Research Article MHM: A Multiple Handshaking MAC …downloads.hindawi.com/journals/ijdsn/2016/9798075.pdf · Research Article MHM: A Multiple Handshaking MAC Protocol for Underwater

Research ArticleMHM A Multiple Handshaking MAC Protocol forUnderwater Acoustic Sensor Networks

Wen Lin12 and Keyu Chen3

1Department of Computer Science Minjiang University Fuzhou 350121 China2Fujian Provincial Key Laboratory of Information Processing and Intelligent Control Fuzhou 350121 China3The Key Laboratory of Underwater Acoustic Communication and Marine Information Technology Xiamen UniversityMinistry of Education Xiamen 361000 China

Correspondence should be addressed to Wen Lin linwen21163com

Received 25 December 2015 Revised 29 March 2016 Accepted 18 April 2016

Academic Editor Ana Alejos

Copyright copy 2016 W Lin and K ChenThis is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Underwater acoustic sensor networks (UWASNs) are effective tools for exploring and observing the ocean Due to thenonnegligible physical restrictions of the underwater acoustic communication most MAC protocols applied in the existingterrestrial wireless networks become inapplicable In this paper we propose a multiple handshaking MAC protocol for UWASNscalledmultihandshakingMAC (MHM)Using themethod ofmultiple handshaking and competitivemechanism of control packetsour protocol is proposed to make the contending nodes share the underwater acoustic channel much more fairly and moreefficiently The main idea of MHM is to allow multiple nodes to transmit and receive data packets at the same time without packetcollisionsWe also propose a competitivemechanism of control packets which can guarantee that there will not be data collisions inthe process of multiple packet transmissions Simulation results show that our protocol can achieve better performance includingthroughput delay and fairness

1 Introduction

The oceans occupy 71 of the earthrsquos area far more thanthe land area The development of the ocean has impor-tant significance for the whole human survival and socialdevelopment With the rapid development of wireless sensornetworks underwater acoustic sensor network (UWASN)which has low cost and high reliability is becoming a newresearch hotspot in recent years [1ndash3] UWASN is consideredas an important way to explore and develop the ocean itcan provide better technical support for resource protectionpollution monitoring disaster warning marine engineeringmarine military and other activities so UWASNs have beenpaid much attention by the governments and researchersaround the world [4]

Themedium access control (MAC) protocol enablesmul-tiple sensor nodes to share common channels without packetcollisions The main function of the MAC protocol is toensure the fair and efficient use of the channel Because of thelimited band in underwater acoustic channel the design ofMAC protocol becomes a key technology in UWASNs MAC

protocol directly affects the performance ofUWASNs such asthe throughput end-to-end delay and energy consumptionof the network Due to some inherent characteristics ofunderwater acoustic channel high propagation delay limitedenergy and multipath effect the design of MAC protocol ofUWASNs faces many challenges [5] Although there are a lotof mature MAC protocols on terrestrial wireless networksthose MAC protocols cannot be directly applied to UWASNsbecause of the different requirements and limitations So it isvery urgent to design new MAC protocols for UWASNs

The MAC protocol generally can be divided into twocategories competitive access and noncompetitive accessprotocol FDMA CDMA and TDMA are three main non-competitive access mechanisms However none of themcan be directly applied to UWASNs Because of the severelimitation and highly dependency on distance between twocommunicating nodes of the available bandwidth frequencydivision is not reasonable in UWASNs CDMA could alsobe difficult to be applied to UWASNs because of the near-far problem TDMA protocol has two disadvantages Thefirst disadvantage is that it requires precise synchronization

Hindawi Publishing CorporationInternational Journal of Distributed Sensor NetworksVolume 2016 Article ID 9798075 12 pageshttpdxdoiorg10115520169798075

2 International Journal of Distributed Sensor Networks

between nodes On the other hand due to the high trans-mission delay the duration of each time slot must be longenough whichwill lead to low network throughputThus thefeasibility of these protocols in UWASNs is unclear [6]

In recent years many competitive access protocols havebeen proposed for UWASNs [7 8] The classical ALOHA-based protocol does not have an effectivemechanism to avoidpacket collisions so it is not suitable for UWASNs [9] In theterrestrial wireless networks the performance of the CSMAprotocol is better than the ALOHA protocol However dueto large propagation delay carrier sense is hard to be realizedin UWASNs [10] Therefore more and more researcherspay attention to the access protocol based on handshakemechanism In the handshake protocol the nodes use thecontrol packet to compete for the use right of channel beforesending data packets Handshake protocols have the problemof hidden terminal and spatial fairness In such protocolsa node schedules its transmissions according to the controlpackets it hearsThese control packets also notify other neigh-bors about the ongoing transmission including the hiddennodes which could reduce collisions significantly Howeverthe handshaking mechanism was not destined for UWASNswhere long propagation delays are prevalent A situation of ahidden terminal problem occurs when a potential interferingnode cannot receive control packets on time and thus thosecontrol packets are unable to inform potential interferingnodes of the forthcoming packet transmissions A defect offairness problemalso existed in the handshake protocol Sincea packetrsquos arrival time is proportional to distance betweenreceiver and transmitter if the transmitter is closer to thereceiver it ismore likely to obtain the right to use the channel

To overcome the hidden terminal problem in UWASNsMolins and Stojanovic introduced the slotted FAMAprotocol[11] In the slotted FAMA protocol time is slotted and anypacket can only be transmitted at the beginning of a slotit is still using control packets to reserve time slots fordata packets Although the protocol can effectively avoidthe problem of hidden terminal however it still has someshortcomings In the slotted FAMA protocol the length oftime slot must be sufficient to ensure that all neighbor nodescan receive control packets in a slot so that they will knowwhether transmitting at the beginning of the next slot willinterfere with an ongoing transmissionThehigh propagationdelaymakes it expensive to exchange Request-To-Send (RTS)and Clear-To-Send (CTS) packets before each data packettransmission After a complete period of control packetsexchanging (two time slots) slotted FAMA only allows onesender-receiver pair access to channel Therefore the longslot length requirement and the inefficient handshakingmechanism affect the throughput and end-to-end delay ofthe UWASNs Finally the protocol does not solve the fairnessproblem

In order to overcome the above problems we propose anew MAC protocol with features of efficiency and fair useof network resources for UWASNs and refer to this MACprotocol as MHM (Multiple Handshaking protocol) MHMcan achieve higher throughput and better fairness by com-bining multiple handshaking mechanism and competitivemechanism of control packets Multiple nodes are allowed

to exchange control packets (RTS or CTS) in the same timeslot so the efficiency of the control packet switching is alsoimproved The multiple handshaking mechanism is develop-ment to avoid collisions of data packets The key idea of thismechanism is to utilize the information of propagation delayto arrange the transmission time of data packets Receivercan receive multiple data packets in a packet train manneras long as multiple senders transmit data packets accordingto the special transmission schedule In our protocol aftersome control packets are exchanged some nodes may needto receive and transmit data packets at the same time Theobjective of competitive mechanism of control packets is tospecify the order in which these nodes transmit and receivecontrol packets In summary in the phase of control packetswitching the node can transmit and receivemultiple controlpackets In the phase of data transmission the node canreceivemultiple data packets thereforeMHMcan reduce thetime during which the control packets occupy the channelimprove the utilization ratio of the channel and ensure betterfairness

The rest of this paper is organized as follow Section 2briefly reviews some related works In Section 3 we presentthe MHM protocol In Section 4 the simulations are carriedout Finally we give our conclusions and further work inSection 5

2 Related Work

There are several existing competitive access protocols forUWASNs Some competitive access protocols use the tradi-tional handshaking mechanism such as MACA MACAUand MACAW However these protocols cannot effectivelysolve the hidden terminal and spatial fairness problem causedby high propagation delay in multihop UWASNsThe FAMAprotocol be proposed in [12] It was shown that using a carriersensing protocol collision avoidance is guaranteed if the fol-lowing conditions hold (1) the length of RTS packets shouldbe greater than the maximum propagation delay and (2) thelength of CTS packets should be greater than the lengthof RTS packets plus twice the maximum propagation delayplus the hardware transmit-to-receive transition time Theseconditions are the basis of the FAMA protocol AlthoughFAMA can ensure the absence of collisions in the networksthe length of control packets becomes excessive on anunderwater acoustic channel Molins and Stojanovic aimedat the hidden terminal problem caused by spatial-temporaluncertainty and proposed the slotted FAMA protocol basedon FAMA [11] Although the slotted FAMA protocol achievesguaranteed collision avoidance for data packets the longslot length requirement and the handshaking mechanismitself affect the throughput In [13] Qian et al proposeda slotted FAMA based MAC protocol for UWASNs calledRC-SFAMA It proposed an RTS competition mechanism toprevent nodes into back-off state caused by the competitionof multiple RTS packets in a slot Via the RTS competitionmechanism useful data transmission can be completed suc-cessfully when the situation of multiple RTS attempts occursIn [14] SF-MAC developed a receiver-based protocol to solve

International Journal of Distributed Sensor Networks 3

spatially unfair problem SF-MAC can avoid the unfairnessof channel access by postponing the CTS packet equal toperiod of RTS contention period The receiver collects RTSpackets from all potential senders during the RTS contentionperiod and calculates the potential sending time of eachof contenders It determines the earliest transmitter with aprobability rule that compares with the first RTS packet Incase of multiple potential senders the SF-MAC can maintaina more exact order of transmission to achieve fairness oftransmission Although the above protocols can improvenetwork performance these protocols do not minimize theeffect of the channel resources waste problem in controlpacket switching therefore they can only be maintained atrelatively low levels in the throughput

In the handshake protocol transmission of controlpackets in UWASNs will decrease the channel utilizationbecause of the consideration of high propagation delay thethroughput of UWASNs is usually poor After a successfulhandshake the traditional handshake mechanism (MACA-base or FAMA-base) can only carry on a data packettransmission The existence of the problem gives rise to theneed to design new handshake mechanism RIPT which is afour-handshaking-based MAC protocol is proposed in [15]to increase the channel utilization of UWASNs Receiversinitiate the 4-way handshake by broadcasting an RTR packetA receiver broadcasts an RTR control packet to its neighborsIf neighbors have packets to send to this receiver theycan replay the data size in another control packet to thereceiver Finally the receiver can schedule each transmissionand then informs the scheduling results in ORDER controlframe Xun et al proposed a protocol called ROPA [16] itcan increase the channel utilization by allowing a sender toinvite its one-hop neighbors to opportunistically transmittheir data packets After the sender finishes transferringits data packets to the receiver it can immediately receivethe incoming data packets from its neighbors In [17] theyproposed a receiver-initiated MAC protocol with packettrain for UWASNs called multireceiver MAC (MR-MAC)protocol The MR-MAC protocol can make more than twonodes to communicate in one handshake held by a mainreceiver By scheduling the packet transmission time thedata packet will be sent in a packet train manner andthe receiver can receive data packet without collision In[18] they proposed a handshake based ordered schedulingMAC (HOSM) protocol The nodes with data packets to betransmitted first reserve the channel in a channel reservationphaseThen an order list is calculated and the data packets ofthese nodes are transmitted according to this order list Theydevelop control packets transmission adjustmentmechanismto reduce collisions of control packets The key idea of thismechanism is to utilize the information of propagation delayto adjust the time instant of control packets transmittingTheabove protocols are too complicated to effectively improvenetwork throughput and need too much control packetswhich will influence the network performance

Some protocols try to improve the network perfor-mance by designing a new handshaking mechanism with-out increasing the number of control packets Chirdchooet al [19] proposed a MACA-based protocol with packet

train to multiple neighbors (MACA-MN) It improves thechannel utilization by forming a train of packets destinedfor multiple neighbors during each round of handshakewhich can reduces the relative proportion of time wasteddue to the propagation delays of control packets Han etal proposed the multisession FAMA (M-FAMA) protocol[20] it allows a sender to open multiple sessions to differentreceivers achieving temporalspatial reuse and yet avoidingcollisions by careful accounting of neighborsrsquo transmissionschedules In [21] author presented a MR-SFAMA protocolfor UWASNs Besides adopting the handshake mechanismof slotted FAMA it uses the multiple receive mechanismwhich significantly improves the network throughput andfairness index However these protocols only can allow nodesto receive or send multiple packets they do not allow nodesto be two roles as both the sender and the receiver Similarto some of the aforementioned protocols our protocol seeksto improve network performance by reducing the proportionof time spent on control packets In our protocol after thereceiver finishes receiving its data packets from multipleneighbors which arrive in a manner of packet train it canimmediately switch its role to transmit data packets to itsdestination nodes Therefore it can achieve better networkperformance and spatial fairness

3 MHM Protocol Design

31 Motivations and Basic Principles Although the slottedFAMA and RC-SFAMA protocol are used for alleviating thehidden and exposed terminal problems in UWASNs theyalso include two disadvantages when they are applied inUWASNs Firstly the need for at least one full round-tripexchange of control packets (RTS-CTS) prior to sendingevery data packet introduces lager latency due to the longpropagation delay At the same time in order to ensure that allhidden terminals can listen to the control packets transmittedby the sender in a time slot the duration of time slot mustbe long enough When the network load is high if a nodereceives multiple RTS packets in a slot the target node inslotted FAMA protocol must enter the back-off state In theRC-SFAMA protocol the receiver only can obtain one datapacket from one of all potential sendersThereforeThis leadsto low utilization of the network As shown in Figure 1 node1 is close to node 2 and node 3 is far from node 2 Thetwo simultaneous transmissions of RTS packets from node1 and node 3 are received separately at node 2 in the slot 1Although node 2 received two RTS packets it can only replyCTS packets to one of the senders (node 2) in next slot Node3 receives the xCTS (a CTS packet intended for another node)packet from node 2 in the waiting time (current slot and thenext one) and it must wait long enough to allow the node 2 toreceive the entire data packet and receive the following ACKpacket from node 2

Our protocol allows the receiving node to receivemultipleRTS packets in a time slot At the beginning of the next timeslot the CTS packet from the receiver is sent to all sendersby means of broadcasting MHM protocol leverages theRTSCTS exchange for learning propagation times between


Node 1

Node 2

Node 3

Slot 1 Slot 2 Slot 3 Slot 4 Slot 5

R

C

D

A

R

CRTS

xRTS

CTS

xCTS

DATA

ACK

Time

R

Time

Time

1

R1

R3

R3

C2

C2

2

D

D

1

1A2

A2

A2

C

Figure 1 An example of RC-SFAMA

Node 1

Time

Node 2

Node 3

Time

Time


R

C

D

A

R

CRTS

xRTS

CTS

xCTS

DATA

ACK

R1

R1

R3

R3

C2

C2

2

D

D

1

1A2

A2

A2

C D3

D3

Figure 2 An example of MHM

a node and its neighbors With loose clock synchronizationamong terminals and a known transmission and propagationtime for each data packet (assumed of fixed size) the receivercan know the time of receiving the data packets sent byall potential transmitting nodes Using the knowledge oftransmitting nodesrsquo propagation time the receiver can designa collision-free transmission schedule of its own At last thereceiver broadcasts the transmission schedule to all of thesenders byCTS packets In the case depicted in Figure 2 node2 receives two RTS packets in the first slot and broadcasts aCTS packet at the beginning of the next slot According tothe transmission order in the CTS packet node 1 and node 3send their data packets to node 2 sequentially Note that node2 activates two types of handshaking for node 1 and node 3 bybroadcasting a CTS packet

In our protocol we use another mechanism to improvethe utilization of the control packets In the slotted FAMAand RC-SFAMA protocol the role of a node can only be asender or receiver If a sender receives an RTS packet fromits neighbor nodes after sending its RTS packet in a time slotthe sendermust ignore the RTS packet from other nodesThereason is that when it sends an RTS packet its role becomes

a sender so it cannot receive other RTS packets before itcompletes its data packet transmission As shown in Figure 3node 2 has a data packet to be sent to the node 3 Howeverat the same slot node 1 also has a data packet to transmit tothe node 2 when node 2 receives the RTS packet transmittedby node 1 node 2 neglects the RTS packet from the node 1because it is communicating with node 3

Since the time to send CTS packets is at the beginning ofa time slot a node transmits CTS packets and receives CTSpackets which can be performed simultaneously in a time slotwithout packet collisions As shown in Figure 4 node 2 canhandshake with the other two nodes simultaneously Node 2has received the data packet fromnode 1 then it sends its datapacket to node 3 immediately Using this mechanism node3 can receive and send data packets after one handshakingtime Based on the above two mechanisms our protocol canimprove the success rate of control packet switching and thenimprove the performance of the network

Another shortcoming of the RC-SFAMA protocol is thefairness problem The mechanism of RTS packets competi-tion makes it essential to also consider the fairness all poten-tial senders Since a nodersquos a competing number is obtained


Node 1

Node 2

Node 3

Node 1 defers it transmissionSlot 1 Slot 2 Slot 3 Slot 4 Slot 5

R

C

D

A

R

CRTS

xRTS

CTS

xCTS

DATA

ACK

Time

RTime

Time

2

R

C3

D2

A3

R1

R1

C3

D2

A32

R2


Time

RTime

Time

2

R

C3

D1

A3

R1

R1

C3

D2

A32

R2

C2

C2

D1

D2

A2

A2

Node 1

Node 2

Node 3


R

C

D

A

R

CRTS

xRTS

CTS

xCTS

DATA

ACK


by random number generation the channel becomes clearearlier when the node has small random number Let us lookback at Figure 1 node 1 and node 3 transmit simultaneouslyan RTS packet to node 2 Because the random number ofnode 3 is greater than the node 1 the node 1 has a greaterchance to handshake with node 1 than node 3 and thisadvantage will be maintained Figure 2 shows an oppositeexample in the MHM protocol node 1 and node 3 can senddata packets in the same slot MHM does not find any factorsthat would affect the probability of access channelThereforeour protocol can improve the fairness of channel access andensure that the network has long survival life

32 UWASNS Model MHM is a kind of access protocolbased on time slot and it requires the node position inthe network to be fixed Therefore this paper assumes thatnetwork model has these following properties

(1) The network topology can be a single-hop or multi-hop topology and all nodes are placed in a randommanner in a network area

(2) Each node acts independently from the others andsends data packets following Poisson distributionNodes can randomly select any node as the destina-tion

(3) There are synchronization requirements between allnodes Nodes work on half-duplex model and nodesrsquolocation is fixed

(4) The optimal routing path is chosen as the onewith thefewest number of hops

33 MHM Protocol The MHM protocol focuses on staticnetworks we assume that all nodesrsquo clocks are synchronizedand any clock drift would still be negligible Under thisassumption each static node can know the propagationdelay between it and its neighbor nodes in the initializationphase In fact the propagation delay can be estimated by ahandshake between the nodes [9]

Similar to the widely known FAMA protocol ourprotocol also employs a four-way handshake (RTSCTSDATAACK) When a node has data packets to transmit itsends an RTS packet at the beginning of the next slot and


Time

R

Time

Time

1

R1

R3

R3

C2

C2

C2

D

D

1

1A2

A2

A2

D3

D3

R1

C1

A1

Time

R2

R2

C1

C1

D2

D2

A1

A1

R2

Node 1

Node 2

Node 3

Node 4

Waiting time New waiting time

R

C

D

A

R

CRTS

xRTS

CTS

xCTS

DATA

ACK

Figure 5 The procedures of MHM protocol

wait the CTS packet from the receiver Each node can onlytransmit an RTS packet in a single transmission cycle Whena node receives an RTS packet it continues listening channeluntil the end of the current slot If the node overhears otherRTS packets during the listening time it also receives thoseRTS packets In the next time slot note that there is animportant modification on how the receiving node shouldrespond with its CTS packets compared with the slottedFAMA and RC-SFAMAprotocol because our protocol needsto handle more than one RTS packet from multiple sendersThe receiving node has two ways to reply the CTS packet Ifthe receiving node has received only one RTS packet in theprevious slot it replies a CTS packet to the sending nodeat the beginning of current slot In contrast if the receivingnode has received multiple RTS packets in the previous slotit broadcasts a CTS packet to a plurality of sending nodes atthe beginning of current slot As shown in Figure 5 node 1and node 3 are neighbor nodes of node 2 node 3 is hiddenfrom node 1 and node 4 is a neighbor node of node 1 At thebeginning of the first slot nodes 1 2 and 3 simultaneouslytransmit an RTS packet Node 2 receives two RTS packetsin the first slot then it broadcasts a CTS packet to node 1and node 3 in the next slot which includes a transmissionschedule of data packets for all senders (node 1 and node 3)In the same slot node 2 also received a CTS control packetfrom node 1

In the third slot there are three nodes that want to senddata packets In order to avoid packet collisions we introducea competitive mechanism of control packets which will befurther discussed in Section 35 Using this mechanism allnodes know the order of sending data packets In the exampleshown in Figure 5 node 1 and node 3 send their data packetsfollowing the transmission schedule in the third slot Node 1sends a data packet directly in the next slot time and node 3must wait for a period of time to send its data packet Thistransmission schedule is to ensure that node 2 can receivetwo data packets without collision which will be furtherdiscussed in Section 34 When node 2 has received data

packets it sends its data packet to node 1 At last node 2broadcasts an ACK packet in the next time slot If node 1 andnode 3 have received the ACKpacket they know that the datatransmission is successful

If a node receives an xRTS packet (an RTS packetintended for another node) in the idle state it must wait twoslots If after this time no carrier is sensed the node returnsto the idle state Different from the slotted FAMA and RC-SFAMA protocol if the CTS packet sent by the neighbornode is received during the waiting period the node shouldreceive the CTS packet and transmit its data packet in thefollowing slot After receiving an xCTS packet the node mustwait long enough to allow other nodes to transmit the entiredata packet and receive the correspondingACKpacket Let uslook back at Figure 5 when node 4 overhears the xRTS packettransmitted by node 1 it must wait two slots (current slot andthe next one) If during this waiting time no control packetsare received node 4 returns to the idle state However node 4receives the xCTSpacket transmitted by node 1 during secondslot it must reset its waiting time to allow node 1 to receivedata packets The waiting time must be long enough to allowthe reception of the subsequent ACK packet After hearingthe ACK packet from node 1 node 4 can return to the idlestate since the data transmission has successfully endedNode3 receives an xRTS packet in the first slot and then it entersthe waiting state but it receives a CTS packet from the node2 in the second slot then node 1 must be out of the waitingstate and ready to transmit its data packets to node 2

34 Multiple Handshaking Mechanism As mentioned abovewhen using the handshake protocol the high propagationdelays can reduce the throughput and increase end-to-enddelay of networks However long propagation delay alsoprovides an opportunity to transmit or receive multiplepackets at the same slot In our protocol when the receivingnode receives several RTS packets from different sendersin a slot it is allowed to broadcast a CTS packet to allsenders at the next slot After a control packet switching cycle


Node 1

Node 2

Node 3

Node 4

R

C

D

A

R

CRTS

xRTS

CTS

xCTS

DATA

ACK

TimeR

Time

1

R2

C1

D2

A2

A2

A2

R1

C1

A1


R1

R2

R2

C1

D2

C2

C2

D1

D1

A1

A1

Time

Time

C2

Figure 6 An example of competitive mechanism of control packets

the receiving node can receive a plurality of data packetsThis mechanism makes it possible to transmit multiple datapackets at the end of the time for one full round-trip exchangeof control packetsThe rule of thumb here is that senders willtransmit those data packets at the beginning of the next slotupon receiving the CTS packet subject to the condition thatthose data packets will not result in data packet collisions atthe receiver If the difference between the propagation delayof senders and the receiver is less than the transmission timeof a data packet data packet collisions may occur at thereceiver In our protocol we use a transmission schedule toavoid the occurrence of the aboveWhen a receiver calculatesand learns that if all senders transmit data packets at thebeginning of time slot this will cause data packet collisionsit will design a transmission schedule and send the scheduleto all senders via a CTS packet Having resolved the timeto transmit its data packet each sender can transmit datapackets in accordance with the schedule Therefore the coreof this mechanism is how to design a suitable transmissionschedule

Next let us introduce how the receiving node computesa data transmission schedule If a receiving node 119903 receivesmultiple RTS packets using the internode propagation delayinformation provided by the initialization phase it cancalculate and arrange the time for all senders to send itsdata packets The sequence of senders transmitting packets isdetermined by the distance between them and the receiver Ifa sender is close to the receiver it can transmit data packets atan earlier timeOtherwise it needs towait for a period of timeto send data packets Let119882

119909119896be the waiting duration of 119896th-

order sender 119909 Let119863119903119909

be the propagation delay between thereceiver and the sender 119909 and let119879

119889be the transmission time

of a data packet When the 1st-order sender receives the CTSpacket according to the arrangement of the transmissionschedule it sends data packet immediately at the beginningof the next slot After receiving the CTS packet the othernodesmust enter into the waiting state and their data packets

will be transmitted after waiting time Therefore119882119909119896

can becalculated as follows

119882119909119896= 119863119903119896minus1+ 119879119889+119882119909119896minus1minus 119863119903119896 (1)

When119882119909119896

is less than or equal to zero the sending node cansend packets directly without the need to enter the waitingstate

35 Competitive Mechanism of Control Packets After aswitching period of control packets the MHM protocolallows some nodes to transmit or receive data packets insubsequent time slots How to arrange the order of nodes totransmit and receive data packets is essential In our protocolwe use a mechanism to solve the above problem we call itcompetitive mechanism of control packets

In the MHM protocol we add an CTS competitionmechanism to all sending or receiving nodes When a nodewants to send a CTS packet it adds a random number intothe CTS packet Then the node sends the CTS packet withthe random number at the beginning of current slot In thisway each CTS packet will have a random number within itWhen network traffic is large for some senders it sends aCTS packet at the beginning of a time slot while it also mayreceive aCTSpacket sent by other nodes in the same time slotThe sender makes its own random number compete with therandom number of the received CTS packets If the randomnumber of the senders is relatively large in the followingtime slot it will send its data packet first and then wait toreceive the data packet from neighboring nodes As shownin Figure 6 the random number of node 2 is larger thannode 1 therefore node 2 sends its data packet to node 1 afterexchanging of control packets and node 1 can only send itsdata packet after receiving the data packet from node 2

However there is still a special case in our protocolWhentwo nodes transmit CTS packets between each other if onenode receives CTS packets sent by broadcast manner it does


not require random number comparison because the levelof CTS packets sent by broadcast manner is higher than thegeneral CTS packetsThe reason is that a broadcast packet cancommunicate with multiple sending nodes at the same timeso the authority is the highest As shown in Figure 5 node 1receives a CTS broadcast packet and then it must send datapackets to node 2 before receiving data packets from node 2

36 Throughput Analysis Let us assume that our analyticalmodel consists of a single receiving node 119909 and 119873 neighbornodes Neighbor nodes are randomly distributed aroundthe receiving node 119909 Because the transmission time ofRTS packets is far less than the internode propagationdelay between neighbor nodes and receiver the collisionprobability of RTS packets is very small In our model weassume that the receiver can receive all the RTS packets ina time slot without collision Each node has a packet readyto send every 1120582 seconds on the average (the arrivals followPoisson distribution with average 120582 packets per second ieexponentially distributed interarrival time) Since our studyonly focuses on the performance of the MAC protocol thechannel is assumed to be error-free in our analytical model

Let us define119875119904as the probability of success (no collisions)

on the channel In the slotted FAMAprotocol the probabilityof no collisions is the probability that no other neighborstransmit within a time slot used by a neighbor node 120596 Theprobability can be expressed as

119875slotted FAMA119904

=

119873minus1

prod1

119890minus120582119879slot = 119890

minus120582(119873minus1)119879slot (2)

Similarly in the RC-SFAMA protocol the probability ofno collisions is the probability that node 120596 sends an RTS andwins the RTS competition if the contenders of node 120596 alsosend RTS packets in the same slot Therefore

119875RC-SFAMA119904

=

119873minus1

sum119894=0

(

119873minus1minus119894

prod1

119890minus120582119879slot (1 minus 119890

minus120582119879slot)119894 1

119894 + 1) (3)

In our protocol we use the multiple handshaking mech-anism to solve the problem of RTS packets competitionthere is no relationship between the 119875

119904of node 120596 and other

neighbors whether to send RTS packets in the same time slotHence

119875MHM119904=

119873minus1

sum119894=0

(


prod1

119890minus120582119879slot (1 minus 119890

minus120582119879slot)119894

) = 1 (4)

Next we estimates the throughput of MHM in the abovenetwork model In this network model we assume that allneighbor nodes have a data packet sent to node 119909 in a giventime 119905

0 119878 is the throughput of each neighbor node and this

throughput can be expressed as

119878 =119879119906

119879ot (5)

where 119879119906is the time while useful data is being sent for

neighbor nodes and 119879ot is the network operation time

Denoting by 119879data the transmission time of each fixed-length data packet the time during which useful data is sentfrom neighbor node 120596 is obtained as

119879119906= 119879data times 119875119904 (6)

Figure 1 describes a complete handshake process and 119879ℎ

is the duration of a successful data packet transmission cyclewhich is given by

119879ℎ= 3119879slot + 119879data + 119863119909119894 + (119879data + 119863119909119894) mod (119879slot) (7)

where 119863119909119894

is the internode propagation delay betweenreceiving node 119909 and neighbor node 119894

The duration of the network operation time is determinedaccording to the access method of the protocol In slottedFAMA and RC-SFAMA protocol when a control packetswitching is completed only one node can send its datapacket so their network operation times are

119879slotted FAMAot =

119873

sum119894=1

(119879119894

back-off + 119879ℎ)

119879RC-SFAMAot = 119879

ℎ+

119873minus1

sum119894=1

(119879119894


(8)

In the MHM protocol the multiple handshaking mech-anism makes it possible to transmit multiple data packets atthe end of the time for one full round-trip exchange of controlpackets Therefore the operation time of network is given by

119879MHMot = 3119879slot + 119863min(119909119894) + 119873119879data

+ (119863min(119909119894) + 119873119879data) mod (119879slot) (9)

where119863min(119909119894) is the minimum single-trip propagation delaybetween the receiving node 119909 and neighbor node 119894

In this situation the theoretical throughput of MHM is

119878

=119879data

3119879slot + 119863min(119909119894) + 119873119879data + (119863min(119909119894) + 119873119879data) mod (119879slot)(10)

Linking all the parts our protocol shows better perfor-mance than the other two protocols in throughput Thisequation is valid for a static single-hop network

4 Simulation Results

41 Simulation Settings In our simulation a multihop net-work is investigated The simulation parameters are as fol-lows we simulate a random network where several under-water acoustic sensor nodes are uniformly distributed in asquare area with the side of 10 km A sending node randomlychooses another node in the network as the destination Eachnode can only transmit a packet in a packet transmissioncycle We model packets traffic as a Poisson arrival processData packets are generated at each node in accordancewith Poisson distribution The bandwidth of the channel


SFAMAS-ALOHA

MHMRC-SFAMA

0

005

01

015

02

025

03

035

04

045

05

Thro

ughp

ut

02 04 06 08 10Offered load

Figure 7 Throughput versus offered load

is set to 1 kbps and the rate of data transmission is 1 kbpsThe transmission range of every node is set to be 1500mThe propagation speed of acoustic signal in underwaterenvironment is about 1500ms The channel is also assumedto be error-free so that all packet losses are purely due to theMAC protocolrsquos performance The data packet size is set to256 bits and all control packets (RTS CTS and ACK) are setto 16 bits All acoustic modems are used in half-duplex mode

We carry out these simulations in the OPNET simulatorTo validate the performance of our protocol in this simu-lation scenario all the nodes are stationary once deployedWe compare our protocol with three previously proposedprotocols namely slotted ALOHA slotted FAMA and RC-SFAMA Note that all the protocols in our simulation studyrequire time synchronization In the process of simulationwe assume that all nodes in the network can keep accuratetime synchronization

42 Simulation Results For a better understanding of ourprotocolrsquos performance we use four metrics as our perfor-mance measure as follows

421 Throughput Figure 7 shows that our protocol alwaysoutperforms other protocols significantly while being ableto maintain a high throughput in heavy load network envi-ronment When the network load is low the slotted ALOHAprotocol can achieve highermaximum throughput than otherprotocols Because of the lack of an effective mechanism toavoid the collisions of data packets the throughput of slottedALOHA actually decreases as the network load increasesHowever handshaking-based protocols could help reducecollisions in UWASNs by alleviating the hidden terminalproblem they maintain stable throughput as the offeredload increases As expected slotted FAMA is very inefficientbecause of the relatively long time slot and low efficiency ofits handshake mechanism RC-SFAMA introduces an RTS

competitionmechanism to overcome the problemofmultipleRTS packet attempts Via the RTS competition mechanismRC-SFAMA protocol has higher throughput than the slottedFAMA protocol However with the increase of the offeredload in these two protocols they restrict channel access toonly one sender-receiver pair during a slot time Nodes onlytransmit or receive a single data packet after each handshakewhich leads to a very low utilization rate of the channel whenthe propagation delay is high

The MHM protocol superior performance over otherthree protocols is due to the multiple handshaking mech-anism and competitive mechanism of CTS It allows thereceiving node receive multiple RTS packets within one cycleof control packets exchanging Consequently all intendedsending nodes may transmit their data packets to the receiverwithout packet collisions After exchanging of control pack-ets our protocol allows nodes to receive and transmit packetsat the following time slot Therefore the two mechanismsdescribed above can improve the utilization rate of controlpackets and thus improve the network throughput Whenthe network throughput is small the throughput of slottedFAMA RC-SFAMA and MHM protocol is very close Thisphenomenon is mainly caused by the following reasons Theprobability of sending and receiving RTS in the same slot isrelatively small Therefore the possibility of receiving nodereceives multiple RTS packets in a time slot relatively low Asthe network offered load increases the method of multiplehandshaking and competitive mechanism of control packetscan reduce the total channel reservation overhead greatly andthus can improve channel utilization

422 The Ratio of RTSDATA The ratio of RTSDATA wasused to evaluate the throughput performance of our protocolin simulation This ratio represents the average number ofRTS packets to be sent before a successful transmission ofa data packet If the ratio is relatively small the probabilityof successful control packets switching is relatively largeIn this case the time during which the control packetsoccupy channel is reduced and the channel utilization willbe improved Figure 8 shows the ratio of slotted FAMA RC-SFAMA and MHM as a function of offered load SlottedALOHA protocol does not require a handshake of controlpackets before transmission of data packets therefore thesimulation experiment will not consider this protocol

At the lower offered load the RTS packets are successfullyreceived by the high probability As offered load increasesmore and more nodes listen to multiple RTS packets orxRTS packets in the slotted FAMA protocol nodes cannotreply CTS packets to these RTS packets so the ratio ofRTSDATA will always be increased In contrast the RC-SFAMA protocol maintain relatively low RTSDATA ratio asthe offered load increases The result can be explained by theefficient handshaking mechanism of RC-SFAMA protocolThe problem that a node cannot receive multiple RTS packetswithin a time slot at higher offered load is eliminated bythe RTS competition mechanism However only the nodewhich wins the RTS competition can send data packetsand the RTS packets sent by other nodes will be invalid


SFAMAMHM

RC-SFAMA

0

2

4

6

8

10

12RT

SD

ATA

ratio


Figure 8 RTSDATA ratio versus offered load

MHMuses themultiple handshakingmechanism to solve theabove problemThe receiver can handshake with all potentialsenders simultaneously therefore MHM achieves a betterperformance in the ratio of RTSDATA

423 Fairness Fairness is a key performance of MAC proto-col which affects the normal operation and survival time ofUWASNs To evaluate the fairness of our protocol we adoptthe Jain Fairness Index define in [22]

FI =(sum1205881198941205881015840119894)2

119899sum (1205881198941205881015840119894)2 (11)

where 120588119894and 1205881015840

119894represent the throughput and the ideal

throughput of node 119894 and 119899 is the number of nodes in thenetwork We assume that all nodes have the same idealthroughput and 0 lt FI le 1 When the index is close to 1this indicates that the protocol has a good fairness

Figure 9 shows the fairness index of four protocolsWith the increase of offered load the fairness index of theslotted ALOHA slotted FAMA and RC-SFAMA is decliningThis is because the back-off algorithm they used will causeunfair In addition to the above reasons in slotted ALOHAdue to the large delay of underwater acoustic networkthe distance between nodes becomes a key factor in thecompetitive channel In RC-SFAMA the RTS competitionmechanism also introduces the unfairness of node accesschannel We discover that the MHM protocol has the bestfairness performance This is explained by the fact thatmultiple handshaking mechanism can improve the fairnessindex It guarantee that a sender who is farther from thereceiver has equal chance of capturing the channel

424 Average Delay In Figure 10 we compare the averagetransmit delay of data packets At low offered load slotted

SFAMAS-ALOHA

MHMRC-SFAMA

0

01

02

03

04

05

06

07

08

09

Fairn

ess i

ndex


Figure 9 Fairness index versus offered load

SFAMAS-ALOHA

MHMRC-SFAMA

0

200

400

600

800

1000

1200

1400Av

erag

e del

ay (s

)


Figure 10 Average delay versus offered load

ALOHA has the best delay performance This is because itdoes not need to exchange control packets before transmit-ting DATA packets However with the increase of offeredload slotted ALOHA has the highest delay because it has noeffective mechanism to avoid data packet collisions The RC-SFAMA protocol is better than slotted FAMA protocol in theperformance of average delay This is because in the case ofRTS packets competition RC-SFAMA can work normallyAt high offered load our protocolrsquos average delay becomessmaller than slotted FAMA and RC-SFAMAThis is the pointwhere the multiple handshaking mechanism can overcomethe overheads incurred by 4-way handshake Our protocolcan use the mechanism more efficiently by sending andreceiving multiple data packets in a transmission cycle


5 Conclusions and Further Work

The long propagation delay of underwater acoustic channelis inevitable and it has great influence on the performance ofhandshakingMACprotocols inUWASNsWehave presentedthe MHM protocol which is a random access handshaking-based protocol It is shown that by using multiple handshakeand competitive mechanism of control packets the MHMprotocol can addresses the channelrsquos long propagation delaycharacteristicThe simulation results have confirmed that ourprotocol can achieve better performance than twootherMACprotocols Future work will mainly focus on the design of anew back-off algorithm the main goal is to reduce the nodersquosback-off time and improve the fairness during the networkwhich is under high level load

Competing Interests

The authors declare that there are no competing interestsregarding the publication of this paper

Acknowledgments

This work was supported in part by the Science Foundationfor Youths of Fujian Province (Grant no 2016J05160) by theNatural Science Foundation of China (Grant no 61501386)and by the Start-Up Foundation for Talent Introduction ofMinjiang University (Grant no MJY15001)

References

[1] I F Akyildiz D Pompili and TMelodia ldquoUnderwater acousticsensor networks research challengesrdquo Ad Hoc Networks vol 3no 3 pp 257ndash279 2005

[2] M Chitre S Shahabudeen and M Stojanovic ldquoUnderwateracoustic communications and networking recent advances andfuture challengesrdquo Marine Technology Society Journal vol 42no 1 pp 103ndash116 2008

[3] J Partan J Kurose and B N Levine ldquoA survey of practicalissues in underwater networksrdquo ACM SIGMOBILE MobileComputing and Communications Review vol 11 no 4 pp 23ndash33 2007

[4] J Heidemann Y Li and A Syed ldquoUnderwater sensor net-working research challenges and potential applicationsrdquo inPro-ceedings of the IEEE Wireless Communications and NetworkingConference (WCNC rsquo06) pp 228ndash235 Las Vegas Nev USAApril 2006

[5] M Stojanovic and J Preisig ldquoUnderwater acoustic communica-tion channels propagation models and statistical characteriza-tionrdquo IEEE Communications Magazine vol 47 no 1 pp 84ndash892009

[6] I F Akyildiz D Pompili and T Melodia ldquoState-of-the-art inprotocol research for underwater acoustic sensor networksrdquoin Proceedings of the 1st ACM International Workshop onUnderwater Networks (WUWNet rsquo06) pp 7ndash16 Los AngelesCalif USA September 2006

[7] Y Noh and S Shin ldquoSurvey on MAC protocols in under-water acoustic sensor networksrdquo in Proceedings of the 14thInternational Symposium on Communications and InformationTechnologies (ISCIT rsquo14) pp 80ndash84 September 2014

[8] K Chen M Ma E Cheng F Yuan and W Su ldquoA survey onMAC protocols for underwater wireless sensor networksrdquo IEEECommunications Surveys and Tutorials vol 16 no 3 pp 1433ndash1447 2014

[9] A A Syed W Ye J Heidemann and B KrishnamacharildquoUnderstanding spatio-temporal uncertainty in medium accesswith ALOHA protocolsrdquo in Proceedings of the 2nd ACMWorkshop on Underwater Networks (WUWNet rsquo07) pp 41ndash48Montreal Canada September 2007

[10] L Kleinrock and F A Tobagi ldquoPacket switching in radiochannels part I-carrier sense multiple-access modes and theirthroughput-delay characteristicsrdquo IEEE Transactions on Com-munications vol 23 no 12 pp 1400ndash1416 1975

[11] M Molins and M Stojanovic ldquoSlotted FAMA a MAC protocolfor underwater acoustic networksrdquo in Proceedings of the IEEEOCEANS Asia Conference pp 1ndash7 Singapore May 2007

[12] C L Fullmer and J J Garcia-Luna-Aceves ldquoFloor acquisitionmultiple access (FAMA) for packet-radio networksrdquo ACMSIGCOMMComputer Communication Review vol 25 no 4 pp262ndash273 1995

[13] L-F Qian S-L Zhang and M-Q Liu ldquoA slotted floor acquisi-tionmultiple access basedMACprotocol for underwater acous-tic networks with RTS competitionrdquo Frontiers of InformationTechnology amp Electronic Engineering vol 16 no 3 pp 217ndash2262015

[14] W-H Liao and C-C Huang ldquoSF-MAC a spatially fair MACprotocol for underwater acoustic sensor networksrdquo IEEE Sen-sors Journal vol 12 no 6 pp 1686ndash1694 2012

[15] H-HNgW-S Soh andMMotani ldquoMACA-U amedia accessprotocol for underwater acoustic networksrdquo in Proceedings ofthe IEEE Global Telecommunications Conference (GLOBECOMrsquo08) pp 1ndash5 New Orleans La USA December 2008

[16] L Xun L Yu F Dong Z Chun-Hua andHHai-Ning ldquoAMACprotocol for underwater acoustic networkrdquo in Proceedings ofthe International Conference on Communication Electronics andAutomation Engineering G Yang Ed vol 181 of Advances inIntelligent Systems and Computing pp 1291ndash1297 Xirsquoan ChinaAugust 2012

[17] W H Liao Y C Lin and S C Kuai ldquoA receiver-initiatedMAC protocol for underwater acoustic sensor networksrdquo inProceedings of the 28th International Conference on InformationNetworking (ICOIN rsquo14) pp 1ndash6 Phuket Thailand February2014

[18] Z Liao D Li and J Chen ldquoA handshake based orderedscheduling MAC protocol for underwater acoustic local areanetworksrdquo International Journal of Distributed Sensor Networksvol 2015 Article ID 984370 15 pages 2015

[19] N Chirdchoo W-S Soh and K C Chua ldquoMACA-MN aMACA-basedMAC protocol for underwater acoustic networkswith packet train for multiple neighborsrdquo in Proceedings of theIEEE 67th Vehicular Technology Conference-Spring (VTC rsquo08)pp 46ndash50 IEEE Singapore May 2008

[20] S Han Y Noh U Lee and M Gerla ldquoM-FAMA a multi-sessionMACprotocol for reliable underwater acoustic streamsrdquoin Proceedings of the IEEE INFOCOM pp 665ndash673 Turin ItalyApril 2013


[21] L Wen ldquoMR-SFAMA a novel MAC protocol for underwateracoustic sensor networksrdquo in Proceedings of the IEEE Interna-tional Conference on Signal Processing Communications andComputing (ICSPCC rsquo15) pp 1ndash4 Ningbo China September2015

[22] R Jain D Chiu and W Hawe ldquoA quantitative measure offairness and discrimination for resource allocation in sharedcomputer systemsrdquo DEC Research 1984

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014


Active and Passive Electronic Components

Control Scienceand Engineering

Journal of



RotatingMachinery


Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design



Shock and Vibration


Civil EngineeringAdvances in

Acoustics and VibrationAdvances in



Electrical and Computer Engineering

Journal of

Advances inOptoElectronics


Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of


Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014


Chemical EngineeringInternational Journal of Antennas and

Propagation




Navigation and Observation



DistributedSensor Networks



between nodes On the other hand due to the high trans-mission delay the duration of each time slot must be longenough whichwill lead to low network throughputThus thefeasibility of these protocols in UWASNs is unclear [6]

In recent years many competitive access protocols havebeen proposed for UWASNs [7 8] The classical ALOHA-based protocol does not have an effectivemechanism to avoidpacket collisions so it is not suitable for UWASNs [9] In theterrestrial wireless networks the performance of the CSMAprotocol is better than the ALOHA protocol However dueto large propagation delay carrier sense is hard to be realizedin UWASNs [10] Therefore more and more researcherspay attention to the access protocol based on handshakemechanism In the handshake protocol the nodes use thecontrol packet to compete for the use right of channel beforesending data packets Handshake protocols have the problemof hidden terminal and spatial fairness In such protocolsa node schedules its transmissions according to the controlpackets it hearsThese control packets also notify other neigh-bors about the ongoing transmission including the hiddennodes which could reduce collisions significantly Howeverthe handshaking mechanism was not destined for UWASNswhere long propagation delays are prevalent A situation of ahidden terminal problem occurs when a potential interferingnode cannot receive control packets on time and thus thosecontrol packets are unable to inform potential interferingnodes of the forthcoming packet transmissions A defect offairness problemalso existed in the handshake protocol Sincea packetrsquos arrival time is proportional to distance betweenreceiver and transmitter if the transmitter is closer to thereceiver it ismore likely to obtain the right to use the channel

To overcome the hidden terminal problem in UWASNsMolins and Stojanovic introduced the slotted FAMAprotocol[11] In the slotted FAMA protocol time is slotted and anypacket can only be transmitted at the beginning of a slotit is still using control packets to reserve time slots fordata packets Although the protocol can effectively avoidthe problem of hidden terminal however it still has someshortcomings In the slotted FAMA protocol the length oftime slot must be sufficient to ensure that all neighbor nodescan receive control packets in a slot so that they will knowwhether transmitting at the beginning of the next slot willinterfere with an ongoing transmissionThehigh propagationdelaymakes it expensive to exchange Request-To-Send (RTS)and Clear-To-Send (CTS) packets before each data packettransmission After a complete period of control packetsexchanging (two time slots) slotted FAMA only allows onesender-receiver pair access to channel Therefore the longslot length requirement and the inefficient handshakingmechanism affect the throughput and end-to-end delay ofthe UWASNs Finally the protocol does not solve the fairnessproblem

In order to overcome the above problems we propose anew MAC protocol with features of efficiency and fair useof network resources for UWASNs and refer to this MACprotocol as MHM (Multiple Handshaking protocol) MHMcan achieve higher throughput and better fairness by com-bining multiple handshaking mechanism and competitivemechanism of control packets Multiple nodes are allowed

to exchange control packets (RTS or CTS) in the same timeslot so the efficiency of the control packet switching is alsoimproved The multiple handshaking mechanism is develop-ment to avoid collisions of data packets The key idea of thismechanism is to utilize the information of propagation delayto arrange the transmission time of data packets Receivercan receive multiple data packets in a packet train manneras long as multiple senders transmit data packets accordingto the special transmission schedule In our protocol aftersome control packets are exchanged some nodes may needto receive and transmit data packets at the same time Theobjective of competitive mechanism of control packets is tospecify the order in which these nodes transmit and receivecontrol packets In summary in the phase of control packetswitching the node can transmit and receivemultiple controlpackets In the phase of data transmission the node canreceivemultiple data packets thereforeMHMcan reduce thetime during which the control packets occupy the channelimprove the utilization ratio of the channel and ensure betterfairness

The rest of this paper is organized as follow Section 2briefly reviews some related works In Section 3 we presentthe MHM protocol In Section 4 the simulations are carriedout Finally we give our conclusions and further work inSection 5

2 Related Work

There are several existing competitive access protocols forUWASNs Some competitive access protocols use the tradi-tional handshaking mechanism such as MACA MACAUand MACAW However these protocols cannot effectivelysolve the hidden terminal and spatial fairness problem causedby high propagation delay in multihop UWASNsThe FAMAprotocol be proposed in [12] It was shown that using a carriersensing protocol collision avoidance is guaranteed if the fol-lowing conditions hold (1) the length of RTS packets shouldbe greater than the maximum propagation delay and (2) thelength of CTS packets should be greater than the lengthof RTS packets plus twice the maximum propagation delayplus the hardware transmit-to-receive transition time Theseconditions are the basis of the FAMA protocol AlthoughFAMA can ensure the absence of collisions in the networksthe length of control packets becomes excessive on anunderwater acoustic channel Molins and Stojanovic aimedat the hidden terminal problem caused by spatial-temporaluncertainty and proposed the slotted FAMA protocol basedon FAMA [11] Although the slotted FAMA protocol achievesguaranteed collision avoidance for data packets the longslot length requirement and the handshaking mechanismitself affect the throughput In [13] Qian et al proposeda slotted FAMA based MAC protocol for UWASNs calledRC-SFAMA It proposed an RTS competition mechanism toprevent nodes into back-off state caused by the competitionof multiple RTS packets in a slot Via the RTS competitionmechanism useful data transmission can be completed suc-cessfully when the situation of multiple RTS attempts occursIn [14] SF-MAC developed a receiver-based protocol to solve










Node 1

Node 2

Node 3


R

C

D

A

R

CRTS

xRTS

CTS

xCTS

DATA

ACK

Time

R

Time

Time

1

R1

R3

R3

C2

C2

2

D

D

1

1A2

A2

A2

C


Node 1

Time

Node 2

Node 3

Time

Time


R

C

D

A

R

CRTS

xRTS

CTS

xCTS

DATA

ACK

R1

R1

R3

R3

C2

C2

2

D

D

1

1A2

A2

A2

C D3

D3








Node 1

Node 2

Node 3


R

C

D

A

R

CRTS

xRTS

CTS

xCTS

DATA

ACK

Time

RTime

Time

2

R

C3

D2

A3

R1

R1

C3

D2

A32

R2


Time

RTime

Time

2

R

C3

D1

A3

R1

R1

C3

D2

A32

R2

C2

C2

D1

D2

A2

A2

Node 1

Node 2

Node 3


R

C

D

A

R

CRTS

xRTS

CTS

xCTS

DATA

ACK











Time

R

Time

Time

1

R1

R3

R3

C2

C2

C2

D

D

1

1A2

A2

A2

D3

D3

R1

C1

A1

Time

R2

R2

C1

C1

D2

D2

A1

A1

R2

Node 1

Node 2

Node 3

Node 4


R

C

D

A

R

CRTS

xRTS

CTS

xCTS

DATA

ACK








Node 1

Node 2

Node 3

Node 4

R

C

D

A

R

CRTS

xRTS

CTS

xCTS

DATA

ACK

TimeR

Time

1

R2

C1

D2

A2

A2

A2

R1

C1

A1


R1

R2

R2

C1

D2

C2

C2

D1

D1

A1

A1

Time

Time

C2





order sender 119909 Let119863119903119909







When119882119909119896











=

119873minus1

prod1

119890minus120582119879slot = 119890

minus120582(119873minus1)119879slot (2)


119875RC-SFAMA119904

=

119873minus1

sum119894=0

(


prod1

119890minus120582119879slot (1 minus 119890

minus120582119879slot)119894 1

119894 + 1) (3)




119875MHM119904=

119873minus1

sum119894=0

(


prod1

119890minus120582119879slot (1 minus 119890

minus120582119879slot)119894

) = 1 (4)




119878 =119879119906

119879ot (5)




119879119906= 119879data times 119875119904 (6)




where 119863119909119894




119873

sum119894=1

(119879119894


119879RC-SFAMAot = 119879

ℎ+

119873minus1

sum119894=1

(119879119894


(8)



+ (119863min(119909119894) + 119873119879data) mod (119879slot) (9)



119878

=119879data






SFAMAS-ALOHA

MHMRC-SFAMA

0

005

01

015

02

025

03

035

04

045

05

Thro

ughp

ut












SFAMAMHM

RC-SFAMA

0

2

4

6

8

10

12RT

SD

ATA

ratio





FI =(sum1205881198941205881015840119894)2

119899sum (1205881198941205881015840119894)2 (11)

where 120588119894and 1205881015840





SFAMAS-ALOHA

MHMRC-SFAMA

0

01

02

03

04

05

06

07

08

09

Fairn

ess i

ndex



SFAMAS-ALOHA

MHMRC-SFAMA

0

200

400

600

800

1000

1200

1400Av

erag

e del

ay (s

)







Competing Interests


Acknowledgments


References


























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation


















Node 1

Node 2

Node 3


R

C

D

A

R

CRTS

xRTS

CTS

xCTS

DATA

ACK

Time

R

Time

Time

1

R1

R3

R3

C2

C2

2

D

D

1

1A2

A2

A2

C


Node 1

Time

Node 2

Node 3

Time

Time


R

C

D

A

R

CRTS

xRTS

CTS

xCTS

DATA

ACK

R1

R1

R3

R3

C2

C2

2

D

D

1

1A2

A2

A2

C D3

D3








Node 1

Node 2

Node 3


R

C

D

A

R

CRTS

xRTS

CTS

xCTS

DATA

ACK

Time

RTime

Time

2

R

C3

D2

A3

R1

R1

C3

D2

A32

R2


Time

RTime

Time

2

R

C3

D1

A3

R1

R1

C3

D2

A32

R2

C2

C2

D1

D2

A2

A2

Node 1

Node 2

Node 3


R

C

D

A

R

CRTS

xRTS

CTS

xCTS

DATA

ACK











Time

R

Time

Time

1

R1

R3

R3

C2

C2

C2

D

D

1

1A2

A2

A2

D3

D3

R1

C1

A1

Time

R2

R2

C1

C1

D2

D2

A1

A1

R2

Node 1

Node 2

Node 3

Node 4


R

C

D

A

R

CRTS

xRTS

CTS

xCTS

DATA

ACK








Node 1

Node 2

Node 3

Node 4

R

C

D

A

R

CRTS

xRTS

CTS

xCTS

DATA

ACK

TimeR

Time

1

R2

C1

D2

A2

A2

A2

R1

C1

A1


R1

R2

R2

C1

D2

C2

C2

D1

D1

A1

A1

Time

Time

C2





order sender 119909 Let119863119903119909







When119882119909119896











=

119873minus1

prod1

119890minus120582119879slot = 119890

minus120582(119873minus1)119879slot (2)


119875RC-SFAMA119904

=

119873minus1

sum119894=0

(


prod1

119890minus120582119879slot (1 minus 119890

minus120582119879slot)119894 1

119894 + 1) (3)




119875MHM119904=

119873minus1

sum119894=0

(


prod1

119890minus120582119879slot (1 minus 119890

minus120582119879slot)119894

) = 1 (4)




119878 =119879119906

119879ot (5)




119879119906= 119879data times 119875119904 (6)




where 119863119909119894




119873

sum119894=1

(119879119894


119879RC-SFAMAot = 119879

ℎ+

119873minus1

sum119894=1

(119879119894


(8)



+ (119863min(119909119894) + 119873119879data) mod (119879slot) (9)



119878

=119879data






SFAMAS-ALOHA

MHMRC-SFAMA

0

005

01

015

02

025

03

035

04

045

05

Thro

ughp

ut












SFAMAMHM

RC-SFAMA

0

2

4

6

8

10

12RT

SD

ATA

ratio





FI =(sum1205881198941205881015840119894)2

119899sum (1205881198941205881015840119894)2 (11)

where 120588119894and 1205881015840





SFAMAS-ALOHA

MHMRC-SFAMA

0

01

02

03

04

05

06

07

08

09

Fairn

ess i

ndex



SFAMAS-ALOHA

MHMRC-SFAMA

0

200

400

600

800

1000

1200

1400Av

erag

e del

ay (s

)







Competing Interests


Acknowledgments


References


























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










Node 1

Node 2

Node 3


R

C

D

A

R

CRTS

xRTS

CTS

xCTS

DATA

ACK

Time

R

Time

Time

1

R1

R3

R3

C2

C2

2

D

D

1

1A2

A2

A2

C


Node 1

Time

Node 2

Node 3

Time

Time


R

C

D

A

R

CRTS

xRTS

CTS

xCTS

DATA

ACK

R1

R1

R3

R3

C2

C2

2

D

D

1

1A2

A2

A2

C D3

D3








Node 1

Node 2

Node 3


R

C

D

A

R

CRTS

xRTS

CTS

xCTS

DATA

ACK

Time

RTime

Time

2

R

C3

D2

A3

R1

R1

C3

D2

A32

R2


Time

RTime

Time

2

R

C3

D1

A3

R1

R1

C3

D2

A32

R2

C2

C2

D1

D2

A2

A2

Node 1

Node 2

Node 3


R

C

D

A

R

CRTS

xRTS

CTS

xCTS

DATA

ACK











Time

R

Time

Time

1

R1

R3

R3

C2

C2

C2

D

D

1

1A2

A2

A2

D3

D3

R1

C1

A1

Time

R2

R2

C1

C1

D2

D2

A1

A1

R2

Node 1

Node 2

Node 3

Node 4


R

C

D

A

R

CRTS

xRTS

CTS

xCTS

DATA

ACK








Node 1

Node 2

Node 3

Node 4

R

C

D

A

R

CRTS

xRTS

CTS

xCTS

DATA

ACK

TimeR

Time

1

R2

C1

D2

A2

A2

A2

R1

C1

A1


R1

R2

R2

C1

D2

C2

C2

D1

D1

A1

A1

Time

Time

C2





order sender 119909 Let119863119903119909







When119882119909119896











=

119873minus1

prod1

119890minus120582119879slot = 119890

minus120582(119873minus1)119879slot (2)


119875RC-SFAMA119904

=

119873minus1

sum119894=0

(


prod1

119890minus120582119879slot (1 minus 119890

minus120582119879slot)119894 1

119894 + 1) (3)




119875MHM119904=

119873minus1

sum119894=0

(


prod1

119890minus120582119879slot (1 minus 119890

minus120582119879slot)119894

) = 1 (4)




119878 =119879119906

119879ot (5)




119879119906= 119879data times 119875119904 (6)




where 119863119909119894




119873

sum119894=1

(119879119894


119879RC-SFAMAot = 119879

ℎ+

119873minus1

sum119894=1

(119879119894


(8)



+ (119863min(119909119894) + 119873119879data) mod (119879slot) (9)



119878

=119879data






SFAMAS-ALOHA

MHMRC-SFAMA

0

005

01

015

02

025

03

035

04

045

05

Thro

ughp

ut












SFAMAMHM

RC-SFAMA

0

2

4

6

8

10

12RT

SD

ATA

ratio





FI =(sum1205881198941205881015840119894)2

119899sum (1205881198941205881015840119894)2 (11)

where 120588119894and 1205881015840





SFAMAS-ALOHA

MHMRC-SFAMA

0

01

02

03

04

05

06

07

08

09

Fairn

ess i

ndex



SFAMAS-ALOHA

MHMRC-SFAMA

0

200

400

600

800

1000

1200

1400Av

erag

e del

ay (s

)







Competing Interests


Acknowledgments


References


























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










Node 1

Node 2

Node 3


R

C

D

A

R

CRTS

xRTS

CTS

xCTS

DATA

ACK

Time

RTime

Time

2

R

C3

D2

A3

R1

R1

C3

D2

A32

R2


Time

RTime

Time

2

R

C3

D1

A3

R1

R1

C3

D2

A32

R2

C2

C2

D1

D2

A2

A2

Node 1

Node 2

Node 3


R

C

D

A

R

CRTS

xRTS

CTS

xCTS

DATA

ACK











Time

R

Time

Time

1

R1

R3

R3

C2

C2

C2

D

D

1

1A2

A2

A2

D3

D3

R1

C1

A1

Time

R2

R2

C1

C1

D2

D2

A1

A1

R2

Node 1

Node 2

Node 3

Node 4


R

C

D

A

R

CRTS

xRTS

CTS

xCTS

DATA

ACK








Node 1

Node 2

Node 3

Node 4

R

C

D

A

R

CRTS

xRTS

CTS

xCTS

DATA

ACK

TimeR

Time

1

R2

C1

D2

A2

A2

A2

R1

C1

A1


R1

R2

R2

C1

D2

C2

C2

D1

D1

A1

A1

Time

Time

C2





order sender 119909 Let119863119903119909







When119882119909119896











=

119873minus1

prod1

119890minus120582119879slot = 119890

minus120582(119873minus1)119879slot (2)


119875RC-SFAMA119904

=

119873minus1

sum119894=0

(


prod1

119890minus120582119879slot (1 minus 119890

minus120582119879slot)119894 1

119894 + 1) (3)




119875MHM119904=

119873minus1

sum119894=0

(


prod1

119890minus120582119879slot (1 minus 119890

minus120582119879slot)119894

) = 1 (4)




119878 =119879119906

119879ot (5)




119879119906= 119879data times 119875119904 (6)




where 119863119909119894




119873

sum119894=1

(119879119894


119879RC-SFAMAot = 119879

ℎ+

119873minus1

sum119894=1

(119879119894


(8)



+ (119863min(119909119894) + 119873119879data) mod (119879slot) (9)



119878

=119879data






SFAMAS-ALOHA

MHMRC-SFAMA

0

005

01

015

02

025

03

035

04

045

05

Thro

ughp

ut












SFAMAMHM

RC-SFAMA

0

2

4

6

8

10

12RT

SD

ATA

ratio





FI =(sum1205881198941205881015840119894)2

119899sum (1205881198941205881015840119894)2 (11)

where 120588119894and 1205881015840





SFAMAS-ALOHA

MHMRC-SFAMA

0

01

02

03

04

05

06

07

08

09

Fairn

ess i

ndex



SFAMAS-ALOHA

MHMRC-SFAMA

0

200

400

600

800

1000

1200

1400Av

erag

e del

ay (s

)







Competing Interests


Acknowledgments


References


























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










Time

R

Time

Time

1

R1

R3

R3

C2

C2

C2

D

D

1

1A2

A2

A2

D3

D3

R1

C1

A1

Time

R2

R2

C1

C1

D2

D2

A1

A1

R2

Node 1

Node 2

Node 3

Node 4


R

C

D

A

R

CRTS

xRTS

CTS

xCTS

DATA

ACK








Node 1

Node 2

Node 3

Node 4

R

C

D

A

R

CRTS

xRTS

CTS

xCTS

DATA

ACK

TimeR

Time

1

R2

C1

D2

A2

A2

A2

R1

C1

A1


R1

R2

R2

C1

D2

C2

C2

D1

D1

A1

A1

Time

Time

C2





order sender 119909 Let119863119903119909







When119882119909119896











=

119873minus1

prod1

119890minus120582119879slot = 119890

minus120582(119873minus1)119879slot (2)


119875RC-SFAMA119904

=

119873minus1

sum119894=0

(


prod1

119890minus120582119879slot (1 minus 119890

minus120582119879slot)119894 1

119894 + 1) (3)




119875MHM119904=

119873minus1

sum119894=0

(


prod1

119890minus120582119879slot (1 minus 119890

minus120582119879slot)119894

) = 1 (4)




119878 =119879119906

119879ot (5)




119879119906= 119879data times 119875119904 (6)




where 119863119909119894




119873

sum119894=1

(119879119894


119879RC-SFAMAot = 119879

ℎ+

119873minus1

sum119894=1

(119879119894


(8)



+ (119863min(119909119894) + 119873119879data) mod (119879slot) (9)



119878

=119879data






SFAMAS-ALOHA

MHMRC-SFAMA

0

005

01

015

02

025

03

035

04

045

05

Thro

ughp

ut












SFAMAMHM

RC-SFAMA

0

2

4

6

8

10

12RT

SD

ATA

ratio





FI =(sum1205881198941205881015840119894)2

119899sum (1205881198941205881015840119894)2 (11)

where 120588119894and 1205881015840





SFAMAS-ALOHA

MHMRC-SFAMA

0

01

02

03

04

05

06

07

08

09

Fairn

ess i

ndex



SFAMAS-ALOHA

MHMRC-SFAMA

0

200

400

600

800

1000

1200

1400Av

erag

e del

ay (s

)







Competing Interests


Acknowledgments


References


























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










Node 1

Node 2

Node 3

Node 4

R

C

D

A

R

CRTS

xRTS

CTS

xCTS

DATA

ACK

TimeR

Time

1

R2

C1

D2

A2

A2

A2

R1

C1

A1


R1

R2

R2

C1

D2

C2

C2

D1

D1

A1

A1

Time

Time

C2





order sender 119909 Let119863119903119909







When119882119909119896











=

119873minus1

prod1

119890minus120582119879slot = 119890

minus120582(119873minus1)119879slot (2)


119875RC-SFAMA119904

=

119873minus1

sum119894=0

(


prod1

119890minus120582119879slot (1 minus 119890

minus120582119879slot)119894 1

119894 + 1) (3)




119875MHM119904=

119873minus1

sum119894=0

(


prod1

119890minus120582119879slot (1 minus 119890

minus120582119879slot)119894

) = 1 (4)




119878 =119879119906

119879ot (5)




119879119906= 119879data times 119875119904 (6)




where 119863119909119894




119873

sum119894=1

(119879119894


119879RC-SFAMAot = 119879

ℎ+

119873minus1

sum119894=1

(119879119894


(8)



+ (119863min(119909119894) + 119873119879data) mod (119879slot) (9)



119878

=119879data






SFAMAS-ALOHA

MHMRC-SFAMA

0

005

01

015

02

025

03

035

04

045

05

Thro

ughp

ut












SFAMAMHM

RC-SFAMA

0

2

4

6

8

10

12RT

SD

ATA

ratio





FI =(sum1205881198941205881015840119894)2

119899sum (1205881198941205881015840119894)2 (11)

where 120588119894and 1205881015840





SFAMAS-ALOHA

MHMRC-SFAMA

0

01

02

03

04

05

06

07

08

09

Fairn

ess i

ndex



SFAMAS-ALOHA

MHMRC-SFAMA

0

200

400

600

800

1000

1200

1400Av

erag

e del

ay (s

)







Competing Interests


Acknowledgments


References


























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation















=

119873minus1

prod1

119890minus120582119879slot = 119890

minus120582(119873minus1)119879slot (2)


119875RC-SFAMA119904

=

119873minus1

sum119894=0

(


prod1

119890minus120582119879slot (1 minus 119890

minus120582119879slot)119894 1

119894 + 1) (3)




119875MHM119904=

119873minus1

sum119894=0

(


prod1

119890minus120582119879slot (1 minus 119890

minus120582119879slot)119894

) = 1 (4)




119878 =119879119906

119879ot (5)




119879119906= 119879data times 119875119904 (6)




where 119863119909119894




119873

sum119894=1

(119879119894


119879RC-SFAMAot = 119879

ℎ+

119873minus1

sum119894=1

(119879119894


(8)



+ (119863min(119909119894) + 119873119879data) mod (119879slot) (9)



119878

=119879data






SFAMAS-ALOHA

MHMRC-SFAMA

0

005

01

015

02

025

03

035

04

045

05

Thro

ughp

ut












SFAMAMHM

RC-SFAMA

0

2

4

6

8

10

12RT

SD

ATA

ratio





FI =(sum1205881198941205881015840119894)2

119899sum (1205881198941205881015840119894)2 (11)

where 120588119894and 1205881015840





SFAMAS-ALOHA

MHMRC-SFAMA

0

01

02

03

04

05

06

07

08

09

Fairn

ess i

ndex



SFAMAS-ALOHA

MHMRC-SFAMA

0

200

400

600

800

1000

1200

1400Av

erag

e del

ay (s

)







Competing Interests


Acknowledgments


References


























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










SFAMAS-ALOHA

MHMRC-SFAMA

0

005

01

015

02

025

03

035

04

045

05

Thro

ughp

ut












SFAMAMHM

RC-SFAMA

0

2

4

6

8

10

12RT

SD

ATA

ratio





FI =(sum1205881198941205881015840119894)2

119899sum (1205881198941205881015840119894)2 (11)

where 120588119894and 1205881015840





SFAMAS-ALOHA

MHMRC-SFAMA

0

01

02

03

04

05

06

07

08

09

Fairn

ess i

ndex



SFAMAS-ALOHA

MHMRC-SFAMA

0

200

400

600

800

1000

1200

1400Av

erag

e del

ay (s

)







Competing Interests


Acknowledgments


References


























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










SFAMAMHM

RC-SFAMA

0

2

4

6

8

10

12RT

SD

ATA

ratio





FI =(sum1205881198941205881015840119894)2

119899sum (1205881198941205881015840119894)2 (11)

where 120588119894and 1205881015840





SFAMAS-ALOHA

MHMRC-SFAMA

0

01

02

03

04

05

06

07

08

09

Fairn

ess i

ndex



SFAMAS-ALOHA

MHMRC-SFAMA

0

200

400

600

800

1000

1200

1400Av

erag

e del

ay (s

)







Competing Interests


Acknowledgments


References


























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation












Competing Interests


Acknowledgments


References


























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation














RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation











RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation









Documents

Research Article MHM: A Multiple Handshaking MAC …downloads.hindawi.com/journals/ijdsn/2016/9798075.pdf · Research Article MHM: A Multiple Handshaking MAC Protocol for Underwater