forwarding -is results in increased throughput incomparison with CoopMAC and IEEE 80211 DCF [71]

Advantages

(i) It promises a higher throughput compared toCoopMAC and IEEE 80211 DCF

48 Cluster-Based Cooperative Routing (CBCR) Protocol-e CBCR protocol has a multihop data-forwarding func-tion realized at the link layer with cooperative links that useM-CMAC -is protocol encompasses two stages namelythe routing relay selection phase and data forwarding phase

(i) Routing relay selection phase all nodes announcetheir presence to their neighbors by broadcastingperiodical beacon messages that carry the MAC ad-dress of the node Each of these constructs a relaytable that contains all neighboring nodes with whichit is able to communicate -e node further broad-casts its neighbor list in case of any change to itsentries since the previous broadcast -e MAC ad-dresses of nodes next to the node X are contained inits relay tablersquos column one and the neighbor nodersquosrow has MAC addresses for neighbors of the adjacentnode Based on its relay table each node in-dependently chooses routing relays Choosing a nodeas a relay node depends on the number of nodes itconnectsmdashit should be the highest number [71]

(ii) Data forwarding phase a node with packets totransmit needs to first verify if the receiver is in thesame cluster and if that receiver has a helper to whichthe packet is thence forwarded but in whose absencethe packet is delivered right to the receiver In casethe intended receiver is in a different cluster therelay table will be checked to see whether that re-ceiver is reachable via other routing relays If it isreachable packets will be sent to the routing relaydirectly or via the helper if at all the relay has one Incase of destination unreachable via relays packetsare multicast to all relays by the node [71]Advantages

(i) It enables multicasting

49 MAC-PHY Cross-Layer Protocol A cooperative cross-layer standard for cooperation at the physical layer in next-generation WMSNs is developed in [70] It provides acomplete MAC layer algorithm that gives a supportive shellfor the PHY-MAC layer Like CoopMAC its scheme de-pends on the middle node for communication betweennodes-eMAC layer protocol has been changed to regulateinformation exchanges at the physical layer -e receivernode receives duplicate packets from source and helpernodes to decode the data [70]

410 MAC-Centric Approach -is cross-layer protocoltargets multimedia applications by using the MPEG-4

Journal of Computer Networks and Communications 27

scheme [21] It is characterized by 4 access categories(AC3ndashAC0) according to their priority in transmission It ismeant for supporting different QoS needs in upcomingvideo applications and enabling MAC layer differentiationfor H264 partitioning [18] Because of low bandwidth delayand other QoS challenges that cause inefficiency in trans-mittingmultimedia data inWMSNs a number of algorithmsdepending on IEEE 80211e have been proposed to supporttransmission of quality videos [72] Selecting an AC dependson the measures for QoS such as the loss rate and delay-us AC3 having the highest priority is mapped to theparameter set concept since the stream is sensitive to bit lossespecially in video transmission [21]

411 Adaptive Cross-Layer Forward Error Correction(ACFEC) In the ACFEC [73] model data packets are ex-changed among nodes through the access point (AP) whichoperates the infrastructure mode in which it adds FEC to videodata [72] -ese data are dealt by encapsulation using astreaming server to the receiver through thewireless AP as RTPpackets An adaptive FEC controller senses the category ofpackets out of the RTP header and recovers the header ofpackets from UDP -e encoder generates some error-cor-recting packets whose number is determined by the sourcepacket number of the block Multimedia transmissions aremonitored by the controller using MAC failure data and itscounter is increased by one if there is a failed transmission-econtroller uses the failure counter to change the packet numberproduced after transmitting a block [18] In case there are lostpackets it changes the redundancy rates and produces extrapackets to replace the lost ones and satisfy the requirements ofthe receiving node [72] -e FEC packet number is enhancedor reduced to satisfy the receiverrsquos requirements and stoppacket losses-is is done through accurate detection of packetlosses and adjustment of redundancy rates If all video datapackets are well received there will not be generation of FECpackets [73]

Advantages

(i) It promises a good QoS through packet loss re-duction and redundancy rate adjustment

412 Balanced Cross-Layer Fuzzy Logic (BCFL) DesignRouting Algorithm in WSNs A new fuzzy logic-basedrouting algorithm (BCFL) [74] was designed using disper-sion of the cross-layer parameter as the fuzzy logic inferencesystem input Each cross-layer parameter has a dynamicweight depending on the value of dispersion -e designcomes with some innovations as follows (i) for fuzzy logicinference system input the absolute parameter value issubstituted by parameter dispersion thereby significantlyreducing algorithmic complexity (ii) dispersion does notchange with the order of magnitude according to the dis-persion formula and (iii) the weight of the parameter de-pends on the size of its dispersion and the two are inverselyproportional to each other -is enables BCFL to have somenobility unlike other algorithms as per the following dis-tinguishing properties making it advantageous

(i) It has got simple if-then rules which remain con-stant even when the constraints increase

(ii) It is capable of dealing with many constraints withno increment in complexity

(iii) It has capacity to yield a more balanced solutionthan other algorithms

(iv) It is easily adaptable to changes in network con-ditions and topology even when the changes arefrequent like in underwater WSNs -e algorithm isuseable in choosing the CH in cluster-based routingprotocols [74]

413 Minimum Hop Disjoint Multipath Routing Algorithmwith Time Slice Load-Balancing Congestion Control SchemeMHDMwTS is a two-phased routing protocol comprisingpath build-up and path acknowledgment phases It consists ofmultiple sources each of which has three build-up disjointpaths namely primary alternate and back-up paths Onactivation the source node requests to build up a route to thenearest hop neighbor which is the path build-up phase [19] Inthis phase step one has source activation in which therequested node for path building adds its number and time-stamp and sends to the least hop-count neighboring node-isgoes on until the least time latency sink having the informationneeded to construct the primary route is reached [75] In steptwo there is path extraction when the new package fromanother path arrives whereby the extracted path is comparedwith the primary path If the node is shared the package will berejected or an alternative path will be searched for to get abackup through comparison of the preceding two paths Phasetwo is path acknowledgment where we have the third step inwhich the sink returns the ACK packet to the sender with pathinformation having nodes and their time information after ithas computed it using the timestamp [1819]

Advantages

(i) -e protocol minimizes end-to-end latency(ii) It enhances congestion control

414 Cross-Layer Optimal Design (CLOD) Authors in [76]come up with a CLOD for scheduling at the data link layerrouting at the network layer and controlling congestion atthe transport layer with an assumption of fixed link capacity-rough congestion control energy efficiency is improvedTransport layer congestion at the nodes is minimized bycompressed sensing (CS) in which transmitted bits are re-duced whereas optimally allocating resources decreasescongestion on the links at the data link layer CLODpromises minimized computational complexity and betterperformance in light traffic scenarios [77] Generating andstoring CLOD is more efficient compared to Gaussianrandom matrices It prolongs network lifetime and savesenergy

Advantages

(i) It does congestion control subsequently increasingenergy efficiency

28 Journal of Computer Networks and Communications

(ii) It reduces computational complexity(iii) It improves performance in light traffic scenarios

415 Challenges and Opportunities of the Cross-Layer Design(CLD)

(i) Different layers in a real network assume differentfunctions andor services A layer only communi-cates with its neighboring layers -e layered modeldegrades system performance because of many

features in wireless transmission QoS parameters inthe layers of WMSNs are shared by routing pro-tocols to optimize the performance Neverthelessthere is a need to design cross-layer models forincreased efficiency in routing [1]

(ii) -e physical layer plays a highly vital role in CLDRate adaptation and channel allocation take place atthe physical layer via signal processing to enhanceQoS End-to-end performance in wireless media isaffected by the changes therein given the effect on

Table 4 A summary of some of the reviewed cross-layer models

Protocol Layers Aims Comments QoS parameters

SDRCS MACPHY Routing real-time traffic Transmits multimedia traffic viacross-layer packet forwarding

-roughputlatency

SUIT Applicationtransport Transmission of non-real-time video Uses fuzzy logic to regulate

congestionTransmission rate

delay

CLASS Any twoDesign serves as a framework fordifferent implementations ofdifferent application scenarios

Scalable very efficient flexible andhas very low propagation delay

Propagation delayjitter

NLQS PHYMACnetwork

Permits packet-level servicedifferentiation as a function of

throughput packet error rate anddelay

Network layer QoS is enhanced -roughput delaypacket error rate

CMVT Applicationnetwork

Encodes video streams to video dataframes via MPEG-4 encoding anddoes route discovery and data

transfer

Greedy forwarding and rollback areused to find source-to-sink paths Energy

CoopMAC PHYMAC Offers spatial diversity among thethree nodes

Helper ID signal finds out howfeasible cooperative communication

is

-roughput ratedelay

M-CMAC PHYMAC Increases end-to-end throughputand packet delivery ratio

Euclidean distances between nodesare calculated and converted to the

data rate for a link

-roughputpacket delivery

ratio

CBCR PHYMACMinimizes control overhead andtime consumed in establishing thecooperative paths than M-CMAC

Energy consumption is moreuniformly distributed in a network

enhancing network lifetime

-roughputpacket deliveryratio energy

MAC-PHY MACPHYMeant to enable PHY layer

cooperation and maximize gains ofcooperation at the MAC layer

Leverages both spatial diversity andcoding gain -roughput delay

MAC-centric MACAPPL

Meant to support QoS needs in newvideo apps and enable MAC layer

differentiation for H264partitioning

Targets multimedia applications byusing the MPEG-4 scheme

Delay packet lossrate

ACFEC MACnetwork(UDP)

Meant to enhance the quality ofvideo streaming over 80211WLANs

and overcome packet losses

Adjusts redundancy rates toovercome channel fluctuations and

detect and reduce packet lossPacket loss rate

MHDMwTS mdash

Meant to provide reliable datatransfer with the multipath routingand load-balancing congestioncontrol method in WMSNs

More reliable than basic routingschemes for transport multimedia

data

Latency packagetransmit rate

BCFL mdash

Introduces dispersion into fuzzylogic-based routing and sets every

cross-layer parameter with adynamic weight

Can be used to select a CH in cluster-based routing protocols andproposes a dispersion formula

Node utilitydispersion

CLODDatalinknetworktransport

Prolongs network lifetime andachieves congestion control anddesigned for lightly loaded WSNs

Assumes fixed link capacity andintegrates compressed sensing

technology

-roughputaverage energy CS

error ratio

Journal of Computer Networks and Communications 29

the purposefulness of the protocols at the networklayer CLD offers solutions to conserving powerminimizing energy and controlling both the flowand congestion in the network as well as fault tol-erance so it is an opportunity for designers con-sidering other layers We further desire to developCLDs that shall make an impact on network oper-ation and get close attention [18]

(iii) CLD interface standardization the architecturemust deliver module functionality although thereexist some queries amongst modules about possibleinterfaces predicted by the necessity to share in-formation among remote protocol layers Technicalchallenges include developing designing andstandardizing of cross-layer interfaces and algo-rithms that satisfy the cross-layer optimization re-quirements among protocol layers

Table 4 summarizes these cross-layer models with a fewcomments

5 Future Research Direction

51 At theMACLayer In WMSNs achieving duty cycling istricky since video traffic is volatile Listening to adjacentnodes wastes energy We need to switch states from awake tosleep -is requires further research

Self-organizing networks and green communicationalthough most CRWSN energy-saving algorithms aredesigned to reduce node transmission power energyconsumption is inevitable during the operational mode ofBSs because of internal processing Scholars are trying withpolicy-makers to encourage a shift to greenWSNs to reduceoperational costs and carbon footprints With self-orga-nizing CRWSNs for automatic switch-off when idle energyis saved which calls for cognitive switching algorithms Sointegrating cognitive switching with radio resource allo-cation in CRWSNs is a promising research area [78]

52 At the Application Layer Considering a WMSN ar-chitecture end users access the network via the BS to andfrom which they send their sensed data We desire to getguaranteed QoS and energy efficiency by designing a newarchitecture in the form of hardware and software whichneeds to be investigated further for a fix of the same [1]

QoS and energy efficiency are very crucial in WMSNsand mostly real-time apps that require guaranteed band-width and throughput in their network lifetime Mostprotocols ignore base station (BS) mobility and WMSNnodes Traffic management telemedicine and battlefieldsurveillance apps require mobile nodes or BSs so designingdynamic routing protocols to be adaptable in these cir-cumstances is necessary

53 Cross-Layer Security of data transmitted on WMSNseg in military surveillance and e-commerce is key withQoS and energy efficiency WMSNs are vulnerable to attacks

like worms and sinkhole Moreover their computationpower is not that high complicating implementation ofstrong secure protocols -is needs further attention

For fairness and priority issues in CRWSNs nodes areassigned diverse priorities depending on importance andurgency of their data For better network performance thereis a need to develop fair resource allocation mechanismswith priority-based fairness in sensors-ere is still potentialin this research area [78]

We must develop multiobjective adaptive protocols tooptimize QoS metrics involved in routing for best trans-mission results to get a trade-off between different opti-mization metrics We need to design standardized cross-layer algorithms which are able to meet cross-layer opti-mization standards

6 Conclusion

We have reviewed the different QoS strategies for WSNs inthe context of IoT from the MAC layer and application layeras well as the cross-layer paradigm For the MAC layer wehave reviewed protocols for WSNs such as the hybrid of thecontention-free and contention-based MAC protocols andthose for WMSNs We further reviewed a number of ap-plication layer protocols including many machine-to-ma-chine requestresponse and publishsubscribe protocols Forsystem optimization cross-layer QoS strategies are veryimportant in wireless communication We have reviewed anumber of cross-layer strategies For all categories reviewedchallenges and opportunities are discussed Finally somepossible future directions are discussed for research andapplication showing a promising potential for future re-search in this relatively new area especially as more WMSNapplications emerge recently

Conflicts of Interest

-e authors hereby declare no conflicts of interest regardingthe publication of this work

Acknowledgments

-is research was funded by the National Natural ScienceFoundation of China (Nos 61602398 61672447 and61711540306)

References

[1] V Bhandary A Malik and S Kumar ldquoRouting in wirelessmultimedia sensor networks a survey of existing protocolsand open research issuesrdquo Journal of Engineering vol 2016Article ID 9608757 27 pages 2016

[2] I F Akyildiz T Melodia and K R Chowdhury ldquoWirelessmultimedia sensor networks applications and testbedsrdquoProceedings of the IEEE vol 96 no 10 pp 1588ndash1605 2008

[3] M Iqbal M Naeem A Ahmed M Awais A Anpalagan andA Ahmad ldquoSwarm intelligence based resource managementfor cooperative cognitive radio network in smart hospitalsrdquoWireless Personal Communications vol 98 no 1 pp 571ndash5922018

30 Journal of Computer Networks and Communications

[4] S A Alvi B Afzal G A Shah L Atzori and W MahmoodldquoInternet of multimedia things vision and challengesrdquo AdHoc Networks vol 33 pp 87ndash111 2015

[5] Q F Hassan A U R Khan and S A Madani Internet ofings Challenges Advances and Applications CRC PressTaylor amp Francis Group A Chapman amp Hall Group BocaRaton FL USA 2018

[6] Q Wang Y Zhao W Wang et al ldquoMultimedia IoT systemsand applicationsrdquo in Proceedings of the 2017 Global Internet ofings Summit (GIoTS) vol 2 Geneva Switzerland June 2017

[7] A Aslam and E Curry ldquoTowards a generalized approach fordeep neural network based event processing for the internet ofmultimedia thingsrdquo IEEE Access vol 6 pp 25573ndash25587 2018

[8] M Jridi T Chapel V Dorez L Bougeant and A Le BotlanldquoSoC-based edge computing gateway in the context of theinternet of multimedia things experimental platformrdquoJournal of Low Power Electronics and Applications vol 8no 1 p 1 2018

[9] H Noura A Chehab L Sleem andM N Rapha ldquoOne roundcipher algorithm for multimedia IoT devicesrdquo MultimediaTools and Applications vol 77 no 14 pp 18383ndash18413 2018

[10] T Balan D Robu and F Sandu ldquoMultihoming for mobileinternet of multimedia thingsrdquo Mobile Information Systemsvol 2017 Article ID 6965028 16 pages 2017

[11] H Shen and G Bai ldquoRouting in wireless multimedia sensornetworks a survey and challenges aheadrdquo Journal of Networkand Computer Applications vol 71 pp 30ndash49 2016

[12] A Ahmad M H Rehmani H Tembine O A Mohammedand A Jamalipour ldquoIEEE access special section Editorial optimization for emerging wireless Networks IoT 5G andsmart grid communication networksrdquo IEEE Access vol 5pp 2096ndash2100 2017

[13] B A Muzakkari M A Mohamed M F A Kadir andZ Mohamad ldquoRecent advances in energy efficient-QoS awareMAC protocols for wireless sensor networksrdquo InternationalJournal of Advanced Computer Research vol 8 no 38pp 212ndash228 2018

[14] M A Yigitel O D Incel and C Ersoy ldquoQoS-aware MACprotocols for wireless sensor networks a surveyrdquo ComputerNetworks vol 55 no 8 pp 1982ndash2004 2011

[15] A M Abbas and O Kure ldquoQuality of Service in mobile ad hocnetworks a surveyrdquo International Journal of Ad Hoc andUbiquitous Computing vol 6 no 2 p 75 2010

[16] M Q Shatnawi ldquoApplication layer protocols for the internetof things a surveyrdquo in Proceedings of the InternationalConference on Engineering amp MIS (ICEMIS) Agadir Mo-rocco September 2016

[17] V Karagiannis P Chatzimisios F Vazquez-gallego andJ Alonso-zarate ldquoA survey on application layer protocols forthe internet of things research motivationrdquo InternationalJournal of Advanced Research in Computer Science vol 8no 3 pp 1ndash10 2015

[18] M S Bernard T Pei Z Li and K Li ldquoQoS strategies forwireless multimedia sensor networks in the context of IoTrdquoLecture Notes of the Institute for Computer Sciences SocialInformatics and Telecommunications Engineering vol 4pp 228ndash253 2019

[19] A AlAmri and M Abdullah ldquoCross-layer quality of serviceprotocols for wireless multimedia sensor networksrdquo in Pro-ceedings of the International Conference on CommunicationManagement and Information Technology pp 649ndash658Cosenza Italy April 2017

[20] J Guo L Sun and R Wang ldquoA cross-layer and multipathbased video transmission scheme for wireless multimedia

sensor networksrdquo Journal of Networks vol 7 no 9pp 1334ndash1340 2012

[21] A Ksentini A Naimi and M Gueroui ldquoToward an im-provement of H264 video transmission over IEEE 80211ethrough a cross-layer architecturerdquo IEEE CommunicationsMagazine vol 44 no 1 pp 107ndash114 2006

[22] W Ye J Heidemann and D Estrin ldquoAn energy-efficientMAC protocol for wireless sensor networksrdquo in Proceedings ofthe 21st International Annual Joint Conference of the IEEEComputer and Communications Societies (INFOCOM 2002)New York NY USA June 2002

[23] J Polastre J Hill and D Culler ldquoVersatile low power mediaaccess for wireless sensor networks categories and subjectdescriptorsrdquo in Proceedings of the Second InternationalConference on Embedded Networked Sensor Systems Sen-Sysrsquo04 pp 95ndash107 Baltimore MA USA November 2004

[24] T Danmanee K N Nakorn and K Rojviboonchai ldquoCU-MAC a duty-cycle MAC protocol for internet of things inwireless sensor networksrdquo ECTI Transactions on ElectricalEngineering Electronics and Communications vol 16 no 2pp 30ndash43 2018

[25] T Van Dam and K Langendoen ldquoAn adaptive energy-efficientMAC protocol for wireless sensor networksrdquo in Proceedings ofthe 1st International Conference on Embedded NetworkedSensor Systems Los Angeles CA USA November 2003

[26] P Xie and J Cui ldquoR-MAC an energy-efficient MAC protocolfor underwater sensor networksrdquo in International Conferenceon Wireless Algorithms Systems and Applications (WASA2007) Chicago IL USA August 2007

[27] A Rahim N Javaid M Aslam U Qasim and Z A KhanldquoAdaptive-reliable medium access control protocol forwireless body area networksrdquo in Proceedings of the AnnualIEEE Communications Society Conference on Sensor Meshand Ad Hoc Communications and Networks (SECON) SeoulKorea June 2012

[28] H S Kwak S Ullah D H Kwak and C H Lee ldquoA powerefficient MAC protocol for wireless body area networksrdquoJournal of the Korean Institute of Intelligent Systems vol 8no 6 pp 131ndash140 2009

[29] C Omeni A J Burdett and M Park ldquoEnergy efficientmedium access protocol for wireless medical body area sensornetworksrdquo in Proceedings of the 4th IEEEEMBS InternationalSummer School and Symposium on Medical Devices andBiosensors pp 29ndash32 Cambridge UK August 2007

[30] L Sitanayah C J Sreenan and K N Brown ldquoER-MAC ahybrid MAC protocol for emergency response wireless sensornetworksrdquo in Proceedings of the Fourth International Con-ference on Sensor Technologies and Applications pp 244ndash249Mestre Italy July 2010

[31] Y Sun and D B Johnson ldquoRI-MAC a receiver-initiatedasynchronous duty cycle MAC protocol for dynamic trafficloads in wireless sensor networksrdquo in Proceedings of the 6thACM conference on Embedded network sensorsystemsmdashSenSysrsquo08 pp 1ndash14 Raleigh NC USA November2008

[32] M Buettner G V Yee E Anderson and R Han ldquoX-MAC ashort preamble MAC protocol for duty-cycled wireless sensornetworksrdquo in Proceedings of the 4th International Conferenceon Embedded Networked Sensor Systems SenSys 2006pp 307ndash320 Boulder Colorado USA October 2006

[33] S Boulfekhar and M Benmohammed ldquoSynchronous receiverinitiated MAC protocol for long-lived sensor networksrdquoComputers amp Electrical Engineering vol 40 no 2 pp 504ndash516 2014

Journal of Computer Networks and Communications 31

[34] S Mehta and K S Kwak ldquoH-MAC a hybrid Mac protocol forwireless sensor networksrdquo International Journal of ComputerNetworks amp Communications vol 2 no 2 pp 108ndash117 2010

[35] I Rhee A Warrier M Aia J Min and M L Sichitiu ldquoZ-MAC a hybrid MAC for wireless sensor networksrdquo IEEEACM Transactions on Networking vol 16 no 3 pp 511ndash5242008

[36] S Arshad A Al-sadi and A Barnawi ldquoZ-MAC performanceevaluation and enhancementsrdquo Procedia Computer Sciencevol 21 pp 485ndash490 2013

[37] Y Liu I Elhanany and H Qi ldquoAn energy-efficient QoS-aware media access control protocol for wireless sensornetworksrdquo in Proceedings of the IEEE International Confer-ence on Mobile ad hoc and Sensor Systems pp 9ndash11 Wash-ington DC USA December 2005

[38] J Decotignie ldquoWiseMAC an ultra low power MAC protocolfor the downlink of infrastructure wireless sensor networksrdquoin Proceedings of the Ninth International Symposium onComputers And Communications pp 244ndash251 AlexandriaEgypt June 2004

[39] R A Rashid ldquoDevelopment of energy aware TDMA-basedMAC protocol for wireless sensor network systemrdquo EuropeanJournal of Scientific Research vol 30 no 4 pp 571ndash578 2009

[40] H P Van Hoesel ldquoA lightweight medium access protocol(LMAC) for wireless sensor networksrdquo in Proceedings of theInternational Workshop on Networked Sensing Systems Bal-timore MD USA November 2004

[41] C Li P Wang H Chen andM Guizani ldquoA cluster based on-demand multi-channel MAC protocol for wireless multi-media sensor networksrdquo in Proceedings of the IEEE In-ternational Conference on Communications pp 2371ndash2376Beijing China May 2008

[42] J L Sobrinho and A S Krishnakumar ldquoQuality-of-service inad hoc carrier sense multiple access wireless networksrdquo IEEEJournal on Selected Areas in Communications vol 17 no 8pp 1353ndash1368 1999

[43] J P John ldquoProviding MAC QoS for multimedia traffic in80211e based multi-hop ad hoc wireless networksrdquo ComputerNetworks vol 51 no 1 pp 153ndash176 2007

[44] S Lofty and N Padmavati ldquoA survey on mobility basedprotocols in WSNsrdquo in Proceedings of the InternationalConference Communication and Manufacturing pp 12ndash15Hangzhou China July 2014

[45] S Deng L Shen and J Li ldquoMobility-based clustering pro-tocol for wireless sensor networks with mobile nodesrdquo IETWireless Sensor Systems vol 1 no 1 pp 39ndash47 2011

[46] M Khalil A Khalid F U Khan and A Shabbir ldquoA review ofrouting protocol selection for wireless sensor networks insmart citiesrdquo in Proceedings of the 24th Asia-Pacific Confer-ence on Communications (APCC) Ningbo China November2014

[47] O Message Q Telemetry and T Mqtt Committee Specifi-cation Draft 02 April 2014

[48] M Collina M Bartolucci A Vanelli-coralli andG E Corazza ldquoInternet of things application layer protocolanalysis over error and delay prone linksrdquo in Proceedings ofthe 7th Advanced Satellite Multimedia Systems Conference andthe 13th Signal Processing for Space CommunicationsWorkshop Livorno Italy September 2014

[49] A Talaminos-barroso M A Estudillo-valderrama L M RoaJ Reina-tosina and F Ortega-ruiz ldquoA machine-to-machineprotocol benchmark for e health applicationsmdashuse case re-spiratory rehabilitationrdquo Computer Methods and Programs inBiomedicine vol 129 pp 1ndash11 2016

[50] V Goyal O Pandey A Sahai and B Waters ldquoAttribute-based encryption for fine-grained access control of encrypteddatardquo in Proceedings of the 13th ACM Conference on Com-puter and Communications Security pp 89ndash98 AlexandriaVA USA October 2006

[51] M Singh M A Rajan V L Shivraj and P BalamuralidharldquoSecure MQTT for internet of things (IoT)rdquo in Proceedings ofthe Fifth International Conference on Communication Systemsand Network Technologies Gwalior India April 2015

[52] A Stanford-Clark and H L Truong MQTT for Sensor Net-works (MQTT-SN) Protocol Specification httpmqttorgnewwp-contentuploads200906MQTT-SN_spec_v12pdf 2013

[53] RFC ldquoExtensible messaging and presence protocol (XMPP)rdquoin Proceedings of the IETF pp 1ndash9 San Diego CA USAAugust 2005

[54] S Vinoski and I Technologies ldquoQueuing protocolrdquo IEEEInternet Computing vol 10 no 6 pp 87ndash89 2006

[55] J L Fernandes I C Lopes J J P C Rodrigues and S UllahldquoPerformance evaluation of restful web services and AMQPprotocolrdquo in Proceedings of the Fifth International Conferenceon Ubiquitous and Future Networks (ICUFN) pp 2ndash7 DaNang Vietnam July 2013

[56] M Asim ldquoA survey on application layer protocols for internetof things (IoT)rdquo International Journal of Advanced Researchin Computer Science vol 8 no 3 pp 996ndash1000 2017

[57] K An A Gokhale D Schmidt S Tambe P Pazandak andG Pardo-castellote ldquoContent-based filtering discovery pro-tocol (CFDP) scalable and efficient OMG DDS discoveryprotocolrdquo in Proceedings of the 8th ACM InternationalConference on Distributed Event-Based Systems New YorkNY USA July 2014

[58] E Web What is a web service pp 52ndash57 2010[59] M Laine RESTful Web Services for the Internet of ings

pp 2ndash4 2012 httpmediaTkk[60] IETF -e WebSocket protocol pp 1ndash71 2011[61] V Pimentel ldquoCommunicating and displaying real-time data

with websocketrdquo IEEE Internet Computing vol 16 no 4pp 45ndash53 2012

[62] Wikipedia 2019 httpsenwikipediaorgwikiStreaming_Text_Oriented_Messaging_Protocol2019

[63] K Raghavendra and S Nireshwalya ldquoApplication layer se-curity issues and its solutionsrdquo IJCSET vol 2 no 6pp 1266ndash1269 2012

[64] Y Xue B Ramamurthy andM C Vuran ldquoSDRCS a service-differentiated real-time communication scheme for eventsensing in wireless sensor networksrdquo Computer Networksvol 55 no 15 pp 3287ndash3302 2011

[65] C Sonmez S Isik M Y Donmez O D Incel and C ErsoyldquoSUIT a cross layer image transport protocol with fuzzy logicbased congestion control for wireless multimedia sensornetworksrdquo in Proceedings of the 5th International Conferenceon New Technologies Mobility and Security (NTMS) pp 1ndash6IEEE Istanbul Turkey May 2012

[66] T Melodia and I Akyildiz ldquoCross-layer QoS-aware com-munication for ultra wide band wireless multimedia sensornetworksrdquo IEEE Journal on Selected Areas in Communica-tions vol 28 no 5 pp 653ndash663 2010

[67] Q Wang and M A Abu-Rgheff ldquoCross-layer signalling fornext-generation wireless systemsrdquo in Proceedings of the 2003IEEE Wireless Communications and Networking 2003WCNC 2003 vol 2 pp 1084ndash1089 IEEE New Orleans LAUSA March 2003

[68] P Liu Z Tao S Narayanan T Korakis and S PanwarldquoCoopMAC a cooperative MAC for wireless LANsrdquo IEEE

32 Journal of Computer Networks and Communications

Journal on Selected Areas in Communications vol 25 no 2pp 340ndash354 2007

[69] H Shan H T Cheng and W Zhuang ldquoCross-layer co-operative MAC protocol in distributed wireless networksrdquoIEEE Transactions onWireless Communications vol 10 no 8pp 2603ndash2615 2011

[70] F Liu T Korakis Z Tao and S Panwar ldquoAMAC-PHY cross-layer protocol for wireless Ad-Hoc networksrdquo in Proceedingsof the 2008 IEEE Wireless Communications and NetworkingConference pp 1792ndash1797 IEEE Las Vegas NV USA March2008

[71] L Jacob and H R Shamna ldquoEfficient cooperative MAC androuting in wireless networksrdquo Transactions on Networks andCommunications vol 3 no 5 2015

[72] S Rao and K Shama ldquoCross layer protocols for multimediatransmission in wireless networksrdquo International Journal ofComputer Science amp Engineering Survey vol 3 no 3pp 15ndash28 2012

[73] L Han S Park S Kang and H Peter ldquoAn adaptive cross-layer FEC mechanism for video transmission over 80211WLANsrdquo in Proceedings of the 2009 International Conferenceon Internet pp 209ndash215 Las Vegas NV USA December2009

[74] N Li J Martınez and V H Dıaz ldquo-e balanced cross-layerdesign routing algorithm in wireless sensor networks usingfuzzy logicrdquo Sensors vol 15 no 8 pp 19541ndash19559 2015

[75] G Sun J Qi Z Zang and Q Xu ldquoA reliable multipathrouting algorithm with related congestion control scheme inwireless multimedia sensor networksrdquo in Proceedings of the2011 3rd International Conference on Computer Research andDevelopment vol 4 pp 229ndash233 IEEE Shanghai ChinaMarch 2011

[76] M Li Y Jing and C Li ldquoA robust and efficient cross-layeroptimal design in wireless sensor networksrdquoWireless PersonalCommunications vol 72 no 4 pp 1889ndash1902 2013

[77] J Yan M Zhou and Z Ding ldquoRecent advances in energy-efficient routing protocols for wireless sensor networks areviewrdquo IEEE Access vol 4 pp 5673ndash5686 2016

[78] A Ahmad S Ahmad M H Rehmani and N U Hassan ldquoAsurvey on radio resource allocation in cognitive radio sensornetworksrdquo IEEE Communications Surveys amp Tutorials vol 17no 2 pp 888ndash917 2015

Journal of Computer Networks and Communications 33

International Journal of

AerospaceEngineeringHindawiwwwhindawicom Volume 2018

RoboticsJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Active and Passive Electronic Components

VLSI Design

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Shock and Vibration

Hindawiwwwhindawicom Volume 2018

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawiwwwhindawicom

Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Control Scienceand Engineering

Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Journal ofEngineeringVolume 2018

SensorsJournal of

Hindawiwwwhindawicom Volume 2018

International Journal of

RotatingMachinery

Hindawiwwwhindawicom Volume 2018

Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Navigation and Observation

International Journal of

Hindawi

wwwhindawicom Volume 2018

Advances in

Multimedia

Submit your manuscripts atwwwhindawicom

Page 17: ReviewArticle - downloads.hindawi.com · ReviewArticle QoSStrategiesforWirelessMultimediaSensorNetworksinthe ContextofIoTattheMACLayer,ApplicationLayer,and Cross-LayerAlgorithms MuwongeSsajjabbiBernard

(ii) Nonconfirmable message here no ACK messagesare sent back

(iii) Acknowledgment (ACK) message this confirmsreceipt of a confirmable message

(iv) Reset message this confirms receipt of an un-processed message in case any critical parts nec-essary for interpreting the message are missing

(v) Piggybacked response the receiver directly re-sponds while getting the ACK message

(vi) Separate response the reply by the receiver is dis-crete from the ACK message

CoAP has no in-built security features It is secured bythe datagram transport layer security (DTLS) that runs ontop of UDP though it was not designed for the IoT -ismakes its suitability debatable since DTLS offers no supportfor multicasting It also needs more packets for handshakingthereby increasing the traffic volume and computationalresources and so shortening the lifespan of mobile devicessince they use batteries [17]

Advantages

(i) It can support multiple users and hence yieldbetter performance via multicasting

Table 2 Comparison of MAC layer protocols used in QoS provisioning

Protocol Features QoS parameter Comments Access Priorityassignment Energy eff Designed for

S-MAC[22] Collision avoidance Energy latency Requires periodic

synchronization CSMA Hybrid Medium WSNs

B-MAC[23] Collision avoidance Energy delay

throughput Flexible interface CSMA Dynamic High WSNs

T-MAC[25] Collision avoidance Energy Complex nonscalable CSMA Dynamic High WSNs

C-MAC[13]

Dynamic channelallocation Energy Multichannel MAC CSMA Dynamic Medium WSNs

CU-MAC[24]

Multichannelmechanism

-roughputlatency High packet delivery ratio TDMA Dynamic mdash WSNs

A-MAC[39]

Per flow serviceguarantees Delay m-ary tree model

dynamic prioritiesTDMACSMA Dynamic Medium WSNs

ER-MAC[30]

Emergencyresponse

Energy delaydelivery ratio

Adapts to traffic andtopology changes

TDMACSMA Hybrid Higher than Z-

MAC WSNs

RI-MAC[31] Collision detection Delay packet

delivery ratioHigh packet delivery ratio

and throughput CSMA Dynamic Medium WSNs

SRI-MAC[33] Collision avoidance Energy delay Good data transfer and

energy consumption TDMA Dynamic High WSNs

H-MAC[34] Slotted sleep time -roughput

latencyData transmission only in

sleep timeCSMAAloha Dynamic Medium WSNs

Z-MAC[36] Collision avoidance Energy delay

throughputVery good channel usage

and error syncCSMATDMA Dynamic Medium WSNs

Q-MAC[37]

Intrainternodescheduling Energy High QoS improved

energy efficiency CSMA Hybrid High WSNs

X-MAC[32]

Overhearingavoidance

Energy delaythroughput

Reduces excessivepreamble by strobing CSMA Adaptive

static Medium WSNs

L-MAC[40] Collision avoidance Network

lifetime energyReduces the number oftransceiver switches TDMA Static Medium WSNs

WiseMAC[38]

No downlinkchannel collision Power delay Significantly low-power

consumptionAlohaCSMA Static Higher than

ZigbeeInfrastructure

WSNsAR-MAC[27] Collision avoidance Energy delay Uses data packets and

control packets TDMA Static Higher thanIEEE 802154 WBANs

BB-MAC[42]

Real-time trafficprioritization Delay Employs round robin in

real-time nodesCSMATDMA Dynamic High WMSNs

COM-MAC [41] Collision avoidance Packet delay

throughputBalances reliability and

retransmissionTDMAFDMA Dynamic High WMSNs

EQoSA[14]

Busty traffic loadaccommodation Energy Provides QoS support for

video and imagesTDMACDMA mdash High WMSNs

QE-MAC[42] Collision avoidance Energy jitter

latencyRuns central polling-like

channel access CSMA Dynamic High WMSNs

R-MAC[26] Collision avoidance Energy delay Reservation based mdash Static High Underwater

SNs

Journal of Computer Networks and Communications 17

(ii) It enables asynchronous message exchange(iii) It is a reliable and simple protocol

32 Message Queue Telemetry Transport (MQTT)Developed by IBM and later improved by OASIS with anintention of decreasing bandwidth MQTT is a simple open-source publishsubscribe protocol analogous to the client-server version It is appropriate for lightweight IoT appli-cations and M2M transmissions in constrained situationslike limited power computation capability and memory orbandwidth -e network bandwidth and the computationalresource usage are decreasing since clients are not updatedmaking publishsubscribe protocols outperform requestresponse protocols via IoT requirements [16] -is protocolcan run over TCPIP and so has the following advantages

(i) Delivering of packets with reliable guarantee(ii) Enabling multicasting(iii) Remote device session establishment(iv) Reduction of transport overhead and protocol ex-

change to minimize network traffic(v) Provision of the notification mechanism when

anomalies occur [47]

MQTTrsquos design targets economical use of bandwidthand battery so it has a major application in FacebookMessenger MQTT ensures QoS and reliability of messagedelivery by providing the following three QoS levels

(1) Fire and forget a message is sent once depending onthe networkrsquos best effort and there is no need forACK (lowest QoS level)

(2) Delivered at least once a message is sent at least onceand an ACK message is needed to avoid duplicates

(3) Delivered exactly once by a four-way handshake itensures a message is sent exactly one time (highestQoS level)-e architecture it has three key components (i)publishersmdashlightweight nodes sending data to thebroker before returning to sleep (ii) broker whichorganizes sensory data in topics and sends them tointerested subscribers and (iii) subscribersmdashIoTapps interested in data sent by sensors through thebroker as shown in Figure 3

MQTT is a simple lightweight protocol that is betterthan CoAP via low and high packet loss probability interms of throughput and exhibits lower latency in lowsampling rates [48] It supports a number of imple-mentations for embedded devices and mobile applicationsand offers web-socket support too Nevertheless it hassome shortcomings

(i) It is only employed in simple data types(ii) It exhibits high latency in high sampling rates(iii) It is unsuitable for real-time apps in such cases(iv) It does not provide automatic discovery andor

sufficient security at the protocol level [49]

33 Secure Message Queue Telemetry Transport (SMQTT)-is is an extension of MQTT that is based on lightweightattribute-based encryption using elliptic curve cryptogra-phy -e protocol has three entities as shown in Figure 4(i) publisher that publishes information under a certaintopic (ii) subscriber that obtains the same topic in-formation via a broker and (iii) public key generator(PKG) or broker that acts as the trusted third party -eprotocol has four phases namely (a) set-up phase (b)encrypt phase (c) publish phase and (d) decrypt phase

(a) Set-up phase publisher and subscriber provide aunique ID with attributes for registration with thePKG -is generates a secret master key set andpublic parameters that it publishes with set U A1A2 An For ciphertext policy-attribute-basedencryption (CP-ABE) the PKG gives a key set todevices with attributes [50]

(b) Encrypt phase based on the access tree the publishercreates an access policy An access tree is sent by thepublisher to the PKG for generating the key-gen-eration policy in the key policy- (KP-) based ABEalgorithm -e publishers know beforehand thetopics and subscribers to access as well as the rulesfor access Subscribers are issued with private keysfor all communication sessions Publishers in theCP-ABE algorithm come up with access trees andrules where the payload is encrypted by the pub-lisher who also issues more info needed for de-cryption and the rules With the help of public keycryptography data are encrypted giving rise to ci-phertext on the basis of KPCP-ABE and the receivergets SUBACK from the broker [51]

(c) Publish phase ciphertext is embedded in theSPublish command as payload and topic name isdefined in the header by the publisher -e brokerreceives the SPublish packet and replies withPUBACK prompting the publisher to respond withthe PUBREL packet -e broker then broadcasts totopic subscribers and deletes the data forwarding thePUBCOMP packet to the sender node

(d) Decrypt phase the encrypted information isdecrypted by the subscriber with the help of a secretkey Considering KP-ABE on verification that itsatisfies the access policy the subscriber invokes thePKG for a corresponding key which it sends afterverifying the request With CP-ABE subscribers usethe private key for decryption after authenticationthereby enabling offline interaction [51]Advantages

(i) -e information being encrypted makes it moresecure

34 MQTT for Sensor Networks (MQTT-SN) -is is a ver-sion of MQTT designed to operate under environments oflow bandwidth high link failures and short message lengthIt is a publishsubscribe (pubsub) protocol designed for

18 Journal of Computer Networks and Communications

UDP-based WSNs to visualize communication for low-power devices [5152] It comprises a publisher device actingas the MQTT-SN client to forward communications to agateway that also changes the MQTT-SN message to theMQTTmessage before delivering it to the broker that sendsit to MQTT-SN subscribers [51]

It has a gateway for changing protocols from MQTT toMQTT-SN (Figure 5) [52] -is protocol supports nodes in

the sleeping mode using an offline keep-alive procedure thatcan be used by battery-operated devices to get into thesleeping mode To support sleeping clients MQTT-SN has anew offline keep-alive procedure with which battery-oper-ated devices can go to the sleeping state in which theirmessages are buffered at the server until they are awake InMQTT-SN clients with no preconfigured gateway addresscan figure out the real address of their operational gateway

Publisher

Subscriber

Subscriber

Broker

Publisher

4G

Figure 3 Architecture of the MQTT protocol

Register by mentioning URI Register by mentioning URI

Access tree

Set-up

Publisher ReceiverPKG(broker)

Key management Key management

Encrypt data using public parametersand random key This is done using

CP-ABE Publish it as

Subscribe (topics)SUBACK

S-Publish (TopicEncrypt (data))S-Publish (TopicEncrypt (data))

Encryptand publish

PUBACK

PUBACK

PUBRELPUBRELPUBCOMP

PUBCOMP

Figure 4 SMQTT protocol

Journal of Computer Networks and Communications 19

within the network through a discovery procedure Simul-taneously we may have several gateways in a single wirelessnetwork that can be cooperative via load sharing

MQTT-SN architecture it is composed of MQTT-SN clients gateways and forwarders (Figure 5)rough the MQTT-SN gateway (GW) the MQTT-SN protocol is used by the MQTT-SN clients toconnect to the MQTT server Integration betweenthe MQTT-SN GW and the MQTT-SN server mightbe possible or not Translating between the MQTT-SN and the MQTT for a stand-alone GW is doneusing the MQTT-SN protocol If a GW is detachedfrom the network it can be accessed by the MQTT-SN clients through a forwarder is just encapsu-lates the received MQTT-SN frames that are on thewireless side before forwarding them to the GW un-altered It then decapsulates the received frames tooresending them to the client as sent by the GW [52]ere are 2 types of gateways (a) transparent gatewayand (b) aggregating gateway as shown in Figure 6

(i) Transparent gateway this launches and sustains adirect MQTT connection between the MQTT-SNclient and the MQTT server which is entirely re-served for end-to-end client-server messaging enumber of servers and that of MQTT-SN clientsconnected to the GW are the same is transparent

GW translates syntax between these protocols and iseasier to implement compared to the aggregatingGW But the MQTT server must support dierentsessions for all connected clients Some MQTTservers are congured to enable few parallel sessions

(ii) Aggregating gateway this maintains a single MQTTconnection to the server on behalf of all clients Allexchanged messages from the MQTT-SN client to theaggregating GW are directed to the GW which sortsthe information to forward to the server e aggre-gating GW decreases the number of MQTT sessionssimultaneously supported by the server in WSNs withmany battery-operated sensors and actuators (SAs)having limited storage and processing capabilities It ishowever more complex to implement compared to atransparent GWe general format of the MQTT-SNprotocol message has two or four message headeroctets and n octets of variable header that is optionaldepending on the message in question MQTT-SN has256 message types (MsgType) with those from0X1Eminus 0XFD 0X19 and 0XFF reserved for futureusage [51]

Advantages

(i) e protocol can operate under environmentsof low bandwidth high link failures and shortmessage length

MQTT-SN gateway

MQTT-SN

MQTT

MQTT-SN

MQTT-SN

MQTT-SN

MQTT-SN client

MQTT-SN clients

MQTT broker

MQTT-SN client

MQTT-SN client

MQTT-SN gateway

MQTT-SN forwarder

Figure 5 MQTT-SN architecture

20 Journal of Computer Networks and Communications

(ii) It is a pubsub protocol designed for UDP-basedWSNs to visualize communication for low-power devices

(iii) It can be optimal for low-cost battery-operatednodes having low processing power and lessstorage

(iv) It supports nodes in the sleeping mode using theobrvbarine keep-alive procedure used by battery-operated devices

35e Extensible Messaging and Presence Protocol (XMPP)e standardization of this protocol by the IETF made itpopular for usage over the Internet especially in instantmessaging and chatting platforms It is however susceptibleto spamming attacks and its failure to have worldwidesupport has made Google too withdraw its support of thestandard [17] e protocol is mainly useable in the client-server architecture but can also be applicable in client-clientXMPP communication [53] Among the application layerprotocols XMPP is the only one that supports both asyn-chronous (publishsubscribe) and synchronous (requestresponse) services while running over TCP but can still runover other protocols

Given the simplicity of its implementation and ease ofusability in XMPP instant messaging (IM) is suitable forXMPP and is implemented using the message stanzaltmessagegt ese stanzas are usable in one-to-one ormultiuser sessions or chatting or in error reporting andsending headlines Attribute type of the stanza denes themessage type e attribute type could be a chat group chatheadline normal or error An instant messaging network inXMPP can support many XMPP clients and servers with agateway congured to enable other IM network users toconnect to other XMPP users e XMPP server mainlycontrols sessions and XML streams to clients and servers inaddition to routing XML stanzas in the streams [53] eTCP ports 5269 and 5222 have been respectively registeredfor the XMPP server-to-server connections and XMPP clientconnections by the Internet Assigned Numbers Authority(IANA) XMPP enables specication of XMPP ExtensionProtocols (XEP) to enhance its usefulness [17]

XMPP security clients connect to the server usingtransport layer security (TLS) and simple authentication andsecurity layer (SASL) protocols which are at the same timeused by the servers for interdomain communications Fordata condentiality and integrity TLS will encrypt thestreams of XML for all sessions To conrm the identity of anyclient that needs to access the server the SASL protocol doesthe authentication of the XML stream For any communi-cating session with a server to start the client needs to resolvethe serverrsquos DNS hostname For protection of the server of theclient from third-party attacks the server maintains the au-thentication details secret ie IP address and access methodand only original connections of the server are required

In server-to-server communications the SASL protocolis employed for authentication and condentiality Server-to-server communications may as well be deactivated from aserver by an administrator depending on the organizationalpolicy To guard against domain-spoong attacks serverdialback can be a good solution since servers support itsecuring more the XML stanzas Nevertheless it cannot beused to authenticate secure or encrypt in-server streamsand the server identication resulting therein is not strongenough to be relied on It further cannot secure against DNSpoisoning attacks and IP session hijacking for remote do-mains [53] High-security domains can use TLS and SASL

Advantages

(i) It is popular for usage over the Internet especially ininstant messaging and chatting platforms

Disadvantages

(i) It is susceptible to spamming attacks(ii) It has failed to have worldwide support making

Google withdraw support of the standard(iii) It is not secure against DNS poisoning attacks and

IP session hijacking for remote domains

36 Advanced Message Queuing Protocol (AMQP) isprotocol can be utilized by a number of platforms formessage exchange since it is open source and written inmany languages It is a publishsubscribe model depending

MQTT-SNclients Broker

MQTT

(a)

MQTT-SNclients Broker

(b)

Figure 6 (a) Transparent gateway and (b) aggregating gateway

Journal of Computer Networks and Communications 21

on a reliable and efficient messaging queue with capacity forinstant messaging or e-mail for different applications es-pecially in IoT environments It differs from others in thesense that message types can be specified and have theirsource traceable and how to do trade-off between perfor-mance security and reliability For reliability of messagetransmission AMQPmakes use of some delivery guaranteeslike at most once at least once and exactly once It can alsohowever utilize the TCP transport layer for guaranteeingreliability -e AMQPrsquos publishsubscribe method has twocomponents namely (i) exchange queue meant for routingof messages to an appropriate order in the queue and (ii)message queue that acts as storage pending their delivery toreceivers [16] In this protocol model the broker can makerouting decisions contrary to othermessaging systems whereapps using the queue have in them embedded the decision-making logic for recipients or senders -e developer mayneed to check each of the affected apps so as to change thelogic for routing and delivery of the messages [54]

Advantages

(i) It is convenient(ii) It has capacity for interoperability between dif-

ferent vendors(iii) It can use open standards to enable connection

between business partners(iv) It can enable innovations based on AMQP foun-

dations [55](v) It can enable supporting of mission-critical ap-

plications in e-commerce(vi) It can guarantee reliable delivery of messages

Disadvantages

(i) It is not very appropriate for real-timeapplications

(ii) It may not offer automatic discovery(iii) Interoperability is not guaranteed(iv) It lacks open-source libraries for constrained

gadgets [49]

37 Data Distribution Service (DDS) A product of theObject Management Group (OMG) DDS is a pubsubprotocol meant for M2M communications with two sub-layers -ese are the data-centric pubsub sublayer re-sponsible for connecting anonymous data publishers tosubscribers and the data-local reconstruction sublayer that isoptional and integrates DDS into the application layer [56]-e DDS architecture connects system components asshown in Figure 7 Publishers and subscribers are discon-nected with respect to (i) time (ii) space (iii) flow andbehavior (iv) platforms and (v) programming languages[57] A DDS data publisher generates topics and publisher-subscriber coupling is represented as topic name data typeschema and publishers and subscribersrsquo QoS attributesDelivering of data is done by the publisher layer -e datawriter and publishers together make a decision about anynecessary alterations for subsequent forwarding to thesubscribers who need to send data to IoT applications Data

readers are responsible for delivering topics to the sub-scribers after reading through them [56]

Some of the major components of the DDS architectureinclude the following [57]

(i) Topic this is a logical path specifying publicationand subscriptionrsquos data type between DataWritersand DataReaders For a successful session to begintopic names types and DataWriters and Data-Readersrsquo QoS should correspond

(ii) Domain this is a logical setting for transmissionthat can be utilized for separating and optimizingnetwork transmission in a setup of similar appli-cations If DDS apps share a domain ID then theycan exchange data among themselves

(iii) DataWriter and DataReader these are terminalobjects for writing and reading messages re-spectively to and from a universal data space

(iv) Participant this is an object that stands for apublisher or subscriber of the DDS app in a do-main acting like a container for other objects

Advantages

(i) It is highly reliable and offers QoS support(ii) It is highly scalable(iii) It is appropriate for real-time apps(iv) It is fault tolerant(v) It has high level of interoperability(vi) It does not need brokers that may act as

bottlenecks(vii) It offers automatic discovery(viii) It has a user-defined data structure for topic(ix) It supports publisher and subscriber decoupling

Disadvantages

(i) It is originally developed for only stand-aloneLANs

(ii) A lot of memory is needed(iii) It has no open-source libraries to support con-

strained devices(iv) It is hard to design and configure compared to

other protocols [49]

38 RESTful Services REpresentational State Transfer (REST)-is is an architectural style based on a set of principles thatdefine and address networked resources It offers web servicesthat enable transmission from the HTTP device to the devicewithin the IoT architecture and enables clients to reach theserver [16]-is protocol utilizes HTTP commands to backupthe requestresponse messaging model HTTP is a popularand secure protocol in the WWW using TLSSSL [16] ARESTful application has the following characteristics

(i) State and functionality are separated into distrib-uted resources

(ii) All resources are distinctively named by the use ofunique HTTP commands (GET POST PUT orDELETE over the Internet)

22 Journal of Computer Networks and Communications

(iii) It is a clientserver stateless layered protocol thatsupports caching REST uses the same methods asHTTP for all requests or responses with POST andGET for creating and retrieving resources PUT forupdating and changing the state of resources andDELETE for removing resourcesAdvantages of REST over arbitrary web services(SOAP) in the IoT context

(i) It reduces overhead(ii) It reduces parsing complexity(iii) It is stateless(iii) It integrates more tightly with HTTP [58](iv) It is easy to implement and learnis together

with its natural t over HTTP makes it apreferred method for Web 20 applications toshowcase their data and secure M2M sessionsin the IoT

(v) Applications that support RESTful web ser-vices exhibit better performance than the re-source-constrained WSN nodes [59]

(vi) REST is important for the IoT since it supportscommercial M2M cloud platforms

(vii) Its implementation in smart phones and tabletapps is also easy since it needs only an HTTPlibrary that is accessible to all operating systemplatforms [56]

Disadvantages

(i) Although it enjoys wide usage in commercialM2M platforms it is not likely to dominatebecause it is hard to implement

(ii) Its HTTP usage makes it incompatible withconstrained M2M devices leaving its usage toend applications

39 WebSockets It is a web-based protocol operating on asingle TCP channel and providing full-duplex transmissions

It permits secure duplex communication between a clientexecuting unreliable code in a restricted environment and adistant node communicating from that code while using theorigin-based security model employed by web browsers overTCP [60] e protocol is not a requestresponse model andis not a publishsubscribe model So unlike other protocolsa web-socket session is established by the client initiating ahandshake to the server to launch the session is onceestablished a full-duplex client-to-server connection beginsasynchronously and runs until terminated by either party[56] It gives a model for browser-based apps utilizing bi-directional communication with no need of many HTTPconnections to servers

WebSocket comprises two parts ie (i) the handshakehaving a clientrsquos message and the serverrsquos handshake responseand (ii) the data transfer HTTP headers should be repeated byapplications in all client requests and server responses arisingfrom the endless polling which might increase the commu-nication overhead subject to the application [61] WebSocketcan be used to come up with scalable real-time web appli-cations since it can enable full-duplex transmission via onesocket across the Web Its security can be ensured with thehelp of TLSSSL is protocol is suitable for resource-con-strained devices and IoTapplications will not have a good userexperience with it since it operates a clientserver architectureNevertheless it can outcompete all TCP-based protocolsbecause of the following advantages

(i) It is t for real-time transmission(ii) It has improved security(iii) It minimizes transmission overhead(iv) It can provide ecient messaging systems using

WebSocket ApplicationMessaging Protocol (WAMP)[17]

310 Streaming (Simple) Text-Oriented Messaging Protocol(STOMP) is is a TCP-based protocol designed for textmessage-oriented middleware (MOM) with the use of

Topic

Datareader

Datareader

Datareader

Datawriter

Datawriter

Datawriter

Subscriber PublisherPublisher

Data bus (DDS domain)

Subscriber

Topic Topic

Figure 7 DDS architecture

Journal of Computer Networks and Communications 23

HTTP-like commands for interoperability among platformslanguages and brokers Data communication takes placebetween the client and the broker in line with commands(like CONNECT DISCONNECT ACK NACK SUB-SCRIBE UNSUBSCRIBE SEND BEGIN COMMIT orABORT) then headers in the form ltkeygt ltvaluegt (one perline) then a blank line and then body content terminatingin a null character Server-client communication takes placeover a MESSAGE RECEIPT or ERROR frame with headerand body content format being analogous MOM productssupporting STOMP include Apache ActiveMQ or FuseMessage Broker HornetQ Open Message Queue(OpenMQ) RabbitMQ and syslog-ng through its STOMPdestination plugin [62] In addition to these other appli-cation layer protocols include simple media control protocol(SMCP) lightweight local automation protocol (LLAP)simple sensor interface (SSI) lightweight M2M (LWM2M)XMPP-IOT and simple object access protocol (SOAP)

Advantages

(i) It is lightweight for constrained networks(ii) It has flexibility to choose quality of services with

the given functionality(iii) It is standardized by OASIS Technical Committee(iv) It is easy and quick to implement

Challenges and opportunities

Here security threats are imminent because of the open-endedness of the application layer [63] -ese could includethe following

(i) -ere is poor or lack of a security design for anapplicationrsquos function Coming up with a gooddesign to address this could be a good ideaopportunity

(ii) Some of the programs might have backdoorsevading otherwise secure controls to allow un-authorized access to network resources We requiremechanisms to patch up such loopholes

(iii) Applications that lack or have weak authenticationmechanisms can be easily targeted by unlicensedusers to abuse the network Such mechanisms needto be strengthened further

(iv) Excessively complicated access control rules maybecause of their misunderstanding or poor writinggive unauthorized access allowing dubious activityrather than protecting the network they purport toprotect -ese should be simple and clear to un-derstand by all users

(v) In booting workstations without disks and managingnetwork devices we usually use the TFTP protocolalthough access does not require authenticating theuser by usernamepassword -is makes an intruderable to gain easy access to configaccess info as longas he can guess the filenames -is needs to be fixed

A few steps can be used to secure the application layerTo protect network data transmitted across the network

applications need to strengthen their encryption and au-thentication mechanisms Apps too need to enable strictcontrols of privileges for data access using good mechanismsto balance between usability and effectiveness [63] Appli-cations handling sensitive data need detailed logging andaudit capability as well as testing and reviewing

Table 3 summarizes the application layer protocols dis-cussed with their weaknesses and differences among them

4 Cross-Layer QoS Strategies

41 Service-Differentiated Real-Time Communication Scheme(SDRCS) -is is an event-driven routing protocol that routesreal-time traffic by featuring a cross-layer packet-forwardingdesign in which it integrates the real-time routing functionalitywith a new priority-based MAC scheme -e protocol ap-proximates distributed packet traversal speed for traffic clas-sification and admission control on the basis of this design Itfurther localizes decision-making by prioritizing packet for-warding to maximize the speed of packets [64]

Advantages

(i) It improves bandwidth utilization bandwidthdegradation due to unschedulable packets can beavoided

(ii) It offers a lightweight packet schedulability estima-tion mechanism Here it uses received signalstrength and admission control and then earlymissed-deadline packet-dropping policies are made

(iii) It delivers high end-to-end throughput via highersource data rates alongside strict end-to-end latencyneeds

(iv) Nodes which meet real-time requirements of lesstraffic load and better channel quality are givenhighest priority when it comes to packet forwarding

-e SDRCS operates via 5 mechanisms namely (1) RSS-(Rich Site SummaryReally Simple Syndication-) basedgrouping (2) admission control (3) prioritized queuing(4) real-time MAC and (5) dynamic forwarding -isprotocol has a queuing policy that employs per-hopdeadline-based queues and the nodes schedule theirpackets using FIFO priority-based queues [1] -e prin-cipal features of this protocol are mentioned as follows

(i) It is event driven and is easily adaptable to anynetwork changes

(ii) It does not need additional hardware for locali-zation or multichannel transmission

(iii) -e cross-layer packet-forwarding strategy can beused when transmitting multimedia traffic

(iv) It is able to efficiently circumvent any voids therein

Early deadline miss (EDM) policy drops all unscheduledpackets -e SDRCS circumvents wrong packet dropsdue to EDM since it adjusts to network changes easily

42 Sensor Fuzzy-Based Image Transport (SUIT) -is pro-tocol regulates congestion using a scheme based on fuzzy

24 Journal of Computer Networks and Communications

logic It transmits packets with reduced quality to allowablelevels if there is congestion enabling it to transmit lower-quality packetsframes without being dropped-e numberof packets delivered per second thus rises and while thisenhances video streaming apps it is more appropriatelymade for those that quickly transmit non-real-time videosuch as in video surveillance SUIT can stream JPEG pic-tures other than the traditional predictive video codingmethods that are not suitable for sensor devices because oftheir constraints of energy memory and computationalspeed Progressive JPEG-PJPEG performs better via packettransmission rate and delay though it degrades the imagequality because of congestion [65] -is protocol employscross-layer methods for interlayer data exchange though itstill employs some layering making it still possible topreserve all advantages of the layered design since it hasapplication and transport layers but has noMAC or routinglayer

For congestion detection and control SUIT bases on thepercentage number of in-bound and out-bound frames perwindow number of contenders and buffer occupancy of thenext-hop node Since buffer management has an effect oneffectiveness and QoS inWMSNs it is useful in the transportlayer so the protocol prioritizes packets for better QoS SUITreorders packets by tracking the sequence numbers of thedelivered packets since they all have a source ID framenumber and sequence number that it uses to detect anymissing packets -e protocol however does not providepacket reliability [65]

Advantages

(i) -e protocol prioritizes packets for better QoS(ii) It can transmit lower-quality packetsframes

without being dropped

Disadvantages

(i) It does not provide packet reliability

43 Network Layer QoS Support Enforced by a Cross-LayerController (NLQS) -is scheme permits packet-level servicedifferentiation as a function of throughput end-to-endpacket error rate and delay [66] (see Figure 8) -is en-hances network layer QoS It comprises a cross-layer controlunit (XLCU) for configuring and controlling networkingfunctionalities at physical MAC and network layers [18]-is depends upon the unified logic which affects choices forapplication layer needs and status of functional blocks whichdo implementation of networking functions [66]

Advantages

(i) Cross-layer interactions can be controlled withoutweakening the upgradability modularity andsimplicity of designing the system

44 Cross-Layer Signaling Shortcuts (CLASS) CLASS has ahigh level of effectiveness flexibility and comprehensive-ness According to [67] it has the following features

(i) Direct signaling among remote layers leading tofaster transmission

(ii) Lightweighted internal message format standard-ized protocols are not suitable for use in internalsignaling since they are not usually lightweight forinstance transmitting against faults in the networkCLASS requires only three fields namely destina-tion address event type and event contents

(iii) -e setup of its exterior messages is standardconsidering external signaling ICMP and TCPIPare suitable for general messages and short notifi-cations respectively

(iv) Message control protocol should ensure optimizedand organized exchange of dense simultaneousmessages across layers for high efficiency andavoidance of possible conflicts -e protocol appliesto many cross-layer signaling situations

Table 3 Differences and problems with application layer protocols

Protocol Architecture Transport QoS options Security Weaknesses

CoAP Requestresponse UDP DTLS No in-built security features DTLS does not supportmulticast

MQTT Publishsubscribe TCP TLSSSL Insufficient security at the protocol level sincepayload values are not encrypted

SMQTT Publishsubscribe TCP CPKP-ABE Key revocation and group pubsub for distributedSMQTT are still a challenge

MQTT-SN Publishsubscribe UDP TLSSSL SSLTLS suffers from attacks like BEAST CRIMERC4 and Heartbleed

XMPP RequestresponsePublishsubscribe TCP X TLSSSL Susceptible to spamming attacks and lacks worldwide

supportAMQP Publishsubscribe TCP TLSSSL Inappropriate for real-time applications

DDS Publishsubscribe TCPUDP TLSSSL Memory intensive and no open-source libraries forconstrained devices

REST Requestresponse HTTP X HTTPS Hard to implement Uses HTTP so incompatiblewith constrained apps

WebSockets PublishsubscribeClientserver TCP X TLSSSL Bad user experience for IoT apps since it runs the

clientserver architectureSTOMP Clientserver TCP HTTP -e broker can act as a bottleneck

Journal of Computer Networks and Communications 25

Advantages

(i) e protocol is scalable and has a very lowpropagation delay

(ii) It is very ecient and exible since it uses directsignaling between any two layers

(iii) It is very comprehensive and eective

45 Cross-Layer and Multipath-Based Video Transmission(CMVT) CMVT is a hybrid of application and networklayers At the application layer it carries out the encoding ofvideo streams to video data frames (I- P- and B-frames) bymeans of the MPEG-4 encoding format At the networklayer route discovery and data transfer take place Twoalgorithms greedy forwarding and rollback are used todiscover a number of paths from the source to the sink nodevia the route discovery method [20] A specied node icalculates the evaluation of its adjacent node j using thefollowing equation

fij (1 minus α)d2(j D) minus d2min(i)d2max(i) minus d

2min(i)

+ αeinit(j) minus eres(j)

einit(j)

(19)

where fij evaluation value for the node i to j d2(j D) isthe distance from the node j to the destination node D

d2min(i) and d2max(i) are the minimum and maximum dis-

tances of neighbors of the node i to D respectively einit(j) isthe initial energy of the node j eres(j) is the current residualenergy of the node j and α is the energy coecient given asfollows

α emax(i) minus emin(i)

emax(i) (20)

where emax(i) and emin(i) are respectively the maximumand minimum energy left of all node irsquos neighbors

e network layer further transmits video streams withCMVT doing status evaluation choosing an appropriatecommunication path for any packets and the QoS guaranteelevel for the path i is calculated as

fi (1 minus ω)hisum hi

+ ωnisum ni

(21)

where fi is the evaluation value for the path i hi are the hopsfor the path i ni is the summation of packets sent via isum ni isthe summation of packets sent by sources and ω is theenergy consumption factor [18]

Advantages

(i) e CMVT protocol is very superior in mediatransmission particularly large-scale WMSNs

To network peers

Flow requirements

-

Atimes

Admissioncontrol

Capacityassignment

Next-hopselection

Coding rate

Data rate

QoScontracts

Cross-layer controlunit (XLCU)

Localization subsystem

Synchronizationsubsystem

Networkcoordinationunit (NCU)

UWB modulator

Channel coder

QoS scheduler

Georouting

Source coder

Rate control

Figure 8 Cross-layer controller architecture

26 Journal of Computer Networks and Communications

46 Cross-Layer Cooperative MAC (CoopMAC)CoopMAC [68] is based on the IEEE 80211 DCF approach-ere is a reduction in interference of neighboring cellsleading to uniform coverage in densely deployed networks[68] A connection is established by two-way authenticationbetween the sender and the receiver When a Clear-to-Send(CTS) packet and short interframe space (SIFS) are obtainedby the sender packets are transmitted right to the receiverwith no valuable cooperation If available the two nodes findout whether they can exchange data -is is done by the useof a helper identification signal to establish how feasiblecooperative communication is It is established in thepresence of a signal and if it is not available direct com-munication is activated For distributed systems helper-initiated cooperation is preferable because of the presence ofthe RTSCTS packets [69]

Advantages

(i) CoopMAC has an advantage over others due to thespatial diversity amongst the three nodes [70]

(ii) It has the potential to achieve significantthroughput and delay with no complexity in thedesign of the system

47 Cooperative MAC Protocol for Multihop Networks (M-CMAC) Just as CoopMAC M-CMAC is made in such away that high data rate stations help low data rate stations toforward traffic for broadcasting [71] -ere are also helperschosen in such a way that 2 fast-hop transmissions replace aslow hop transmission -e helper via whom there existsleast delay from the sender to the receiver is considered bestbecause of theminimum two-hop transmission rate and so isthe chosen neighbor to the duo Here the working as-sumption is that all nodes have their position coordinatesknown and thus the Euclidean distances between all pairscan be calculated and converted to the data rate for that linkAll nodes have cooperative tables (CT) of potential helperswith destination and helper MAC address Euclidean dis-tance and total distance through the helper Any sender withdata to transmit checks out for the existence of any helper inthe CT for the destination and if available forwards RTS forchannel reservation for single-hop duration M-CMAC hasgot an RTS format having a shape with five fields similar tothe one given as follows

Framecontrol Duration Source

addressDestination

addressHelperaddress

-e destination nodersquos address is kept by the source in thepart for the helper address that is in turn saved as the desti-nation address Nodes inspect helper and destination addressfields on receipt of RTS whereby the node acts as the othernodersquos helper in case the helper address field is not the sameand the helper transmits CTS back to the source if at all it wantsto forward data On delivery of the CTS packet the sourcetransmits to the helper which in turn sends the packets to thedestination that also sends ACK to the helper on receipt of thepacket In M-CMAC there exists a higher level of channelreuse (parallel transmissions) due to the increased number of

nodes therein that subsequently raises availability of helpers fordata forwarding -is results in increased throughput incomparison with CoopMAC and IEEE 80211 DCF [71]

Advantages

(i) It promises a higher throughput compared toCoopMAC and IEEE 80211 DCF

48 Cluster-Based Cooperative Routing (CBCR) Protocol-e CBCR protocol has a multihop data-forwarding func-tion realized at the link layer with cooperative links that useM-CMAC -is protocol encompasses two stages namelythe routing relay selection phase and data forwarding phase

(i) Routing relay selection phase all nodes announcetheir presence to their neighbors by broadcastingperiodical beacon messages that carry the MAC ad-dress of the node Each of these constructs a relaytable that contains all neighboring nodes with whichit is able to communicate -e node further broad-casts its neighbor list in case of any change to itsentries since the previous broadcast -e MAC ad-dresses of nodes next to the node X are contained inits relay tablersquos column one and the neighbor nodersquosrow has MAC addresses for neighbors of the adjacentnode Based on its relay table each node in-dependently chooses routing relays Choosing a nodeas a relay node depends on the number of nodes itconnectsmdashit should be the highest number [71]

(ii) Data forwarding phase a node with packets totransmit needs to first verify if the receiver is in thesame cluster and if that receiver has a helper to whichthe packet is thence forwarded but in whose absencethe packet is delivered right to the receiver In casethe intended receiver is in a different cluster therelay table will be checked to see whether that re-ceiver is reachable via other routing relays If it isreachable packets will be sent to the routing relaydirectly or via the helper if at all the relay has one Incase of destination unreachable via relays packetsare multicast to all relays by the node [71]Advantages

(i) It enables multicasting

49 MAC-PHY Cross-Layer Protocol A cooperative cross-layer standard for cooperation at the physical layer in next-generation WMSNs is developed in [70] It provides acomplete MAC layer algorithm that gives a supportive shellfor the PHY-MAC layer Like CoopMAC its scheme de-pends on the middle node for communication betweennodes-eMAC layer protocol has been changed to regulateinformation exchanges at the physical layer -e receivernode receives duplicate packets from source and helpernodes to decode the data [70]

410 MAC-Centric Approach -is cross-layer protocoltargets multimedia applications by using the MPEG-4

Journal of Computer Networks and Communications 27

scheme [21] It is characterized by 4 access categories(AC3ndashAC0) according to their priority in transmission It ismeant for supporting different QoS needs in upcomingvideo applications and enabling MAC layer differentiationfor H264 partitioning [18] Because of low bandwidth delayand other QoS challenges that cause inefficiency in trans-mittingmultimedia data inWMSNs a number of algorithmsdepending on IEEE 80211e have been proposed to supporttransmission of quality videos [72] Selecting an AC dependson the measures for QoS such as the loss rate and delay-us AC3 having the highest priority is mapped to theparameter set concept since the stream is sensitive to bit lossespecially in video transmission [21]

411 Adaptive Cross-Layer Forward Error Correction(ACFEC) In the ACFEC [73] model data packets are ex-changed among nodes through the access point (AP) whichoperates the infrastructure mode in which it adds FEC to videodata [72] -ese data are dealt by encapsulation using astreaming server to the receiver through thewireless AP as RTPpackets An adaptive FEC controller senses the category ofpackets out of the RTP header and recovers the header ofpackets from UDP -e encoder generates some error-cor-recting packets whose number is determined by the sourcepacket number of the block Multimedia transmissions aremonitored by the controller using MAC failure data and itscounter is increased by one if there is a failed transmission-econtroller uses the failure counter to change the packet numberproduced after transmitting a block [18] In case there are lostpackets it changes the redundancy rates and produces extrapackets to replace the lost ones and satisfy the requirements ofthe receiving node [72] -e FEC packet number is enhancedor reduced to satisfy the receiverrsquos requirements and stoppacket losses-is is done through accurate detection of packetlosses and adjustment of redundancy rates If all video datapackets are well received there will not be generation of FECpackets [73]

Advantages

(i) It promises a good QoS through packet loss re-duction and redundancy rate adjustment

412 Balanced Cross-Layer Fuzzy Logic (BCFL) DesignRouting Algorithm in WSNs A new fuzzy logic-basedrouting algorithm (BCFL) [74] was designed using disper-sion of the cross-layer parameter as the fuzzy logic inferencesystem input Each cross-layer parameter has a dynamicweight depending on the value of dispersion -e designcomes with some innovations as follows (i) for fuzzy logicinference system input the absolute parameter value issubstituted by parameter dispersion thereby significantlyreducing algorithmic complexity (ii) dispersion does notchange with the order of magnitude according to the dis-persion formula and (iii) the weight of the parameter de-pends on the size of its dispersion and the two are inverselyproportional to each other -is enables BCFL to have somenobility unlike other algorithms as per the following dis-tinguishing properties making it advantageous

(i) It has got simple if-then rules which remain con-stant even when the constraints increase

(ii) It is capable of dealing with many constraints withno increment in complexity

(iii) It has capacity to yield a more balanced solutionthan other algorithms

(iv) It is easily adaptable to changes in network con-ditions and topology even when the changes arefrequent like in underwater WSNs -e algorithm isuseable in choosing the CH in cluster-based routingprotocols [74]

413 Minimum Hop Disjoint Multipath Routing Algorithmwith Time Slice Load-Balancing Congestion Control SchemeMHDMwTS is a two-phased routing protocol comprisingpath build-up and path acknowledgment phases It consists ofmultiple sources each of which has three build-up disjointpaths namely primary alternate and back-up paths Onactivation the source node requests to build up a route to thenearest hop neighbor which is the path build-up phase [19] Inthis phase step one has source activation in which therequested node for path building adds its number and time-stamp and sends to the least hop-count neighboring node-isgoes on until the least time latency sink having the informationneeded to construct the primary route is reached [75] In steptwo there is path extraction when the new package fromanother path arrives whereby the extracted path is comparedwith the primary path If the node is shared the package will berejected or an alternative path will be searched for to get abackup through comparison of the preceding two paths Phasetwo is path acknowledgment where we have the third step inwhich the sink returns the ACK packet to the sender with pathinformation having nodes and their time information after ithas computed it using the timestamp [1819]

Advantages

(i) -e protocol minimizes end-to-end latency(ii) It enhances congestion control

414 Cross-Layer Optimal Design (CLOD) Authors in [76]come up with a CLOD for scheduling at the data link layerrouting at the network layer and controlling congestion atthe transport layer with an assumption of fixed link capacity-rough congestion control energy efficiency is improvedTransport layer congestion at the nodes is minimized bycompressed sensing (CS) in which transmitted bits are re-duced whereas optimally allocating resources decreasescongestion on the links at the data link layer CLODpromises minimized computational complexity and betterperformance in light traffic scenarios [77] Generating andstoring CLOD is more efficient compared to Gaussianrandom matrices It prolongs network lifetime and savesenergy

Advantages

(i) It does congestion control subsequently increasingenergy efficiency

28 Journal of Computer Networks and Communications

(ii) It reduces computational complexity(iii) It improves performance in light traffic scenarios

415 Challenges and Opportunities of the Cross-Layer Design(CLD)

(i) Different layers in a real network assume differentfunctions andor services A layer only communi-cates with its neighboring layers -e layered modeldegrades system performance because of many

features in wireless transmission QoS parameters inthe layers of WMSNs are shared by routing pro-tocols to optimize the performance Neverthelessthere is a need to design cross-layer models forincreased efficiency in routing [1]

(ii) -e physical layer plays a highly vital role in CLDRate adaptation and channel allocation take place atthe physical layer via signal processing to enhanceQoS End-to-end performance in wireless media isaffected by the changes therein given the effect on

Table 4 A summary of some of the reviewed cross-layer models

Protocol Layers Aims Comments QoS parameters

SDRCS MACPHY Routing real-time traffic Transmits multimedia traffic viacross-layer packet forwarding

-roughputlatency

SUIT Applicationtransport Transmission of non-real-time video Uses fuzzy logic to regulate

congestionTransmission rate

delay

CLASS Any twoDesign serves as a framework fordifferent implementations ofdifferent application scenarios

Scalable very efficient flexible andhas very low propagation delay

Propagation delayjitter

NLQS PHYMACnetwork

Permits packet-level servicedifferentiation as a function of

throughput packet error rate anddelay

Network layer QoS is enhanced -roughput delaypacket error rate

CMVT Applicationnetwork

Encodes video streams to video dataframes via MPEG-4 encoding anddoes route discovery and data

transfer

Greedy forwarding and rollback areused to find source-to-sink paths Energy

CoopMAC PHYMAC Offers spatial diversity among thethree nodes

Helper ID signal finds out howfeasible cooperative communication

is

-roughput ratedelay

M-CMAC PHYMAC Increases end-to-end throughputand packet delivery ratio

Euclidean distances between nodesare calculated and converted to the

data rate for a link

-roughputpacket delivery

ratio

CBCR PHYMACMinimizes control overhead andtime consumed in establishing thecooperative paths than M-CMAC

Energy consumption is moreuniformly distributed in a network

enhancing network lifetime

-roughputpacket deliveryratio energy

MAC-PHY MACPHYMeant to enable PHY layer

cooperation and maximize gains ofcooperation at the MAC layer

Leverages both spatial diversity andcoding gain -roughput delay

MAC-centric MACAPPL

Meant to support QoS needs in newvideo apps and enable MAC layer

differentiation for H264partitioning

Targets multimedia applications byusing the MPEG-4 scheme

Delay packet lossrate

ACFEC MACnetwork(UDP)

Meant to enhance the quality ofvideo streaming over 80211WLANs

and overcome packet losses

Adjusts redundancy rates toovercome channel fluctuations and

detect and reduce packet lossPacket loss rate

MHDMwTS mdash

Meant to provide reliable datatransfer with the multipath routingand load-balancing congestioncontrol method in WMSNs

More reliable than basic routingschemes for transport multimedia

data

Latency packagetransmit rate

BCFL mdash

Introduces dispersion into fuzzylogic-based routing and sets every

cross-layer parameter with adynamic weight

Can be used to select a CH in cluster-based routing protocols andproposes a dispersion formula

Node utilitydispersion

CLODDatalinknetworktransport

Prolongs network lifetime andachieves congestion control anddesigned for lightly loaded WSNs

Assumes fixed link capacity andintegrates compressed sensing

technology

-roughputaverage energy CS

error ratio

Journal of Computer Networks and Communications 29

the purposefulness of the protocols at the networklayer CLD offers solutions to conserving powerminimizing energy and controlling both the flowand congestion in the network as well as fault tol-erance so it is an opportunity for designers con-sidering other layers We further desire to developCLDs that shall make an impact on network oper-ation and get close attention [18]

(iii) CLD interface standardization the architecturemust deliver module functionality although thereexist some queries amongst modules about possibleinterfaces predicted by the necessity to share in-formation among remote protocol layers Technicalchallenges include developing designing andstandardizing of cross-layer interfaces and algo-rithms that satisfy the cross-layer optimization re-quirements among protocol layers

Table 4 summarizes these cross-layer models with a fewcomments

5 Future Research Direction

51 At theMACLayer In WMSNs achieving duty cycling istricky since video traffic is volatile Listening to adjacentnodes wastes energy We need to switch states from awake tosleep -is requires further research

Self-organizing networks and green communicationalthough most CRWSN energy-saving algorithms aredesigned to reduce node transmission power energyconsumption is inevitable during the operational mode ofBSs because of internal processing Scholars are trying withpolicy-makers to encourage a shift to greenWSNs to reduceoperational costs and carbon footprints With self-orga-nizing CRWSNs for automatic switch-off when idle energyis saved which calls for cognitive switching algorithms Sointegrating cognitive switching with radio resource allo-cation in CRWSNs is a promising research area [78]

52 At the Application Layer Considering a WMSN ar-chitecture end users access the network via the BS to andfrom which they send their sensed data We desire to getguaranteed QoS and energy efficiency by designing a newarchitecture in the form of hardware and software whichneeds to be investigated further for a fix of the same [1]

QoS and energy efficiency are very crucial in WMSNsand mostly real-time apps that require guaranteed band-width and throughput in their network lifetime Mostprotocols ignore base station (BS) mobility and WMSNnodes Traffic management telemedicine and battlefieldsurveillance apps require mobile nodes or BSs so designingdynamic routing protocols to be adaptable in these cir-cumstances is necessary

53 Cross-Layer Security of data transmitted on WMSNseg in military surveillance and e-commerce is key withQoS and energy efficiency WMSNs are vulnerable to attacks

like worms and sinkhole Moreover their computationpower is not that high complicating implementation ofstrong secure protocols -is needs further attention

For fairness and priority issues in CRWSNs nodes areassigned diverse priorities depending on importance andurgency of their data For better network performance thereis a need to develop fair resource allocation mechanismswith priority-based fairness in sensors-ere is still potentialin this research area [78]

We must develop multiobjective adaptive protocols tooptimize QoS metrics involved in routing for best trans-mission results to get a trade-off between different opti-mization metrics We need to design standardized cross-layer algorithms which are able to meet cross-layer opti-mization standards

6 Conclusion

We have reviewed the different QoS strategies for WSNs inthe context of IoT from the MAC layer and application layeras well as the cross-layer paradigm For the MAC layer wehave reviewed protocols for WSNs such as the hybrid of thecontention-free and contention-based MAC protocols andthose for WMSNs We further reviewed a number of ap-plication layer protocols including many machine-to-ma-chine requestresponse and publishsubscribe protocols Forsystem optimization cross-layer QoS strategies are veryimportant in wireless communication We have reviewed anumber of cross-layer strategies For all categories reviewedchallenges and opportunities are discussed Finally somepossible future directions are discussed for research andapplication showing a promising potential for future re-search in this relatively new area especially as more WMSNapplications emerge recently

Conflicts of Interest

-e authors hereby declare no conflicts of interest regardingthe publication of this work

Acknowledgments

-is research was funded by the National Natural ScienceFoundation of China (Nos 61602398 61672447 and61711540306)

References

[1] V Bhandary A Malik and S Kumar ldquoRouting in wirelessmultimedia sensor networks a survey of existing protocolsand open research issuesrdquo Journal of Engineering vol 2016Article ID 9608757 27 pages 2016

[2] I F Akyildiz T Melodia and K R Chowdhury ldquoWirelessmultimedia sensor networks applications and testbedsrdquoProceedings of the IEEE vol 96 no 10 pp 1588ndash1605 2008

[3] M Iqbal M Naeem A Ahmed M Awais A Anpalagan andA Ahmad ldquoSwarm intelligence based resource managementfor cooperative cognitive radio network in smart hospitalsrdquoWireless Personal Communications vol 98 no 1 pp 571ndash5922018

30 Journal of Computer Networks and Communications

[4] S A Alvi B Afzal G A Shah L Atzori and W MahmoodldquoInternet of multimedia things vision and challengesrdquo AdHoc Networks vol 33 pp 87ndash111 2015

[5] Q F Hassan A U R Khan and S A Madani Internet ofings Challenges Advances and Applications CRC PressTaylor amp Francis Group A Chapman amp Hall Group BocaRaton FL USA 2018

[6] Q Wang Y Zhao W Wang et al ldquoMultimedia IoT systemsand applicationsrdquo in Proceedings of the 2017 Global Internet ofings Summit (GIoTS) vol 2 Geneva Switzerland June 2017

[7] A Aslam and E Curry ldquoTowards a generalized approach fordeep neural network based event processing for the internet ofmultimedia thingsrdquo IEEE Access vol 6 pp 25573ndash25587 2018

[8] M Jridi T Chapel V Dorez L Bougeant and A Le BotlanldquoSoC-based edge computing gateway in the context of theinternet of multimedia things experimental platformrdquoJournal of Low Power Electronics and Applications vol 8no 1 p 1 2018

[9] H Noura A Chehab L Sleem andM N Rapha ldquoOne roundcipher algorithm for multimedia IoT devicesrdquo MultimediaTools and Applications vol 77 no 14 pp 18383ndash18413 2018

[10] T Balan D Robu and F Sandu ldquoMultihoming for mobileinternet of multimedia thingsrdquo Mobile Information Systemsvol 2017 Article ID 6965028 16 pages 2017

[11] H Shen and G Bai ldquoRouting in wireless multimedia sensornetworks a survey and challenges aheadrdquo Journal of Networkand Computer Applications vol 71 pp 30ndash49 2016

[12] A Ahmad M H Rehmani H Tembine O A Mohammedand A Jamalipour ldquoIEEE access special section Editorial optimization for emerging wireless Networks IoT 5G andsmart grid communication networksrdquo IEEE Access vol 5pp 2096ndash2100 2017

[13] B A Muzakkari M A Mohamed M F A Kadir andZ Mohamad ldquoRecent advances in energy efficient-QoS awareMAC protocols for wireless sensor networksrdquo InternationalJournal of Advanced Computer Research vol 8 no 38pp 212ndash228 2018

[14] M A Yigitel O D Incel and C Ersoy ldquoQoS-aware MACprotocols for wireless sensor networks a surveyrdquo ComputerNetworks vol 55 no 8 pp 1982ndash2004 2011

[15] A M Abbas and O Kure ldquoQuality of Service in mobile ad hocnetworks a surveyrdquo International Journal of Ad Hoc andUbiquitous Computing vol 6 no 2 p 75 2010

[16] M Q Shatnawi ldquoApplication layer protocols for the internetof things a surveyrdquo in Proceedings of the InternationalConference on Engineering amp MIS (ICEMIS) Agadir Mo-rocco September 2016

[17] V Karagiannis P Chatzimisios F Vazquez-gallego andJ Alonso-zarate ldquoA survey on application layer protocols forthe internet of things research motivationrdquo InternationalJournal of Advanced Research in Computer Science vol 8no 3 pp 1ndash10 2015

[18] M S Bernard T Pei Z Li and K Li ldquoQoS strategies forwireless multimedia sensor networks in the context of IoTrdquoLecture Notes of the Institute for Computer Sciences SocialInformatics and Telecommunications Engineering vol 4pp 228ndash253 2019

[19] A AlAmri and M Abdullah ldquoCross-layer quality of serviceprotocols for wireless multimedia sensor networksrdquo in Pro-ceedings of the International Conference on CommunicationManagement and Information Technology pp 649ndash658Cosenza Italy April 2017

[20] J Guo L Sun and R Wang ldquoA cross-layer and multipathbased video transmission scheme for wireless multimedia

sensor networksrdquo Journal of Networks vol 7 no 9pp 1334ndash1340 2012

[21] A Ksentini A Naimi and M Gueroui ldquoToward an im-provement of H264 video transmission over IEEE 80211ethrough a cross-layer architecturerdquo IEEE CommunicationsMagazine vol 44 no 1 pp 107ndash114 2006

[22] W Ye J Heidemann and D Estrin ldquoAn energy-efficientMAC protocol for wireless sensor networksrdquo in Proceedings ofthe 21st International Annual Joint Conference of the IEEEComputer and Communications Societies (INFOCOM 2002)New York NY USA June 2002

[23] J Polastre J Hill and D Culler ldquoVersatile low power mediaaccess for wireless sensor networks categories and subjectdescriptorsrdquo in Proceedings of the Second InternationalConference on Embedded Networked Sensor Systems Sen-Sysrsquo04 pp 95ndash107 Baltimore MA USA November 2004

[24] T Danmanee K N Nakorn and K Rojviboonchai ldquoCU-MAC a duty-cycle MAC protocol for internet of things inwireless sensor networksrdquo ECTI Transactions on ElectricalEngineering Electronics and Communications vol 16 no 2pp 30ndash43 2018

[25] T Van Dam and K Langendoen ldquoAn adaptive energy-efficientMAC protocol for wireless sensor networksrdquo in Proceedings ofthe 1st International Conference on Embedded NetworkedSensor Systems Los Angeles CA USA November 2003

[26] P Xie and J Cui ldquoR-MAC an energy-efficient MAC protocolfor underwater sensor networksrdquo in International Conferenceon Wireless Algorithms Systems and Applications (WASA2007) Chicago IL USA August 2007

[27] A Rahim N Javaid M Aslam U Qasim and Z A KhanldquoAdaptive-reliable medium access control protocol forwireless body area networksrdquo in Proceedings of the AnnualIEEE Communications Society Conference on Sensor Meshand Ad Hoc Communications and Networks (SECON) SeoulKorea June 2012

[28] H S Kwak S Ullah D H Kwak and C H Lee ldquoA powerefficient MAC protocol for wireless body area networksrdquoJournal of the Korean Institute of Intelligent Systems vol 8no 6 pp 131ndash140 2009

[29] C Omeni A J Burdett and M Park ldquoEnergy efficientmedium access protocol for wireless medical body area sensornetworksrdquo in Proceedings of the 4th IEEEEMBS InternationalSummer School and Symposium on Medical Devices andBiosensors pp 29ndash32 Cambridge UK August 2007

[30] L Sitanayah C J Sreenan and K N Brown ldquoER-MAC ahybrid MAC protocol for emergency response wireless sensornetworksrdquo in Proceedings of the Fourth International Con-ference on Sensor Technologies and Applications pp 244ndash249Mestre Italy July 2010

[31] Y Sun and D B Johnson ldquoRI-MAC a receiver-initiatedasynchronous duty cycle MAC protocol for dynamic trafficloads in wireless sensor networksrdquo in Proceedings of the 6thACM conference on Embedded network sensorsystemsmdashSenSysrsquo08 pp 1ndash14 Raleigh NC USA November2008

[32] M Buettner G V Yee E Anderson and R Han ldquoX-MAC ashort preamble MAC protocol for duty-cycled wireless sensornetworksrdquo in Proceedings of the 4th International Conferenceon Embedded Networked Sensor Systems SenSys 2006pp 307ndash320 Boulder Colorado USA October 2006

[33] S Boulfekhar and M Benmohammed ldquoSynchronous receiverinitiated MAC protocol for long-lived sensor networksrdquoComputers amp Electrical Engineering vol 40 no 2 pp 504ndash516 2014

Journal of Computer Networks and Communications 31

[34] S Mehta and K S Kwak ldquoH-MAC a hybrid Mac protocol forwireless sensor networksrdquo International Journal of ComputerNetworks amp Communications vol 2 no 2 pp 108ndash117 2010

[35] I Rhee A Warrier M Aia J Min and M L Sichitiu ldquoZ-MAC a hybrid MAC for wireless sensor networksrdquo IEEEACM Transactions on Networking vol 16 no 3 pp 511ndash5242008

[36] S Arshad A Al-sadi and A Barnawi ldquoZ-MAC performanceevaluation and enhancementsrdquo Procedia Computer Sciencevol 21 pp 485ndash490 2013

[37] Y Liu I Elhanany and H Qi ldquoAn energy-efficient QoS-aware media access control protocol for wireless sensornetworksrdquo in Proceedings of the IEEE International Confer-ence on Mobile ad hoc and Sensor Systems pp 9ndash11 Wash-ington DC USA December 2005

[38] J Decotignie ldquoWiseMAC an ultra low power MAC protocolfor the downlink of infrastructure wireless sensor networksrdquoin Proceedings of the Ninth International Symposium onComputers And Communications pp 244ndash251 AlexandriaEgypt June 2004

[39] R A Rashid ldquoDevelopment of energy aware TDMA-basedMAC protocol for wireless sensor network systemrdquo EuropeanJournal of Scientific Research vol 30 no 4 pp 571ndash578 2009

[40] H P Van Hoesel ldquoA lightweight medium access protocol(LMAC) for wireless sensor networksrdquo in Proceedings of theInternational Workshop on Networked Sensing Systems Bal-timore MD USA November 2004

[41] C Li P Wang H Chen andM Guizani ldquoA cluster based on-demand multi-channel MAC protocol for wireless multi-media sensor networksrdquo in Proceedings of the IEEE In-ternational Conference on Communications pp 2371ndash2376Beijing China May 2008

[42] J L Sobrinho and A S Krishnakumar ldquoQuality-of-service inad hoc carrier sense multiple access wireless networksrdquo IEEEJournal on Selected Areas in Communications vol 17 no 8pp 1353ndash1368 1999

[43] J P John ldquoProviding MAC QoS for multimedia traffic in80211e based multi-hop ad hoc wireless networksrdquo ComputerNetworks vol 51 no 1 pp 153ndash176 2007

[44] S Lofty and N Padmavati ldquoA survey on mobility basedprotocols in WSNsrdquo in Proceedings of the InternationalConference Communication and Manufacturing pp 12ndash15Hangzhou China July 2014

[45] S Deng L Shen and J Li ldquoMobility-based clustering pro-tocol for wireless sensor networks with mobile nodesrdquo IETWireless Sensor Systems vol 1 no 1 pp 39ndash47 2011

[46] M Khalil A Khalid F U Khan and A Shabbir ldquoA review ofrouting protocol selection for wireless sensor networks insmart citiesrdquo in Proceedings of the 24th Asia-Pacific Confer-ence on Communications (APCC) Ningbo China November2014

[47] O Message Q Telemetry and T Mqtt Committee Specifi-cation Draft 02 April 2014

[48] M Collina M Bartolucci A Vanelli-coralli andG E Corazza ldquoInternet of things application layer protocolanalysis over error and delay prone linksrdquo in Proceedings ofthe 7th Advanced Satellite Multimedia Systems Conference andthe 13th Signal Processing for Space CommunicationsWorkshop Livorno Italy September 2014

[49] A Talaminos-barroso M A Estudillo-valderrama L M RoaJ Reina-tosina and F Ortega-ruiz ldquoA machine-to-machineprotocol benchmark for e health applicationsmdashuse case re-spiratory rehabilitationrdquo Computer Methods and Programs inBiomedicine vol 129 pp 1ndash11 2016

[50] V Goyal O Pandey A Sahai and B Waters ldquoAttribute-based encryption for fine-grained access control of encrypteddatardquo in Proceedings of the 13th ACM Conference on Com-puter and Communications Security pp 89ndash98 AlexandriaVA USA October 2006

[51] M Singh M A Rajan V L Shivraj and P BalamuralidharldquoSecure MQTT for internet of things (IoT)rdquo in Proceedings ofthe Fifth International Conference on Communication Systemsand Network Technologies Gwalior India April 2015

[52] A Stanford-Clark and H L Truong MQTT for Sensor Net-works (MQTT-SN) Protocol Specification httpmqttorgnewwp-contentuploads200906MQTT-SN_spec_v12pdf 2013

[53] RFC ldquoExtensible messaging and presence protocol (XMPP)rdquoin Proceedings of the IETF pp 1ndash9 San Diego CA USAAugust 2005

[54] S Vinoski and I Technologies ldquoQueuing protocolrdquo IEEEInternet Computing vol 10 no 6 pp 87ndash89 2006

[55] J L Fernandes I C Lopes J J P C Rodrigues and S UllahldquoPerformance evaluation of restful web services and AMQPprotocolrdquo in Proceedings of the Fifth International Conferenceon Ubiquitous and Future Networks (ICUFN) pp 2ndash7 DaNang Vietnam July 2013

[56] M Asim ldquoA survey on application layer protocols for internetof things (IoT)rdquo International Journal of Advanced Researchin Computer Science vol 8 no 3 pp 996ndash1000 2017

[57] K An A Gokhale D Schmidt S Tambe P Pazandak andG Pardo-castellote ldquoContent-based filtering discovery pro-tocol (CFDP) scalable and efficient OMG DDS discoveryprotocolrdquo in Proceedings of the 8th ACM InternationalConference on Distributed Event-Based Systems New YorkNY USA July 2014

[58] E Web What is a web service pp 52ndash57 2010[59] M Laine RESTful Web Services for the Internet of ings

pp 2ndash4 2012 httpmediaTkk[60] IETF -e WebSocket protocol pp 1ndash71 2011[61] V Pimentel ldquoCommunicating and displaying real-time data

with websocketrdquo IEEE Internet Computing vol 16 no 4pp 45ndash53 2012

[62] Wikipedia 2019 httpsenwikipediaorgwikiStreaming_Text_Oriented_Messaging_Protocol2019

[63] K Raghavendra and S Nireshwalya ldquoApplication layer se-curity issues and its solutionsrdquo IJCSET vol 2 no 6pp 1266ndash1269 2012

[64] Y Xue B Ramamurthy andM C Vuran ldquoSDRCS a service-differentiated real-time communication scheme for eventsensing in wireless sensor networksrdquo Computer Networksvol 55 no 15 pp 3287ndash3302 2011

[65] C Sonmez S Isik M Y Donmez O D Incel and C ErsoyldquoSUIT a cross layer image transport protocol with fuzzy logicbased congestion control for wireless multimedia sensornetworksrdquo in Proceedings of the 5th International Conferenceon New Technologies Mobility and Security (NTMS) pp 1ndash6IEEE Istanbul Turkey May 2012

[66] T Melodia and I Akyildiz ldquoCross-layer QoS-aware com-munication for ultra wide band wireless multimedia sensornetworksrdquo IEEE Journal on Selected Areas in Communica-tions vol 28 no 5 pp 653ndash663 2010

[67] Q Wang and M A Abu-Rgheff ldquoCross-layer signalling fornext-generation wireless systemsrdquo in Proceedings of the 2003IEEE Wireless Communications and Networking 2003WCNC 2003 vol 2 pp 1084ndash1089 IEEE New Orleans LAUSA March 2003

[68] P Liu Z Tao S Narayanan T Korakis and S PanwarldquoCoopMAC a cooperative MAC for wireless LANsrdquo IEEE

32 Journal of Computer Networks and Communications

Journal on Selected Areas in Communications vol 25 no 2pp 340ndash354 2007

[69] H Shan H T Cheng and W Zhuang ldquoCross-layer co-operative MAC protocol in distributed wireless networksrdquoIEEE Transactions onWireless Communications vol 10 no 8pp 2603ndash2615 2011

[70] F Liu T Korakis Z Tao and S Panwar ldquoAMAC-PHY cross-layer protocol for wireless Ad-Hoc networksrdquo in Proceedingsof the 2008 IEEE Wireless Communications and NetworkingConference pp 1792ndash1797 IEEE Las Vegas NV USA March2008

[71] L Jacob and H R Shamna ldquoEfficient cooperative MAC androuting in wireless networksrdquo Transactions on Networks andCommunications vol 3 no 5 2015

[72] S Rao and K Shama ldquoCross layer protocols for multimediatransmission in wireless networksrdquo International Journal ofComputer Science amp Engineering Survey vol 3 no 3pp 15ndash28 2012

[73] L Han S Park S Kang and H Peter ldquoAn adaptive cross-layer FEC mechanism for video transmission over 80211WLANsrdquo in Proceedings of the 2009 International Conferenceon Internet pp 209ndash215 Las Vegas NV USA December2009

[74] N Li J Martınez and V H Dıaz ldquo-e balanced cross-layerdesign routing algorithm in wireless sensor networks usingfuzzy logicrdquo Sensors vol 15 no 8 pp 19541ndash19559 2015

[75] G Sun J Qi Z Zang and Q Xu ldquoA reliable multipathrouting algorithm with related congestion control scheme inwireless multimedia sensor networksrdquo in Proceedings of the2011 3rd International Conference on Computer Research andDevelopment vol 4 pp 229ndash233 IEEE Shanghai ChinaMarch 2011

[76] M Li Y Jing and C Li ldquoA robust and efficient cross-layeroptimal design in wireless sensor networksrdquoWireless PersonalCommunications vol 72 no 4 pp 1889ndash1902 2013

[77] J Yan M Zhou and Z Ding ldquoRecent advances in energy-efficient routing protocols for wireless sensor networks areviewrdquo IEEE Access vol 4 pp 5673ndash5686 2016

[78] A Ahmad S Ahmad M H Rehmani and N U Hassan ldquoAsurvey on radio resource allocation in cognitive radio sensornetworksrdquo IEEE Communications Surveys amp Tutorials vol 17no 2 pp 888ndash917 2015

Journal of Computer Networks and Communications 33

International Journal of

AerospaceEngineeringHindawiwwwhindawicom Volume 2018

RoboticsJournal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Active and Passive Electronic Components

VLSI Design

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Shock and Vibration

Hindawiwwwhindawicom Volume 2018

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Electrical and Computer Engineering

Journal of

Advances inOptoElectronics

Hindawiwwwhindawicom

Volume 2018

Hindawi Publishing Corporation httpwwwhindawicom Volume 2013Hindawiwwwhindawicom

The Scientific World Journal

Volume 2018

Control Scienceand Engineering

Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom

Journal ofEngineeringVolume 2018

SensorsJournal of

Hindawiwwwhindawicom Volume 2018

International Journal of

RotatingMachinery

Hindawiwwwhindawicom Volume 2018

Modelling ampSimulationin EngineeringHindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Chemical EngineeringInternational Journal of Antennas and

Propagation

International Journal of

Hindawiwwwhindawicom Volume 2018

Hindawiwwwhindawicom Volume 2018

Navigation and Observation

International Journal of

Hindawi

wwwhindawicom Volume 2018

Advances in

Multimedia

Submit your manuscripts atwwwhindawicom

Page 18: ReviewArticle - downloads.hindawi.com · ReviewArticle QoSStrategiesforWirelessMultimediaSensorNetworksinthe ContextofIoTattheMACLayer,ApplicationLayer,and Cross-LayerAlgorithms MuwongeSsajjabbiBernard
Page 19: ReviewArticle - downloads.hindawi.com · ReviewArticle QoSStrategiesforWirelessMultimediaSensorNetworksinthe ContextofIoTattheMACLayer,ApplicationLayer,and Cross-LayerAlgorithms MuwongeSsajjabbiBernard
Page 20: ReviewArticle - downloads.hindawi.com · ReviewArticle QoSStrategiesforWirelessMultimediaSensorNetworksinthe ContextofIoTattheMACLayer,ApplicationLayer,and Cross-LayerAlgorithms MuwongeSsajjabbiBernard
Page 21: ReviewArticle - downloads.hindawi.com · ReviewArticle QoSStrategiesforWirelessMultimediaSensorNetworksinthe ContextofIoTattheMACLayer,ApplicationLayer,and Cross-LayerAlgorithms MuwongeSsajjabbiBernard
Page 22: ReviewArticle - downloads.hindawi.com · ReviewArticle QoSStrategiesforWirelessMultimediaSensorNetworksinthe ContextofIoTattheMACLayer,ApplicationLayer,and Cross-LayerAlgorithms MuwongeSsajjabbiBernard
Page 23: ReviewArticle - downloads.hindawi.com · ReviewArticle QoSStrategiesforWirelessMultimediaSensorNetworksinthe ContextofIoTattheMACLayer,ApplicationLayer,and Cross-LayerAlgorithms MuwongeSsajjabbiBernard
Page 24: ReviewArticle - downloads.hindawi.com · ReviewArticle QoSStrategiesforWirelessMultimediaSensorNetworksinthe ContextofIoTattheMACLayer,ApplicationLayer,and Cross-LayerAlgorithms MuwongeSsajjabbiBernard
Page 25: ReviewArticle - downloads.hindawi.com · ReviewArticle QoSStrategiesforWirelessMultimediaSensorNetworksinthe ContextofIoTattheMACLayer,ApplicationLayer,and Cross-LayerAlgorithms MuwongeSsajjabbiBernard
Page 26: ReviewArticle - downloads.hindawi.com · ReviewArticle QoSStrategiesforWirelessMultimediaSensorNetworksinthe ContextofIoTattheMACLayer,ApplicationLayer,and Cross-LayerAlgorithms MuwongeSsajjabbiBernard
Page 27: ReviewArticle - downloads.hindawi.com · ReviewArticle QoSStrategiesforWirelessMultimediaSensorNetworksinthe ContextofIoTattheMACLayer,ApplicationLayer,and Cross-LayerAlgorithms MuwongeSsajjabbiBernard
Page 28: ReviewArticle - downloads.hindawi.com · ReviewArticle QoSStrategiesforWirelessMultimediaSensorNetworksinthe ContextofIoTattheMACLayer,ApplicationLayer,and Cross-LayerAlgorithms MuwongeSsajjabbiBernard
Page 29: ReviewArticle - downloads.hindawi.com · ReviewArticle QoSStrategiesforWirelessMultimediaSensorNetworksinthe ContextofIoTattheMACLayer,ApplicationLayer,and Cross-LayerAlgorithms MuwongeSsajjabbiBernard
Page 30: ReviewArticle - downloads.hindawi.com · ReviewArticle QoSStrategiesforWirelessMultimediaSensorNetworksinthe ContextofIoTattheMACLayer,ApplicationLayer,and Cross-LayerAlgorithms MuwongeSsajjabbiBernard
Page 31: ReviewArticle - downloads.hindawi.com · ReviewArticle QoSStrategiesforWirelessMultimediaSensorNetworksinthe ContextofIoTattheMACLayer,ApplicationLayer,and Cross-LayerAlgorithms MuwongeSsajjabbiBernard
Page 32: ReviewArticle - downloads.hindawi.com · ReviewArticle QoSStrategiesforWirelessMultimediaSensorNetworksinthe ContextofIoTattheMACLayer,ApplicationLayer,and Cross-LayerAlgorithms MuwongeSsajjabbiBernard
Page 33: ReviewArticle - downloads.hindawi.com · ReviewArticle QoSStrategiesforWirelessMultimediaSensorNetworksinthe ContextofIoTattheMACLayer,ApplicationLayer,and Cross-LayerAlgorithms MuwongeSsajjabbiBernard
Page 34: ReviewArticle - downloads.hindawi.com · ReviewArticle QoSStrategiesforWirelessMultimediaSensorNetworksinthe ContextofIoTattheMACLayer,ApplicationLayer,and Cross-LayerAlgorithms MuwongeSsajjabbiBernard