Underwater acoustic sensor networks: research challenges · 2005-11-26 · Underwater acoustic sensor networks: research challenges Ian F. Akyildiz *, Dario Pompili, Tommaso Melodia

Ad Hoc Networks 3 (2005) 257–279

www.elsevier.com/locate/adhoc

Underwater acoustic sensor networks: research challenges

Ian F. Akyildiz *, Dario Pompili, Tommaso Melodia

Broadband and Wireless Networking Laboratory, School of Electrical and Computer Engineering,

Georgia Institute of Technology, Atlanta, GA 30332, USA

Received 15 July 2004; received in revised form 20 September 2004; accepted 21 January 2005Available online 2 February 2005

Abstract

Underwater sensor nodes will find applications in oceanographic data collection, pollution monitoring, offshoreexploration, disaster prevention, assisted navigation and tactical surveillance applications. Moreover, unmanned orautonomous underwater vehicles (UUVs, AUVs), equipped with sensors, will enable the exploration of natural under-sea resources and gathering of scientific data in collaborative monitoring missions. Underwater acoustic networking isthe enabling technology for these applications. Underwater networks consist of a variable number of sensors and vehi-cles that are deployed to perform collaborative monitoring tasks over a given area.

In this paper, several fundamental key aspects of underwater acoustic communications are investigated. Different archi-tectures for two-dimensional and three-dimensional underwater sensor networks are discussed, and the characteristics of theunderwater channel are detailed. The main challenges for the development of efficient networking solutions posed by theunderwater environment are detailed and a cross-layer approach to the integration of all communication functionalitiesis suggested. Furthermore, open research issues are discussed and possible solution approaches are outlined.� 2005 Published by Elsevier B.V.

Keywords: Underwater acoustic sensor networks; Underwater networking; Acoustic communications

1. Introduction

Underwater sensor networks are envisioned toenable applications for oceanographic data col-

1570-8705/$ - see front matter � 2005 Published by Elsevier B.V.doi:10.1016/j.adhoc.2005.01.004

* Corresponding author. Tel.: +1 404 894 5141; fax: +1 404894 7883.

E-mail addresses: [email protected] (I.F. Akyildiz), [email protected] (D. Pompili), [email protected] (T.Melodia).

lection, pollution monitoring, offshore explora-tion, disaster prevention, assisted navigationand tactical surveillance applications. Multipleunmanned or autonomous underwater vehicles(UUVs, AUVs), equipped with underwater sen-sors, will also find application in explorationof natural undersea resources and gathering ofscientific data in collaborative monitoring mis-sions. To make these applications viable, thereis a need to enable underwater communications

mailto:[email protected]




258 I.F. Akyildiz et al. / Ad Hoc Networks 3 (2005) 257–279

among underwater devices. Underwater sensornodes and vehicles must possess self-configura-tion capabilities, i.e., they must be able tocoordinate their operation by exchanging config-uration, location and movement information,and to relay monitored data to an onshorestation.

Wireless underwater acoustic networking is theenabling technology for these applications. Under-Water Acoustic Sensor Networks (UW-ASNs)consist of a variable number of sensors andvehicles that are deployed to perform collaborativemonitoring tasks over a given area. To achievethis objective, sensors and vehicles self-organizein an autonomous network which canadapt to the characteristics of the ocean environ-ment [1].

The above described features enable a broadrange of applications for underwater acoustic sen-sor networks:

• Ocean sampling networks. Networks of sensorsand AUVs, such as the Odyssey-class AUVs[2], can perform synoptic, cooperative adaptivesampling of the 3D coastal ocean environment[3]. Experiments such as the Monterey Bayfield experiment [4] demonstrated the advanta-ges of bringing together sophisticated newrobotic vehicles with advanced ocean modelsto improve the ability to observe and pre-dict the characteristics of the oceanic envi-ronment.

• Environmental monitoring. UW-ASNs can per-form pollution monitoring (chemical, biologi-cal and nuclear). For example, it may bepossible to detail the chemical slurry of antibi-otics, estrogen-type hormones and insecticidesto monitor streams, rivers, lakes and oceanbays (water quality in situ analysis) [51]. Moni-toring of ocean currents and winds, improvedweather forecast, detecting climate change,under-standing and predicting the effect ofhuman activities on marine ecosystems, biolog-ical monitoring such as tracking of fishes ormicro-organisms, are other possible applica-tions. For example, in [52], the design and con-struction of a simple underwater sensornetwork is described to detect extreme tempera-

ture gradients (thermoclines), which are consid-ered to be a breeding ground for certain marinemicro-organisms.

• Undersea explorations. Underwater sensor net-works can help detecting underwater oilfieldsor reservoirs, determine routes for laying under-sea cables, and assist in exploration for valuableminerals.

• Disaster prevention. Sensor networks that mea-sure seismic activity from remote locations canprovide tsunami warnings to coastal areas [42],or study the effects of submarine earthquakes(seaquakes).

• Assisted navigation. Sensors can be used to iden-tify hazards on the seabed, locate dangerousrocks or shoals in shallow waters, mooring posi-tions, submerged wrecks, and to performbathymetry profiling.

• Distributed tactical surveillance. AUVs andfixed underwater sensors can collaborativelymonitor areas for surveillance, reconnaissance,targeting and intrusion detection systems. Forexample, in [15], a 3D underwater sensor net-work is designed for a tactical surveillancesystem that is able to detect and classify subma-rines, small delivery vehicles (SDVs) and diversbased on the sensed data from mechanical,radiation, magnetic and acoustic microsensors.With respect to traditional radar/sonar sys-tems, underwater sensor networks can reach ahigher accuracy, and enable detection andclassification of low signature targets by alsocombining measures from different types ofsensors.

• Mine reconnaissance. The simultaneous opera-tion of multiple AUVs with acoustic and opti-cal sensors can be used to perform rapidenvironmental assessment and detect mine-likeobjects.

Underwater networking is a rather unexploredarea although underwater communications havebeen experimented since World War II, when, in1945, an underwater telephone was developed inthe United States to communicate with submarines[39]. Acoustic communications are the typicalphysical layer technology in underwater networks.In fact, radio waves propagate at long distances

I.F. Akyildiz et al. / Ad Hoc Networks 3 (2005) 257–279 259

through conductive sea water only at extra low fre-quencies (30–300 Hz), which require large anten-nae and high transmission power. For example,the Berkeley Mica 2 Motes, the most popularexperimental platform in the sensor networkingcommunity, have been reported to have a trans-mission range of 120 cm in underwater at433 MHz by experiments performed at the Ro-botic Embedded Systems Laboratory (RESL) atthe University of Southern California. Opticalwaves do not suffer from such high attenuationbut are affected by scattering. Moreover, transmis-sion of optical signals requires high precision inpointing the narrow laser beams. Thus, links inunderwater networks are based on acoustic wire-

less communications [45].The traditional approach for ocean-bottom or

ocean-column monitoring is to deploy underwatersensors that record data during the monitoringmission, and then recover the instruments [37].This approach has the following disadvantages:

• No real-time monitoring. The recorded data can-not be accessed until the instruments are recov-ered, which may happen several months afterthe beginning of the monitoring mission. Thisis critical especially in surveillance or in envi-ronmental monitoring applications such as seis-mic monitoring.

• No on-line system reconfiguration. Interactionbetween onshore control systems and the mon-itoring instruments is not possible. Thisimpedes any adaptive tuning of the instruments,nor is it possible to reconfigure the system afterparticular events occur.

• No failure detection. If failures or misconfigura-

tions occur, it may not be possible to detectthem before the instruments are recovered. Thiscan easily lead to the complete failure of a mon-itoring mission.

• Limited storage capacity. The amount of datathat can be recorded during the monitoring mis-sion by every sensor is limited by the capacity ofthe onboard storage devices (memories, harddisks).

Therefore, there is a need to deploy underwaternetworks that will enable real-time monitoring of

selected ocean areas, remote configuration andinteraction with onshore human operators. Thiscan be obtained by connecting underwater instru-ments by means of wireless links based on acousticcommunication.

Many researchers are currently engaged indeveloping networking solutions for terrestrialwireless ad hoc and sensor networks. Althoughthere exist many recently developed network pro-tocols for wireless sensor networks, the uniquecharacteristics of the underwater acoustic commu-nication channel, such as limited bandwidthcapacity and variable delays [38], require very effi-cient and reliable new data communicationprotocols.

Major challenges in the design of underwateracoustic networks are:

• The available bandwidth is severely limited;• The underwater channel is severely im-paired, especially due to multi-path and fad-ing;

• Propagation delay in underwater is five ordersof magnitude higher than in radio frequency(RF) terrestrial channels, and extremelyvariable;

• High bit error rates and temporary losses ofconnectivity (shadow zones) can be experienced,due to the extreme characteristics of the under-water channel;

• Battery power is limited and usually batteriescannot be recharged, also because solar energycannot be exploited;

• Underwater sensors are prone to failuresbecause of fouling and corrosion.

In this survey, we discuss several fundamentalkey aspects of underwater acoustic communica-tions. We discuss the communication architectureof underwater sensor networks as well as the fac-tors that influence underwater network design.The ultimate objective of this paper is to encour-age research efforts to lay down fundamental basisfor the development of new advanced communica-tion techniques for efficient underwater communi-cation and networking for enhanced oceanmonitoring and exploration applications. InTable 3, we report a list of research laboratories


and ongoing research projects related to underwa-ter communications and explorations.

The remainder of this paper is organized asfollows. In Sections 2 and 3 we introduce the com-munication architecture and design challenges,respectively, of underwater acoustic sensornetworks. In Section 4, we investigate the under-water acoustic communication channel andsummarize the associated physical layer challengesfor underwater networking. In Sections 5–9 we dis-cuss physical, data link, network, transport andapplication layer issues in underwater sensor net-works, respectively. In Section 10 we describesome experimental implementations of underwatersensor networks while in Section 11 we draw themain conclusions.

2. Underwater acoustic sensor networks:

communication architecture

In this section, we describe the communicationarchitecture of underwater acoustic sensornetworks. In particular, we introduce referencearchitectures for two-dimensional and three-dimensional underwater networks, and presentseveral types of autonomous underwater vehicles(AUVs) which can enhance the capabilities ofunderwater sensor networks.

The network topology is in general a crucialfactor in determining the energy consumption, thecapacity and the reliability of a network. Hence,the network topology should be carefully engi-neered and post-deployment topology optimization

should be performed, when possible.Underwater monitoring missions can be extre-

mely expensive due to the high cost of underwaterdevices. Hence, it is important that the deployednetwork be highly reliable, so as to avoid failureof monitoring missions due to failure of single ormultiple devices. For example, it is crucial to avoiddesigning the network topology with single pointsof failure that could compromise the overall func-tioning of the network.

The network capacity is also influenced by thenetwork topology. Since the capacity of the under-water channel is severely limited, as will be dis-cussed in Section 4, it is very important to

organize the network topology such a way thatno communication bottleneck is introduced.

The communication architectures introducedhere are used as a basis for discussion of the chal-lenges associated with underwater acoustic sensornetworks. The underwater sensor network topol-ogy is an open research issue in itself that needsfurther analytical and simulative investigationfrom the research community. In the remainderof this section, we discuss the followingarchitectures:

• Static two-dimensional UW-ASNs for ocean bot-tom monitoring. These are constituted by sensornodes that are anchored to the bottom of theocean, as discussed in Section 2.1. Typicalapplications may be environmental monitoring,or monitoring of underwater plates in tectonics[21].

• Static three-dimensional UW-ASNs for ocean-

column monitoring. These include networks ofsensors whose depth can be controlled by meansof techniques discussed in Section 2.2, and maybe used for surveillance applications or moni-toring of ocean phenomena (ocean bio–geo-chemical processes, water streams, pollution).

• Three-dimensional networks of autonomous

underwater vehicles (AUVs). These networksinclude fixed portions composed of anchoredsensors and mobile portions constituted byautonomous vehicles, as detailed in Section 2.3.

2.1. Two-dimensional underwater sensor networks

A reference architecture for two-dimensionalunderwater networks is shown in Fig. 1. A groupof sensor nodes are anchored to the bottom ofthe ocean with deep ocean anchors. Underwatersensor nodes are interconnected to one or moreunderwater sinks (uw-sinks) by means of wirelessacoustic links. Uw-sinks, as shown in Fig. 1, arenetwork devices in charge of relaying data fromthe ocean bottom network to a surface station.To achieve this objective, uw-sinks are equippedwith two acoustic transceivers, namely a vertical

and a horizontal transceiver. The horizontal trans-ceiver is used by the uw-sink to communicate with

Fig. 1. Architecture for 2D underwater sensor networks.


the sensor nodes in order to: (i) send commandsand configuration data to the sensors (uw-sink tosensors); (ii) collect monitored data (sensors touw-sink). The vertical link is used by the uw-sinksto relay data to a surface station. In deep waterapplications, vertical transceivers must be longrange transceivers as the ocean can be as deep as10 km. The surface station is equipped with anacoustic transceiver that is able to handle multipleparallel communications with the deployed uw-sinks. It is also endowed with a long range RFand/or satellite transmitter to communicate withthe onshore sink (os-sink) and/or to a surface sink

(s-sink).Sensors can be connected to uw-sinks via direct

links or through multi-hop paths. In the formercase, each sensor directly sends the gathered datato the selected uw-sink. However, in UW-ASNs,the power necessary to transmit may decay withpowers greater than two of the distance [44], andthe uw-sink may be far from the sensor node.Consequently, although direct link connection isthe simplest way to network sensors, it may notbe the most energy efficient solution. Further-more, direct links are very likely to reduce the net-

work throughput because of increased acousticinterference due to high transmission power. Incase of multi-hop paths, as in terrestrial sensornetworks [10], the data produced by a source sen-sor is relayed by intermediate sensors until itreaches the uw-sink. This may result in energysavings and increased network capacity, but in-creases the complexity of the routing functional-ity. In fact, every network device usually takespart in a collaborative process whose objective isto diffuse topology information such that efficientand loop free routing decisions can be made ateach intermediate node. This process involves sig-naling and computation. Since energy and capacityare precious resources in underwater environments,as discussed above, in UW-ASNs the objective isto deliver event features by exploiting multi-hoppaths and minimizing the signaling overhead neces-sary to construct underwater paths at the sametime.

2.2. Three-dimensional underwater sensor networks

Three dimensional underwater networks areused to detect and observe phenomena that cannot


be adequately observed by means of ocean bottomsensor nodes, i.e., to perform cooperative samplingof the 3D ocean environment. In three-dimen-sional underwater networks, sensor nodes float atdifferent depths in order to observe a given phe-nomenon. One possible solution would be toattach each uw-sensor node to a surface buoy, bymeans of wires whose length can be regulated soas to adjust the depth of each sensor node [15].However, although this solution allows easy andquick deployment of the sensor network, multiplefloating buoys may obstruct ships navigating onthe surface, or they can be easily detected anddeactivated by enemies in military settings. Fur-thermore, floating buoys are vulnerable to weatherand tampering or pilfering.

For these reasons, a different approach can beto anchor sensor devices to the bottom of theocean. In this architecture, depicted in Fig. 2, eachsensor is anchored to the ocean bottom andequipped with a floating buoy that can be inflatedby a pump. The buoy pushes the sensor towardsthe ocean surface. The depth of the sensor can thenbe regulated by adjusting the length of the wire

Fig. 2. Architecture for 3D und

that connects the sensor to the anchor, by meansof an electronically controlled engine that resideson the sensor. A challenge to be addressed in suchan architecture is the effect of ocean currents onthe described mechanism to regulate the depth ofthe sensors.

Many challenges arise with such an architec-ture, that need to be solved in order to enable3D monitoring, including:

• Sensing coverage. Sensors should collabora-tively regulate their depth in order to achieve3D coverage of the ocean column, accordingto their sensing ranges. Hence, it must be possi-ble to obtain sampling of the desired phenome-non at all depths.

• Communication coverage. Since in 3D underwa-ter networks there may be no notion of uw-sink,sensors should be able to relay information tothe surface station via multi-hop paths. Thus,network devices should coordinate their depthsin such a way that the network topology isalways connected, i.e., at least one path fromevery sensor to the surface station always exists.

erwater sensor networks.


Sensing and communication coverage in a 3Denvironment are rigorously investigated in [40].The diameter, minimum and maximum degree ofthe reachability graph that describes the networkare derived as a function of the communicationrange, while different degrees of coverage for the3D environment are characterized as a function ofthe sensing range. These techniques could beexploited to investigate the coverage issues inUW-ASNs.

2.3. Sensor networks with autonomous

underwater vehicles

AUVs can function without tethers, cables, orremote control, and therefore they have a multi-tude of applications in oceanography, environ-mental monitoring, and underwater resourcestudy. Previous experimental work has shown thefeasibility of relatively inexpensive AUV subma-rines equipped with multiple underwater sensorsthat can reach any depth in the ocean [2]. Hence,they can be used to enhance the capabilities ofunderwater sensor networks in many ways. Theintegration and enhancement of fixed sensor net-works with AUVs is an almost unexplored re-search area which requires new networkcoordination algorithms such as:

• Adaptive sampling. This includes control strate-gies to command the mobile vehicles to placeswhere their data will be most useful. Thisapproach is also known as adaptive sampling

and has been proposed in pioneering monitoringmissions such as [4]. For example, the densityof sensor nodes can be adaptively increasedin a given area when a higher sampling rate isneeded for a given monitored phenomenon.

• Self-configuration. This includes control proce-dures to automatically detect connectivity holesdue to node failures or channel impairment andrequest the intervention of an AUV. Further-more, AUVs can either be used for installationand maintenance of the sensor network infra-structure or to deploy new sensors. They canalso be used as temporary relay nodes to restoreconnectivity.

One of the design objectives of AUVs is tomake them rely on local intelligence and lessdependent on communications from online shores[25]. In general, control strategies are needed forautonomous coordination, obstacle avoidanceand steering strategies. Solar energy systems allowincreasing the lifetime of AUVs, i.e., it is not nec-essary to recover and recharge the vehicle on a dai-ly basis. Hence, solar powered AUVs can acquirecontinuous information for periods of time of theorder of months [27].

Several types of AUVs exist as experimentalplatforms for underwater experiments. Some ofthem resemble small-scale submarines (such asthe Odyssey-class AUVs [2] developed at MIT).Others are simpler devices that do not encompasssuch sophisticated capabilities. For example, drift-ers and gliders are oceanographic instruments of-ten used in underwater explorations. Drifterunderwater vehicles drift with local current andhave the ability to move vertically through thewater column. They are used for taking measure-ments at preset depths [24]. Underwater gliders[18] are battery powered autonomous underwatervehicles that use hydraulic pumps to vary their vol-ume by a few hundred cubic centimeters in orderto generate the buoyancy changes that power theirforward gliding. When they emerge on the surface,global positioning system (GPS) is used to locatethe vehicle. This information can be relayed tothe onshore station while operators can interactby sending control information to the gliders.Depth capabilities range from 200 m to 1500 mwhile operating lifetimes range from a few weeksto several months. These long durations are possi-ble because gliders move very slowly, typically25 cm/s (0.5 knots). In [34], a control strategy forgroups of gliders to cooperatively move and recon-figure in response to a sensed distributed environ-ment is presented. The proposed framework allowspreserving the symmetry of the group of gliders.The group is constrained to maintain a uniformdistribution as needed, but is free to spin and pos-sibly wiggle with the current. In [20], results are re-ported on the application of the theory in [34] on afleet of autonomous underwater gliders during theexperiment on Monterey Bay in 2003 [4].

Fig. 3. Internal architecture of an underwater sensor node.


3. Underwater acoustic sensor networks: design

challenges

In this section, we describe the design challengesof underwater acoustic sensor networks. In partic-ular, we itemize the main differences between terres-trial and underwater sensor networks, we detail keydesign issues and deployment challenges for under-water sensors, and we give motivations for a cross-layer design approach to improve the networkefficiency in the critical underwater environment.

3.1. Differences with terrestrial sensor networks

The main differences between terrestrial andunderwater sensor networks are as follows:

• Cost. While terrestrial sensor nodes areexpected to become increasingly inexpensive,underwater sensors are expensive devices. Thisis especially due to the more complex underwa-ter transceivers and to the hardware protectionneeded in the extreme underwater environment.

• Deployment. While terrestrial sensor networksare densely deployed, in underwater, thedeployment is deemed to be more sparse, dueto the cost involved and to the challenges asso-ciated to the deployment itself.

• Power. The power needed for acoustic under-water communications is higher than in terres-trial radio communications due to higherdistances and to more complex signal process-ing at the receivers to compensate for theimpairments of the channel.

• Memory. While terrestrial sensor nodes havevery limited storage capacity, uw-sensors mayneed to be able to do some data caching asthe underwater channel may be intermittent.

• Spatial correlation. While the readings from ter-restrial sensors are often correlated, this is moreunlikely to happen in underwater networks dueto the higher distance among sensors.

3.2. Underwater sensors

The typical internal architecture of an underwa-ter sensor is shown in Fig. 3. It consists of a main

controller/CPU which is interfaced with an ocean-ographic instrument or sensor through a sensorinterface circuitry. The controller receives datafrom the sensor and it can store it in the onboardmemory, process it, and send it to other networkdevices by controlling the acoustic modem. Theelectronics are usually mounted on a frame whichis protected by a PVC housing. Sometimes all sen-sor components are protected by bottom-mountedinstrument frames that are designed to permit azi-muthally omnidirectional acoustic communica-tions, and protect sensors and modems frompotential impact of trawling gear, especially inareas subjected to fishing activities. In [16], theprotecting frame is designed so as to deflect trawl-ing gear on impact, by housing all componentsbeneath a low-profile pyramidal frame.

Underwater sensors include sensors to measurethe quality of water and to study its characteristicssuch as temperature, density, salinity (interfero-metric and refractometric sensors), acidity, chemi-cals, conductivity, pH (magnetoelastic sensors),oxygen (Clark-type electrode), hydrogen, dissolvedmethane gas (METS), and turbidity. Disposablesensors exist that detect ricin, the highly poisonousprotein found in castor beans and thought to be apotential terrorism agent. DNA microarrays canbe used to monitor both abundance and activitylevel variations among natural microbial popula-tions. Other existing underwater sensors includehydrothermal sulfide, silicate, voltammetric sensors


for spectrophotometry, gold-amalgam electrodesensors for sediment measurements of metalions (ion-selective analysis), amperometric micro-sensors for H2S measurements for studies ofanoxygenic photosynthesis, sulfide oxidation, andsulfate reduction of sediments. In addition, force/torque sensors for underwater applications requir-ing simultaneous measurements of several forcesand moments have also been developed, as wellas quantum sensors to measure light radiationand sensors for measurements of harmful algalblooms.

The challenges related to the deployment of lowcost, low scale underwater sensors, are listed asfollows:

• It is necessary to develop less expensive, robust‘‘nano-sensors’’, e.g., sensors based on nano-technology, which involves development ofmaterials and systems at the atomic, molecular,or macromolecular levels in the dimensionrange of approximately 1–500 nm.

• It is necessary to devise periodical cleaningmechanisms against corrosion and fouling,which may impact the lifetime of underwaterdevices. For example, some sensors for pCO2,pH and nitrate measurement, and fluorometersand spectral radiometers, may be limited bybio-fouling, especially on a long time scale.

• There is a need for robust, stable sensors on ahigh range of temperatures since sensor driftof underwater devices may be a concern. To thisend, protocols for in situ calibration of sensorsto improve accuracy and precision of sampleddata must be developed.

• There is a need for new integrated sensors forsynoptic sampling of physical, chemical, andbiological parameters to improve the under-standing of processes in marine systems.

3.3. A Cross-layer protocol stack

A protocol stack for uw-sensors should com-bine power awareness and management, and pro-mote cooperation among the sensor nodes. Itshould consist of physical layer, data link layer,network layer, transport layer, and application

layer functionalities. The protocol stack shouldalso include a power management plane, a coordina-tion plane, and a localization plane. The powermanagement plane is responsible for networkfunctionalities aimed at minimizing the energyconsumption (e.g., sleep modes, power control,etc.). The coordination plane is responsible forall functionalities that require coordination amongsensors (e.g., coordination of the sleep modes, dataaggregation, 3D topology optimization). Thelocalization plane is responsible for providingabsolute or relative localization information tothe sensor node, when needed by the protocolstack or by the application.

While all the research on underwater network-ing so far has followed the traditional layered ap-proach for network design, it is an increasinglyaccepted opinion in the wireless networking com-munity that the improved network efficiency, espe-cially in critical environments, can be obtained witha cross-layer design approach. These techniqueswill entail a joint design of different network func-tionalities, from modem design to MAC and rout-ing, from channel coding and modulation to sourcecompression and transport layer, with the objectiveto overcome the shortcomings of a layered ap-proach that lacks of information sharing acrossprotocol layers, forcing the network to operate ina suboptimal mode. Hence, while in the followingsections for the sake of clarity we present the chal-lenges associated with underwater sensor networksfollowing the traditional layered approach, we be-lieve that the underwater environment particularlyrequires for cross-layer design solutions that allowa more efficient use of the scarce availableresources. However, although we advocate inte-grating functionalities to improve network perfor-mance and to avoid duplication of functions bymeans of cross-layer design, it is important to con-sider the ease of design by following a modular

design approach. This also allows improving andupgrading particular functionalities without theneed to re-design the entire communication system.

Although systematic research on cross-layer de-sign for underwater communications is missing, astudy on the interaction between physical andMAC layers is presented in [29], where a methodis proposed based on the sonar equation [49] to

0200

400600

8001000

0

10

20

30

40

5020

30

40

50

60

70

80

90

Distance [m]Frequency [KHz]

Aco

ustic

Pow

er A

ttenu

atio

n [d

B]

Fig. 4. Path loss of short range shallow UW-A channels vs distance and frequency in band 1–50 kHz.


estimate the battery lifetime and power costfor shallow water underwater acoustic sensornetworks for civilian applications. The battery life-time is modeled as dependent on four key parame-ters, namely internode distance, transmissionfrequency, frequency of data updates and numberof nodes per cluster. Interestingly, since in shallowwater the acoustic propagation loss increases withincreasing frequency and distance (as shown inFig. 4), it is proposed to assign lower frequenciesto sensor nodes that are closer to the sink, sincethey also have to relay data on behalf of more dis-tant nodes. This way, the energy consumption issomehow equalized and the network lifetime isprolonged.

3.4. Real-time vs delay-tolerant networking

As in terrestrial sensor networks, depending onthe application, there may be very differentrequirements for data delivery. For example, sur-veillance application may need very fast reactionto events and thus networking protocols thatprovide guaranteed delay-bounded delivery are re-quired. Hence, it is necessary to develop protocolsthat deal with the characteristics of the underwaterenvironment in order to quickly restore connectiv-

ity when it is lost and that react to unpaired orcongested links by taking appropriate action(e.g., dynamical rerouting) in order to meet thegiven delay bound. Conversely, other applicationsmay produce large bundles of data to be deliveredto the onshore sink without particular delay con-straints. With this respect, the Delay-TolerantNetworking Research Group (DTNRG) [5,19]developed mechanisms to resolve the intermittentconnectivity, long or variable delay, asymmetricdata rates, and high error rates by using a store

and forward mechanism based on a middlewarebetween the application layer and the lower layers.Similar methodologies may be particularly usefulfor applications, such as those that record seismicactivity, that have a very low duty cycle and pro-duce, when activated, large bundles of data thatneed to be relayed to a monitoring station whereit can be analyzed to predict future activity. Onthe other hand, sensor networks intended fordisaster prevention such as those that provideearthquake or tsunami warnings, require immedi-ate delivery of information and hence real-timeprotocols. Therefore, the design of networkingsolutions for underwater acoustic sensor networksshould always be aware of the difference betweenreal-time and delay-tolerant applications, and


jointly tune existing solutions to the applicationneeds and to the characteristics of the underwaterenvironment.

4. Basics of acoustic propagation

Underwater acoustic communications aremainly influenced by path loss, noise, multi-path,Doppler spread, and high and variable propagation

delay. All these factors determine the temporal

and spatial variability of the acoustic channel,and make the available bandwidth of the Under-Water Acoustic channel (UW-A) limited anddramatically dependent on both range and fre-quency. Long-range systems that operate over sev-eral tens of kilometers may have a bandwidth ofonly a few kHz, while a short-range system operat-ing over several tens of meters may have more thana hundred kHz of bandwidth. In both cases thesefactors lead to low bit rate [14], in the order of tensof kbit/s for existing devices.

Underwater acoustic communication links canbe classified according to their range as very long,long, medium, short, and very short links [45]. Table1 shows typical bandwidths of the underwaterchannel for different ranges. Acoustic links are alsoroughly classified as vertical and horizontal,according to the direction of the sound ray with re-spect to the ocean bottom. As will be shown later,their propagation characteristics differ consider-ably, especially with respect to time dispersion,multi-path spreads, and delay variance. In the fol-lowing, as usually done in oceanic literature, shal-low water refers to water with depth lower than100 m, while deep water is used for deeper oceans.

Hereafter we analyze the factors that influenceacoustic communications in order to state the

Table 1Available bandwidth for different ranges in UW-A channels

Range [km] Bandwidth [kHz]

Very long 1000 <1Long 10–100 2–5Medium 1–10 �10Short 0.1–1 20–50Very short <0.1 >100

challenges posed by the underwater channel forunderwater sensor networking. These include:

• Path loss:� Attenuation. Is mainly provoked by absorp-tion due to conversion of acoustic energy intoheat. The attenuation increases with distanceand frequency. Fig. 4 shows the acousticattenuation with varying frequency and dis-tance for a short range shallow water UW-Achannel, according to the propagation modelin [49]. The attenuation is also caused by scat-tering and reverberation (on rough ocean sur-face and bottom), refraction, and dispersion(due to the displacement of the reflectionpoint caused by wind on the surface). Waterdepth plays a key role in determining theattenuation.

� Geometric spreading. This refers to thespreading of sound energy as a result of theexpansion of the wavefronts. It increases withthe propagation distance and is independentof frequency. There are two common kindsof geometric spreading: spherical (omni-direc-tional point source), which characterizes deepwater communications, and cylindrical (hori-zontal radiation only), which characterizesshallow water communications.

• Noise:� Man made noise. This is mainly caused bymachinery noise (pumps, reduction gears,power plants), and shipping activity (hullfouling, animal life on hull, cavitation), espe-cially in areas encumbered with heavy vesseltraffic.

� Ambient noise. Is related to hydrodynamics(movement of water including tides, current,storms, wind, and rain), and to seismic andbiological phenomena. In [23], boat noiseand snapping shrimps have been found tobe the primary sources of noise in shallowwater by means of measurement experimentson the ocean bottom.

• Multi-path:� Multi-path propagation may be responsiblefor severe degradation of the acoustic com-munication signal, since it generates inter-symbol interference (ISI).


� The multi-path geometry depends on the linkconfiguration. Vertical channels are charac-terized by little time dispersion, whereas hor-izontal channels may have extremely longmulti-path spreads.

� The extent of the spreading is a strong func-tion of depth and the distance between trans-mitter and receiver.

• High delay and delay variance:� The propagation speed in the UW-A channelis five orders of magnitude lower than in theradio channel. This large propagation delay(0.67 s/km) can reduce the throughput of thesystem considerably.

� The very high delay variance is even moreharmful for efficient protocol design, as it pre-vents from accurately estimating the roundtrip time (RTT), which is the key parameterfor many common communication protocols.

• Doppler spread:� The Doppler frequency spread can be signif-icant in UW-A channels [45], thus causing adegradation in the performance of digitalcommunications: high data rate transmis-sions cause adjacent symbols to interfereat the receiver. This requires sophisticatedsignal processing to deal with the generatedISI.

� The Doppler spreading generates a simple fre-quency translation, which is relatively easyfor a receiver to compensate for; and a contin-uous spreading of frequencies that constitutesa non-shifted signal, which is more difficult tocompensate for.

� If a channel has a Doppler spread withbandwidth B and a signal has symbol dura-tion T, then there are approximately BT

uncorrelated samples of its complex enve-lope. When BT is much less than unity, thechannel is said to be underspread and theeffects of the Doppler fading can be ignored,while, if greater than unity, it is said to beoverspread [32].

Most of the described factors are caused by thechemical-physical properties of the water mediumsuch as temperature, salinity and density, and bytheir spatio-temporal variations. These variations,

together with the wave guide nature of the chan-nel, cause the acoustic channel to be highly tempo-

rally and spatially variable. In particular, thehorizontal channel is by far more rapidly varyingthan the vertical channel, in both deep and shallowwater.

5. Physical layer

Until the beginning of the last decade, due tothe challenging characteristics of the underwaterchannel, underwater modem development wasbased on non-coherent frequency shift keying(FSK) modulation, since it relies on energy detec-tion. Thus, it does not require phase tracking,which is a very difficult task mainly because ofthe Doppler-spread in the UW-A channel, de-scribed in Section 4. In FSK modulation schemesdeveloped for underwater communications, themulti-path effects are suppressed by inserting timeguards between successive pulses to ensure that thereverberation, caused by the rough ocean surfaceand bottom, vanishes before each subsequentpulse is received. Dynamic frequency guards canalso be used between frequency tones to adaptthe communication to the Doppler spreading ofthe channel. Although non-coherent modulationschemes are characterized by a high power effi-ciency, their low bandwidth efficiency makes themunsuitable for high data rate multiuser networks.Hence, coherent modulation techniques have beendeveloped for long-range, high-throughput sys-tems. In the last years, fully coherent modulationtechniques, such as phase shift keying (PSK) andquadrature amplitude modulation (QAM), havebecome practical due to the availability of power-ful digital processing. Channel equalization tech-niques are exploited to leverage the effect of theinter-symbol interference (ISI), instead of tryingto avoid or suppress it. Decision-feedback equaliz-ers (DFEs) track the complex, relatively slowlyvarying channel response and thus provide highthroughput when the channel is slowly varying.Conversely, when the channel varies faster, it isnecessary to combine the DFE with a PhaseLocked Loop (PLL) [46], which estimates andcompensates for the phase offset in a rapid, stable

Table 2Evolution of modulation technique

Type Year Rate [kbps] Band [kHz] Range [km]a

FSK 1984 1.2 5 3sPSK 1989 500 125 0.06dFSK 1991 1.25 10 2dPSK 1993 0.3–0.5 0.3–1 200d–90sPSK 1994 0.02 20 0.9sFSK 1997 0.6–2.4 5 10d–5sDPSK 1997 20 10 1dPSK 1998 1.67–6.7 2–10 4d–2s16-QAM 2001 40 10 0.3sa The subscripts d and s stand for deep and shallow water.


manner. The use of decision feedback equalizationand phase-locked loops is driven by the complexityand time variability of ocean channel impulse re-sponses. Table 2 presents the evolution from

Table 3Research laboratories and ongoing research projects related to under

Research lab or project name Research area

BWN-Lab @ GeorgiaTech Underwater acoustic

MIT & Woods Hole O.I. Underwater acoustic

Front Project @ UConn Spatial sampling of o

Autonomous Ocean Sampling Networks II Adaptive ocean sampAdaptive Sampling and Prediction (ASAP) Adaptive ocean sampSensor Networks for Undersea Seismic

Experimentation (SNUSE) @ USCUnderwater acoustic

AUV Lab @ MIT Sea Grant AUVsOcean Engineering @ FAU Advanced marine systAOSN Autonomous ocean saAcoustic Research Laboratory (ARL) Underwater Acoustic

ACME Acoustic communicatmonitoring underwateenvironments in costa

Underwater Acoustic Research Group @Loughborough University

Underwater communi

Underwater Technologies Laboratory @Florida Tech

Integrated sustained osystem

Underwater Research Lab�s @ Simon FraserUniversity

Underwater acousticsimaging, bottom andcolumn surveys with Aprocessing and target

Autonomous Undersea Systems Institute(AUSI)

Applications of AUVsensors

non-coherent modems to the recent coherentmodems.

Differential phase shift keying (DPSK) serves asan intermediate solution between incoherent andfully coherent systems in terms of bandwidth effi-ciency. DPSK encodes information relative to theprevious symbol rather than to an arbitrary fixedreference in the signal phase and may be referredto as a partially coherent modulation. While thisstrategy substantially alleviates carrier phase-track-ing requirements, the penalty is an increased errorprobability over PSK at an equivalent data rate.

With respect to Table 2, it is worth noticing thatearly phase-coherent systems achieved higherbandwidth efficiencies (bit rate/occupied band-width) than their incoherent counterparts, butthey did not outperform incoherent modulationschemes yet. In fact, coherent systems had lower

water acoustic sensor networks

URL

sensor networks http://www.ece.gatech.edu/research/labs/bwn/UWASN/

networks http://www.mit.edu/people/millitsa/research.html

cean http://www.nopp.uconn.edu/ADCP/index.html

ling http://www.princeton.edu/dcsl/aosn/ling http://www.princeton.edu/dcsl/asap/sensor networks http://www.isi.edu/ilense/snuse/

http://auvlab.mit.edu/ems http://www.oe.fau.edu/research/ams.htmlmplng networks http://www.mbari.org/aosn/Communications http://www.arl.nus.edu.sg/web/research/

acommsion network forrl areas

http://flipper.ncl.ac.uk/acme/

cations http://sonar-fs.lboro.ac.uk/

cean observing http://my.fit.edu/swood/subsea.html

, sonar, bottomwaterUVs, signaldetection

http://www.ensc.sfu.ca/research/url/

s, platforms and http://www.ausi.org/

http://www.ece.gatech.edu/research/labs/bwn/UWASN/


http://auvlab.mit.edu/


http://www.oe.fau.edu/research/ams.html


http://www.mbari.org/aosn/MontereyBay2003/MontereyBay2003Default.htm

http://www.dtnrg.org

http://www.npaci.edu/DICE/SRB/

http://dx.doi.org/10.1016/j.adhoc.2004.10.008


http://www.mbari.org/aosn/

http://www.arl.nus.edu.sg/web/research/acomms

http://www.arl.nus.edu.sg/web/research/acomms

http://flipper.ncl.ac.uk/acme/

http://sonar-fs.lboro.ac.uk/

http://my.fit.edu/swood/subsea.html

http://www.ensc.sfu.ca/research/url/

http://www.ausi.org/


performance than incoherent systems for long-haul transmissions on horizontal channels untilISI compensation via decision-feedback equalizersfor optimal channel estimation was implemented[47]. However, these filtering algorithms are com-plex and not suitable for real-time communica-tions, as they do not meet real-time constraints.Hence, sub-optimal filters have to be considered,but the imperfect knowledge of the channel im-pulse response that they provide leads to channelestimation errors, and ultimately to decreasedperformance.

Another promising solution for underwatercommunications is the orthogonal frequency divi-sion multiplexing (OFDM) spread spectrum tech-nique, which is particularly efficient when noise isspread over a large portion of the available band-width. OFDM is frequently referred to as multi-carrier modulation because it transmits signalsover multiple sub-carriers simultaneously. In par-ticular, sub-carriers that experience higher SNR,are allotted with a higher number of bits, whereasless bits are allotted to sub-carriers experiencingattenuation, according to the concept of bit load-ing, which requires channel estimation. Since thesymbol duration for each individual carrier in-creases, OFDM systems perform robustly in severemulti-path environments, and achieve a high spec-tral efficiency.

Many of the techniques discussed above requireunderwater channel estimation, which can beachieved by means of probe packets [30]. An accu-rate estimate of the channel can be obtained with ahigh probing rate and/or with a large probe packetsize, which however result in high overhead, and inthe consequent drain of channel capacity andenergy.

5.1. Open research issues

In order to enable physical layer solutions spe-cifically tailored to underwater acoustic sensor net-works, the following open research issues need tobe addressed:

• It is necessary to develop inexpensive trans-mitter/receiver modems for underwatercommunications.

• Research is needed on design of low-complexitysub-optimal filters characterized by rapid con-vergence, to enable real-time underwater com-munications with decreased energy expenditure.

• There is a need to overcome stability problem inthe coupling between the phase locked loop(PLL) and the decision feedback equalizer(DCE).

6. Data link layer

In this section we discuss techniques for multi-ple access in UW-ASNs and present open researchissues to address the requirements of the data linklayer in an underwater environment. Channel ac-cess control in UW-ASNs poses additional chal-lenges due to the peculiarities of the underwaterchannel, in particular limited bandwidth, and highand variable delay.

Frequency division multiple access (FDMA) isnot suitable for UW-ASNs due to the narrowbandwidth in UW-A channels and the vulnerabil-ity of limited band systems to fading and multi-path.

Time division multiple access (TDMA) shows alimited bandwidth efficiency because of the longtime guards required in the UW-A channel. Infact, long time guards must be designed to accountfor the large propagation delay and delay varianceof the underwater channel, discussed in Section 4,in order to minimize packet collisions from adja-cent time slots. Moreover, the variable delaymakes it very challenging to realize a precise syn-chronization, with a common timing reference,which is required for TDMA.

Carrier sense multiple access (CSMA) preventscollisions with the ongoing transmission at thetransmitter side. To prevent collisions at the recei-ver side, however, it is necessary to add a guardtime between transmissions dimensioned accord-ing to the maximum propagation delay in the net-work. This makes the protocol dramaticallyinefficient for UW-ASNs.

The use of contention-based techniques thatrely on handshaking mechanisms such as RTS/CTS in shared medium access (e.g., MACA [31],


IEEE 802.11) is impractical in underwater, for thefollowing reasons: (i) large delays in the propaga-tion of RTS/CTS control packets lead to lowthroughput; (ii) due to the high propagation delayof UW-A channels, when carrier sense is used, asin 802.11, it is more likely that the channel besensed idle while a transmission is ongoing, sincethe signal may not have reached the receiver yet;(iii) the high variability of delay in handshakingpackets makes it impractical to predict the startand finish time of the transmissions of other sta-tions. Thus, collisions are highly likely to occur.

Many novel access schemes have been designedfor terrestrial sensor networks, whose objective,similarly to underwater sensor networks, is to pre-vent collisions in the access channel, thus maximiz-ing the network efficiency. These similarities wouldsuggest to tune and apply those efficient schemes inthe underwater environment; on the other hand,the main focus in medium access control in terres-trial wireless sensor networks is on energy-latencytradeoffs. Some proposed schemes aim at decreas-ing the energy consumption by using sleep sched-ules with virtual clustering. However, thesetechniques may not be suitable for an environmentwhere dense sensor deployment cannot be as-sumed, as discussed in Section 2. Moreover, theadditional challenges due to the underwater chan-nel, such as variable and high propagation delays,and very limited available bandwidth, furthercomplicate the medium access problem in under-water environments.

Code division multiple access (CDMA) is quiterobust to frequency selective fading caused byunderwater multi-paths, since it distinguishessimultaneous signals transmitted by multiple de-vices by means of pseudo-noise codes that are usedfor spreading the user signal over the entire avail-able band. This allows exploiting the time diversityin the UW-A channel by leveraging Rake filters

[43] at the receiver. These filters are designed tomatch the pulse spreading, the pulse shape andthe channel impulse response, so as to compensatefor the effect of multi-path. CDMA allows reduc-ing the number of packet retransmissions, whichresults in decreased battery consumption and in-creased network throughput. For example, in[22], two code-division spread-spectrum access

techniques for underwater communications inshallow water are compared, namely direct se-quence spread spectrum (DSSS) and frequencyhopping spread spectrum (FHSS). AlthoughFHSS is more prone to the Doppler shift effect,since transmissions take place in narrow bands,this scheme is more robust to multiple access inter-ference (MAI) than DSSS. Furthermore, althoughFHSS is shown to lead to a higher bit error ratethan DHSS, it results in simple receivers and pro-vides robustness to the near–far problem, thuspotentially simplifying the power control function-ality. One of the most attractive access techniquesin the recent underwater literature combinesmulti carrier transmission with the DSSS CDMA[30], as it may offer higher spectral efficiencythan its single carrier counterpart, and increasethe flexibility to support integrated high data rateapplications with different quality of servicerequirements. The main idea is to spread each datasymbol in the frequency domain by transmittingall the chips of a spread symbol at the same timeinto a large number of narrow subchannels. Thisway, high data rate can be supported by increasingthe duration of each symbol, which drasticallyreduces ISI.

In conclusion, although the high delay spreadwhich characterizes the horizontal link in under-water channels makes it difficult to maintain syn-chronization among the stations, especially whenorthogonal code techniques are used [30], CDMAis a promising multiple access technique for under-water acoustic networks. This is particularly truein shallow water, where multi-paths and Dopp-ler-spreading play a key role in the communicationperformance.

In [41], a protocol is proposed for networkswith autonomous underwater vehicles. The pro-posed scheme is based on organizing the networkin multiple clusters, each composed of adjacentvehicles. Inside each cluster, TDMA is used withlong band guards, to overcome the effect of prop-agation delay in underwater. In this case, TDMAis not highly inefficient since vehicles in the samecluster are close to one another. Hence, the effectof propagation delay is limited. Interferenceamong different clusters is avoided by assigningdifferent spreading codes to different clusters. The


proposed protocol sketches also some mechanismsto reorganize clusters after node mobility.

In order to meet a required bit error rate at thedata link layer of the deployed underwater sensornetworks, it is mandatory to provide error controlfunctionalities for the transmitted data, since pathloss and multi-path fading affecting UW-A chan-nels lead to high bit error rates (on the order of10�2–10�5 [48,44]). While automatic repeat request

(ARQ) techniques appear not to be suitable forthe underwater environment, because they incur ahigh latency, additional energy cost, and signalingoverhead due to retransmissions; forward errorcorrection (FEC) techniques can be effectivelyemployed in such an environment. The objectiveof these techniques is to protect data by introduc-ing redundant bits in the transmission so that thereceiver can correct detected bit errors. Thisway retransmissions are not necessary althoughboth the transmitter and the receiver incur addi-tional processing power drain for encoding anddecoding, respectively. There is a trade-off betweenthe robustness of the adopted FEC technique,which depends on the amount of redundant bits in-jected in the channel, and the channel efficiency. Apossible solution to maximize the underwater chan-nel efficiency such a way to effectively exploit itsscarce bandwidth would be to dynamically choosethe optimal amount of redundant bits accordingto measurements of the state of the underwaterchannel.


In order to enable data link layer solutions spe-cifically tailored to underwater acoustic sensor net-works, the following open research issues need tobe addressed:

• In case CDMA is adopted, which we stronglyadvocate, it is necessary to design access codeswith high auto-correlation and low cross-corre-lation properties to achieve minimum interfer-ence among users. This needs to be achievedeven when the transmitting and receiving nodesare not synchronized.

• Research on optimal data packet length isneeded to maximize the network efficiency.

• It is necessary to design low-complexity encod-ers and decoders to limit the processing powerrequired for forward error correction (FEC)functionalities. Researchers should evaluatethe feasibility and the energy-efficiency of non-convolutional error control coding schemes.

• Distributed protocols should be devised toreduce the activity of a device when its batteryis depleting without compromising on networkavailability.

7. Network layer

The network layer is in charge of determiningthe path between a source (the sensor that samplesa physical phenomenon) and a destination node(usually the surface station). In general, whilemany impairments of the underwater acousticchannel are adequately addressed at the physicaland data link layers, some other characteristics,such as the extremely long propagation delays,are better addressed at the network layer.

In the last few years there has been an intensivestudy in routing protocols for ad hoc wireless net-works [7] and sensor networks [9]. However, dueto the different nature of the underwater environ-ment and applications, there are several draw-backs with respect to the suitability of theexisting solutions for underwater acoustic net-works. The existing routing protocols are usuallydivided into three categories, namely proactive,reactive and geographical routing protocols:

• Proactive protocols (e.g., DSDV [36], OLSR[26]). These protocols attempt to minimize themessage latency induced by route discovery,by maintaining up-to-date routing informationat all times from each node to every other node.This is obtained by broadcasting control pack-ets that contain routing table information(e.g., distance vectors). These protocols pro-voke a large signaling overhead to establishroutes for the first time and each time the net-work topology is modified because of mobilityor node failures, since updated topology infor-mation has to be propagated to all the nodesin the network. This way, each node is able to


establish a path to any other node in the net-work, which may not be needed in UW-ASNs.For this reason, proactive protocols are notsuitable for underwater networks.

• Reactive protocols (e.g., AODV [35], DSR [28]).A node initiates a route discovery process onlywhen a route to a destination is required. Oncea route has been established, it is maintained bya route maintenance procedure until it is nolonger desired. These protocols are more suit-able for dynamic environments but incur ahigher latency and still require source-initiatedflooding of control packets to establish paths.Thus, both proactive and reactive protocolsincur excessive signaling overhead due to theirextensive reliance on flooding. Reactive proto-cols are deemed to be unsuitable for UW-ASNsas they also cause a high latency in the estab-lishment of paths, which may be even amplifiedunderwater by the slow propagation of acousticsignals. Furthermore, links are likely to beasymmetrical, due to bottom characteristicsand variability in sound speed channel. Hence,protocols that rely on symmetrical links, suchas most of the reactive protocols, are unsuitedfor the underwater environment. Moreover,the topology of UW-ASNs is unlikely to varydynamically on a short time scale.

• Geographical routing protocols (e.g., GFG [12],PTKF [33]). These protocols establish source–destination paths by leveraging localizationinformation, i.e., each node selects its nexthop based on the position of its neighbors andof the destination node. Although these tech-niques are very promising, it is still not clearhow accurate localization information can beobtained in the underwater environment withlimited energy expenditure. In fact, fine-grainedlocalization usually requires strict synchroniza-tion among nodes, which is difficult to achieveunderwater due to the variable propagationdelay. In addition, global positioning system(GPS) receivers, which may be used in terres-trial systems to accurately estimate the geo-graphical location of sensor nodes, do notwork properly underwater. In fact, GPS useswaves in the 1.5 GHz band and those wavesdo not propagate in water.

Some recent papers propose network layer pro-tocols specifically tailored to underwater acousticnetworks. In [50], a routing protocol is proposedthat autonomously establishes the underwater net-work topology, controls network resources andestablishes network flows. The protocol relies ona centralized network manager running on the sur-face station. The manager implements networkmanagement and routing agents that periodicallyprobe the nodes to estimate the channel character-istics. This information is exploited by the man-ager to establish efficient data delivery paths in acentralized fashion, which allows avoiding conges-tion and providing forms of quality of serviceguarantee. The performance evaluation of the pro-posed mechanisms has not been thoroughly car-ried out yet.

In [17], a framework is provided for 3D positionbased routing in ad hoc networks. It is assumedthat each node knows its 3D position and the po-sition of the destination node, and a cell structureis leveraged in order to aggregate the topologicalinformation at each node. Although it is claimedthat the mechanism can be applied to ocean sensornetworks, all the experiments performed assumeradio frequency communications among terrestrialmobile devices.

In [44], it is shown with simple acoustic propa-gation models [13] that multi-hop routing saves en-ergy in underwater networks with respect to singlehop communications, especially with distances inthe order of some kilometers. Based on this, a sim-ple ad hoc underwater network is designed andsimulated, where routes are established by a cen-tral manager based on neighborhood informationgathered by all nodes by means of poll packets.

In general, while most developed protocols forterrestrial ad hoc networks, mostly due to scalabil-ity and mobility concerns, are based on packet

switching, i.e., the routing function is performedseparately for each single packet and paths aredynamically established, virtual circuit routingtechniques can be considered in UW-ASNs. Inthese techniques, paths are established a priori be-tween each source and sink, and each packet fol-lows the same path. This may require some formof centralized coordination, and implies a less flex-ible architecture, but allows exploiting powerful


optimization tools on a centralized manager (e.g.,the surface station) to achieve optimal perfor-mance at the network layer (e.g., minimum delaypaths, energy efficient paths), with minimum com-munication signaling overhead.

Furthermore, routing schemes that account forthe 3D underwater environment need to be de-vised. Especially, in the 3D case the effect of cur-rents should be taken into account, since theintensity and the direction of currents are depen-dent on the depth of the sensor node. Thus, under-water currents can modify the relative position ofsensor devices and also cause connectivity holes,especially when ocean-column monitoring is per-formed in deep waters.


There exist many open research issues for thedevelopment of efficient routing solutions forunderwater acoustic sensor networks, as outlinedbelow:

• There is a need to develop algorithms to pro-vide strict or loose latency bounds for time crit-ical applications. To this respect, it should beconsidered that while the delay for an acousticsignal to propagate from one node to anothermainly depends on the distance of the twonodes, the delay variance also depends on thenature of the link, i.e., the delay variance in hor-izontal acoustic links is generally larger than invertical links due to multi-paths [45].

• For delay-tolerant applications, there is a needto develop mechanisms to handle loss of con-nectivity without provoking immediate retrans-missions. Strict integration with transport anddata link layer mechanisms may be advanta-geous to this end.

• It is necessary to devise routing algorithms thatare robust with respect to the intermittent con-nectivity of acoustic channels. The quality ofacoustic links is highly unpredictable, since itmainly depends on fading and multi-path,which are hard phenomena to model.

• Accurate modeling is needed to better under-stand the dynamics of data transmission at the

network layer. Moreover, credible simulationmodels and tools need to be developed.

• Algorithms and protocols need to be developedthat detect and deal with disconnections due tofailures, unforeseen mobility of nodes or batterydepletion. These solutions should be local so asto avoid communication with the surface sta-tion and global reconfiguration of the network,and should minimize the signaling overhead.

• Local route optimization algorithms are neededto react to consistent variations in the metricsdescribing the energy efficiency of the underwa-ter channel. These variations can be caused byincreased bit error rates due to acoustic noise,or relative displacement of communicatingnodes due to variable currents.

• Mechanisms are needed to integrate AUVs inunderwater networks and to enable commu-nication between sensors and AUVs. In par-ticular, all the information available tosophisticated AUV devices (trajectory, localiza-tion) could be exploited to minimize the signal-ing needed for reconfigurations.

• In case of geographical routing protocols, it isnecessary to devise efficient underwater locationdiscovery techniques.

8. Transport layer

The transport layer of UW-ASNs is a totallyunexplored area. In this section we discuss the fun-damental challenges for the development of an effi-cient reliable transport layer protocol whichaddresses the requirements of UW-ASNs. We alsodiscuss some existing reliable data transport solu-tions for wireless sensor networks, along with theirshortcomings in the underwater environment.

Noticeably, in sensor networks, reliable eventdetection at the sink should be based on collectiveinformation provided by source nodes and not onany individual report from each single source [8].Hence, conventional end-to-end reliability defini-tions and solutions can be inapplicable in theunderwater sensor field, and could lead to wasteof scarce sensor resources. On the other hand,the absence of a reliable transport mechanism alto-


gether can seriously impair event detection due tothe underwater challenges. Thus, the UW-ASNparadigm necessitates a new event transport reli-

ability notion rather than the traditional end-to-end approaches.

A transport layer protocol is needed in UW-ASNs not only to achieve reliable collective trans-

port of event features, but also to perform flow

control and congestion control. The primary objec-tive is to save scarce sensor resources and increasethe network efficiency. A reliable transport proto-col should guarantee that the applications be ableto correctly identify event features estimated bythe sensor network. Congestion control is neededto prevent the network from being congested byexcessive data with respect to the network capac-ity, while flow control is needed to avoid thatnetwork devices with limited memory are over-whelmed by data transmissions.

Most existing TCP implementations areunsuited for the underwater environment, sincethe flow control functionality is based on a win-dow-based mechanism that relies on an accurateesteem of the round trip time (RTT), which istwice the end-to-end delay from source to destina-tion. The underwater RTT can be modeled as astochastic variable with a high mean value, whichreflects the sum of the high delays on the linkscomposing the end-to-end path, and a high delayvariance, which reflects the sum of the high delayvariances on the composing link. This high-mean/high-variance RTT would affect thethroughput of most TCP implementations. Fur-thermore, the high variability of the RTT wouldmake it hard to effectively set the timeout of thewindow-based mechanism that most current TCPimplementations adopt.

Rate-based transport protocols seem alsounsuited for this challenging environment. In fact,although they do not adopt a window-based mech-anism, they still rely on feedback control messagessent back by the destination to dynamically adaptthe transmission rate, i.e., to decrease the transmis-sion rate when packet loss is experienced or to in-crease it otherwise. The high delay and delayvariance can thus cause instability in the feedbackcontrol.

Furthermore, due to the unreliability of theacoustic channel, it is necessary to distinguish be-tween packet losses due to the high bit error rateof the acoustic channel, from those caused bypackets being dropped from the queues of sensornodes due to network congestion. Most TCPimplementations, which are designed for wirednetworks, assume that congestion is the only causefor packet loss. Due to this assumption, when apacket loss occurs, they reduce the transmissionrate to avoid injecting more packets in the net-work. Conversely, in UW-ASNs as in terrestrialwireless networks, it is important to discriminatelosses due to impairments of the channel fromthose caused by congestion. When congestion isthe cause of the packet loss, the transmission rateshould be decreased to avoid overwhelming thenetwork, while in case of losses due to bad channelquality, the transmission rate should not be de-creased to preserve throughput efficiency.

For these reasons, it may be necessary to devisecompletely new strategies to achieve underwaterflow control and reliability.

Several solutions have been proposed to addressthe transport layer problems in terrestrial wirelesssensor networks. For example, in [8], event-to-sinkreliable transport (ESRT) protocol is proposed toachieve reliable event detection with minimum en-ergy expenditure. However, the ESRT mechanismrelies on spatial correlation among event flowswhich may not be easily leveraged in underwateracoustic sensor networks. In fact, in terrestrial sen-sor networks nodes are densely deployed, and thusthe physical readings of spatially close nodes maybe correlated (spatial correlation). Conversely,underwater sensor nodes may be more expensiveand complex devices, and are usually more spar-sely deployed. Hence, correlation among sensorreadings from different sensors may not be signifi-cant in UW-ASNs.

Transport layer functionalities can be tightlyintegrated with data link layer functionalities in across-layer module. The purpose of such an inte-grated module is to make the information aboutthe condition of the variable underwater channelavailable also at the transport layer. In fact, usu-ally the state of the channel is known only at the


physical and channel access sub-layers, while thedesign principle of layer separation makes thisinformation transparent to the higher layers. Thisintegration allows maximizing the efficiency of thetransport functionalities, and the behavior of datalink and transport layer protocols can be dynami-cally adapted to the variability of the underwaterenvironment.


In order to develop a new efficient cross-layerreliable protocol specifically tailored to underwa-ter acoustic sensor networks, the following issuesmust be studied:

• New flow control strategies need to be devisedin order to tackle the high delay and delay var-iance of the control messages sent back by thereceivers.

• New effective mechanisms tailored to the under-water acoustic channel need to be developed, inorder to efficiently infer the cause of packet losses.

• New event transport reliability metric defini-tions need to be proposed, based on the eventmodel and on the underwater acoustic channelmodel.

• Optimal update policies for the sensor reportingrate are needed, to prevent congestion and max-imize the network throughput efficiency as wellas the transport reliability in bandwidth limitedunderwater networks.

• The effects of multiple event occurrences on thereliability and network performance require-ments must be studied, as well as efficient mech-anisms to deal with it.

• It is necessary to statistically model loss of con-nectivity events in order to devise mechanisms,to enable delay-tolerant applications tailoredto the specific underwater requirements.

• Different functionalities at the data link andtransport layer such as channel access, reliabil-ity and flow control, should be jointly designedand studied. A cross-layer approach is highlyrecommended to accordingly optimize thesemechanisms and make them adaptable to the var-iability of the characteristics of the underwaterchannel.

9. Application layer

Although many application areas for underwa-ter sensor networks can be outlined, to the best ofour knowledge the definition of an applicationlayer protocol for UW-ASNs remains largelyunexplored.

The purpose of an application layer is multi-fold: (i) to provide a network management proto-col that makes hardware and software details ofthe lower layers transparent to management appli-cations; (ii) to provide a language for querying thesensor network as a whole; (iii) to assign tasks andto advertise events and data.

No efforts in these areas have been made to datethat address the specific needs of the underwateracoustic environment. A deeper understanding ofthe application areas and of the communicationproblems in underwater sensor networks is crucialto outline some design principles on how to extendor reshape existing application layer protocols [10]for terrestrial sensor networks.

Some of the latest developments in middlewaremay be studied and adapted to realize a versatileapplication layer for underwater sensor networks.For example, the San Diego Supercomputing Cen-ter Storage Resource Broker (SRB) [6,11] is a cli-ent-server middleware that provides a uniforminterface for connecting to heterogeneous data re-sources over a network, and accessing replicateddata sets. SRB provides a way to access data setsand resources based on their attributes and/or log-ical names rather than their names or physicallocations.

10. Implementations of underwater sensor networks

A few experimental implementations of under-water acoustic sensor networks have been reportedin the last few years. In this section we describetwo of them, one mainly concerned with militaryapplications and the other with oceanographicobservations.

The Front-Resolving Observational Networkwith Telemetry (FRONT) project at the Universityof Connecticut relies on acoustic telemetry andranging advances pursued by the US Navy


referred to as ‘‘telesonar’’ technology [16]. TheSeaweb network for FRONT Oceanographic Sen-sors involves telesonar modems deployed in con-junction with three types of nodes, namelysensors, gateways and repeaters. Sensors are ocean-ographic instruments serially connected to anacoustic modem. Gateways are surface buoys thatrelay data from the subsurface network to theshore. Repeaters are acoustic modems that relaydata packets. In the various Seaweb/FRONTexperiments, 20 sensors and repeaters have beendeployed in shallow water (20–60 m deep). Bymeans of long range ocean bottom active sensors,acoustic correlation current profilers (ACCP),sampling of the 3D water column is achieved witha 2D network architecture (Section 2). The net-work enables sensor-to-shore data delivery andshore-to-sensor remote control.

Researchers from different fields gathered at theMonterey Bay Aquarium Research Institute(MBARI) in August 2003 for a month-long exper-iment to quantify gains in predictive skills for prin-cipal circulation trajectories, i.e., to studyupwelling of cold, nutrient-rich water in the Mon-terey Bay. Autonomous vehicles (AUVs, gliders),as well as other ships, vessels and platforms, en-abled unexampled observational capabilities thatare reported on the experiment web site [4]. Exten-sive data are reported that show the variation ofthe characteristics of the circulation of water dur-ing the various days of the experiment.

11. Conclusions

In this paper, we presented an overview of thestate of the art in underwater acoustic sensor net-work. We described the challenges posed by thepeculiarities of the underwater channel with partic-ular reference to monitoring applications for theocean environment. We discussed characteristicsof the underwater channel and outlined future re-search directions for the development of efficientand reliable underwater acoustic sensor networks.The ultimate objective of this paper is to encour-age research efforts to lay down fundamental basisfor the development of new advanced communica-tion techniques for efficient underwater communi-

cation and networking for enhanced oceanmonitoring and exploration applications. Westrongly advocated the use of a cross-layer ap-proach to jointly optimize the main networkingfunctionalities in order to design communicationsuites that are adaptable to the variability of thecharacteristics of the underwater channel and opti-mally exploit the extremely scarce resources.

Acknowledgement

The authors wish to thank Dr. Ozgur B. Akan,Dr. Eylem Ekici, Vehbi C. Gungor and MehmetC. Vuran for their valuable comments that im-proved the quality of this paper. This work wassupported by the Office of Naval Research undercontract N00014-02-1-0564.

References

[1] UnderWater Sensor Networks at BWN Laboratory, Geor-gia Institute of Technology, Available from <http://www.ece.gatech.edu/research/labs/bwn/UWASN/>.

[2] AUV Laboratory at MIT Sea Grant, Available from<http://auvlab.mit.edu/>.

[3] Ocean Engineering at Florida Atlantic University, Avail-able from <http://www.oe.fau.edu/research/ams.html>.

[4] Second field test for the AOSN program, Monterey BayAugust 2003, Available from <http://www.mbari.org/aosn/MontereyBay2003/MontereyBay2003Default.htm>.

[5] The Delay Tolerant Networking Research Group, Avail-able from <http://www.dtnrg.org>.

[6] SDSC Storage Resource Broker, Available from <http://www.npaci.edu/DICE/SRB/>.

[7] M. Abolhasan, T. Wysocki, E. Dutkiewicz, A review ofrouting protocols for mobile ad hoc networks, Ad HocNetworks 2 (1) (2004) 1–22.

[8] O.B. Akan, I.F. Akyildiz, Event-to-sink reliable transportin wireless sensor networks, IEEE/ACM Transactions onNetworking, in press.

[9] K. Akkaya, M. Younis, A survey on routing protocols forwireless sensor networks, Ad Hoc Networks 3 (3) (2005)325–349, in this issue.

[10] I.F. Akyildiz, W. Su, Y. Sankarasubramaniam, E. Cayirci,Wireless sensor networks: A survey, Computer Networks38 (4) (2002) 393–422.

[11] C. Baru, R. Moore, A. Rajasekar, M. Wan, The SDSCstorage resource broker, in: Proceedings of the 1998Conference of the Centre for Advanced Studies onCollaborative Research, 1998.







http://www.dtnrg.org




[12] P. Bose, P. Morin, I. Stojmenovic, J. Urrutia, Routingwith guaranteed delivery in ad hoc wireless networks,ACM Wireless Networks 7 (6) (2001) 609–616.

[13] L. Brekhovskikh, Y. Lysanov, Fundamentals of OceansAcoustics, Springer, New York, 2001.

[14] J. Catipovic, Performance limitations in underwater acous-tic telemetry, IEEE Journal of Oceanic Engineering 15 (3)(1990) 205–216.

[15] E. Cayirci, H. Tezcan, Y. Dogan, V. Coskun, Wirelesssensor networks for underwater surveillance systems, AdHoc Networks, in press; doi:10.1016/j.adhoc.2004.10.008.

[16] D.L. Codiga, J.A. Rice, P.A. Baxley, Networked acousticmodems for real-time data delivery from distributedsubsurface instruments in the coastal ocean: Initial systemdevelopment and performance, Journal of Atmosphericand Oceanic Technology, in press.

[17] S. Das(Bit), R. Ragupathy, Routing in manet and sensornetworks—a 3d position based approach, submitted forpublication.

[18] R. Davis, C. Eriksen, C. Jones, Autonomous buoyancy-driven underwater gliders, in: G. Griffiths (Ed.), TheTechnology and Applications of Autonomous UnderwaterVehicles, Taylor and Francis, London, 2002.

[19] K. Fall. A delay-tolerant network architecture for chal-lenged internets, in: Proceedings of ACM SIGCOMM�03,August 2003.

[20] E. Fiorelli, N. Leonard, P. Bhatta, D. Paley, R. Bachma-yer, D. Fratantoni, Multi-AUV control and adaptivesampling in Monterey Bay, in: Proceedings of IEEEAutonomous Underwater Vehicles 2004: Workshop onMultiple AUV Operations (AUV04), June 2004.

[21] L. Freitag, M. Stojanovic, Acoustic communications forregional undersea observatories, in: Proceedings of Ocean-ology International, London, UK, March 2002.

[22] L. Freitag, M. Stojanovic, S. Singh, M. Johnson, Analysisof channel effects on direct-sequence and frequency-hoppedspread-spectrum acoustic communication, IEEE Journal ofOceanic Engineering 26 (4) (2001) 586–593.

[23] S.A.L. Glegg, R. Pirie, A. LaVigne, A study of ambientnoise in shallow water, Florida Atlantic University Tech-nical Report, 2000.

[24] M. Hinchey, Development of a small autonomous under-water drifter, in: Proceedings of IEEE NECEC�04, October2004.

[25] B.M. Howe, T. McGinnis, H. Kirkham, Sensor networksfor cabled ocean observatories, Geophysical ResearchAbstracts 5 (2003).

[26] P. Jacquet, P. Muhlethaler, T. Clausen, A. Laouiti, A.Qayyum, L. Viennot, Optimized link state routing protocolfor ad hoc networks, in: Proceedings of IEEE INMIC,Pakistan, 2001.

[27] J. Jalbert, D. Blidberg, M. Ageev, Some design consider-ations for a solar powered AUV: Energy management andits impact on operational characteristics, Unmanned Sys-tems 15 (4) (1997) 26–31.

[28] D.B. Johnson, D.A. Maltz, J. Broch, DSR: The dynamicsource routing protocol for multi-hop wireless ad hoc

networks, in: C.E. Perkins (Ed.), Ad Hoc Networking,Addison-Wesley, Reading, MA, 2001, pp. 139–172.

[29] R. Jurdak, C.V. Lopes, P. Baldi, Battery lifetime esti-mation and optimization for underwater sensor networks,in: Sensor Network Operations, IEEE Press, New York,2004.

[30] D.N. Kalofonos, M. Stojanovic, J.G. Proakis, Perfor-mance of adaptive MC-CDMA detectors in rapidly fadingrayleigh channels, IEEE Transactions on Wireless Com-munications 2 (2) (2003) 229–239.

[31] P. Karn, MACA—a new channel access method for packetradio, in: Proceedings of the 9th ARRL ComputerNetworking Conference, London, Ontario, Canada, 1990.

[32] D.B. Kilfoyle, A.B. Baggeroer, The state of the art inunderwater acoustic telemetry, IEEE Journal of OceanicEngineering 25 (1) (2000) 4–27.

[33] T. Melodia, D. Pompili, I.F. Akyildiz, Optimal localtopology knowledge for energy efficient geographicalrouting in sensor networks, in: Proceedings of IEEEINFOCOM�04, Hong Kong SAR, PRC, March 2004.

[34] P. Ogren, E. Fiorelli, N.E. Leonard, Cooperative controlof mobile sensor networks: Adaptive gradient climbing in adistributed environment, IEEE Transactions on AutomaticControl 49 (8) (2004) 1292–1302.

[35] C.E. Perkins, E.M. Belding-Royer, S. Das, Ad hoc ondemand distance vector (AODV) routing, IETF RFC3561.

[36] C.E. Perkins, T.J. Watson, Highly dynamic destinationsequenced distance vector routing (DSDV) for mobilecomputers, in: Proceedings of ACM SIGCOMM�94, Lon-don, UK, 1994.

[37] J.G. Proakis, J.A. Rice, E.M. Sozer, M. Stojanovic,Shallow water acoustic networks, in: J.G. Proakis (Ed.),Encyclopedia of Telecommunications, John Wiley andSons, 2003.

[38] J.G. Proakis, E.M. Sozer, J.A. Rice, M. Stojanovic,Shallow water acoustic networks, IEEE CommunicationsMagazine 39 (11) (2001) 114–119.

[39] A. Quazi, W. Konrad, Underwater acoustic communica-tion, IEEE Communications Magazine 20 (2) (1982)24–30.

[40] V. Ravelomanana, Extremal properties of three-dimen-sional sensor networks with applications, IEEE Transac-tions on Mobile Computing 3 (3) (2004) 246–257.

[41] F. Salva-Garau, M. Stojanovic, Multi-cluster protocol forad hoc mobile underwater acoustic networks, in: Proceed-ings of IEEE OCEANS�03, San Francisco, CA.

[42] N.N. Soreide, C.E. Woody, S.M. Holt, Overview of oceanbased buoys and drifters: Present applications and futureneeds, in: 16th International Conference on InteractiveInformation and Processing Systems (IIPS) for Meteorol-ogy, Oceanography, and Hydrology, January 2004.

[43] E.M. Sozer, J.G. Proakis, M. Stojanovic, J.A. Rice, A.Benson, M. Hatch, Direct sequence spread spectrum basedmodem for underwater acoustic communication andchannel measurements, in: Proceedings of OCEANS�99,November 1999.

http://dx.doi.org/10.1016/j.adhoc.2004.10.008


[44] E.M. Sozer, M. Stojanovic, J.G. Proakis, Underwateracoustic networks, IEEE Journal of Oceanic Engineering25 (1) (2000) 72–83.

[45] M. Stojanovic, Acoustic (underwater) communications, in:J.G. Proakis (Ed.), Encyclopedia of Telecommunications,Wiley, New York, 2003.

[46] M. Stojanovic, J. Catipovic, J.G. Proakis, Phase coherentdigital communications for underwater acoustic channels,IEEE Journal of Oceanic Engineering 19 (1) (1994) 100–111.

[47] M. Stojanovic, J.G. Proakis, J. Catipovic, Analysis of theimpact of channel estimation errors on the performance ofa decision feedback equalizer in multipath fading channels,IEEE Transactions on Communications 43 (2/3/4) (1995)877–886.

[48] M. Stojanovic, J.G. Proakis, J. Catipovic, Performance ofhigh-rate adaptive equalization on a shallow water acousticchannel, Journal of the Acoustical Society of America 100(4) (1996) 2213–2219.

[49] R.J. Urick, Principles of Underwater Sound, McGraw-Hill, 1983.

[50] G. Xie, J.H. Gibson, A network layer protocol for UANsto address propagation delay induced performance limita-tions, in: Proceedings of IEEE OCEANS�01, vol. 4,Honolulu, HI, November 2001, pp. 2087–2094.

[51] X. Yang, K.G. Ong, W.R. Dreschel, K. Zeng, C.S.Mungle, C.A. Grimes, Design of a wireless sensor networkfor long-term, in-situ monitoring of an aqueous environ-ment, Sensors 2 (2002) 455–472.

[52] B. Zhang, G.S. Sukhatme, A.A. Requicha, Adaptivesampling for marine microorganism monitoring, in:IEEE/RSJ International Conference on Intelligent Robotsand Systems, 2004.

Ian F. Akyildiz is the Ken Byers Dis-tinguished Chair Professor with theSchool of Electrical and ComputerEngineering, Georgia Institute ofTechnology and Director of Broad-band and Wireless Networking Labo-ratory. He is the Editor-in-Chief ofComputer Networks (Elsevier) and AdHoc Networks (Elsevier) Journal. He isan IEEE FELLOW (1995), an ACMFELLOW (1996). He served as aNational Lecturer for ACM from 1989until 1998 and received the ACM

Outstanding Distinguished Lecturer Award for 1994. Hereceived the 1997 IEEE Leonard G. Abraham Prize award(IEEE Communications Society) for his paper entitled ‘‘Mul-timedia Group Synchronization Protocols for Integrated Ser-vices Architectures’’ published in the IEEE Journal of SelectedAreas in Communications (JSAC) in January 1996. He receivedthe 2002 IEEE Harry M. Goode Memorial award (IEEEComputer Society) with the citation ‘‘for significant and pio-neering contributions to advanced architectures and protocolsfor wireless and satellite networking’’. He received the 2003IEEE Best Tutorial Award (IEEE Communicaton Society) forhis paper entitled ‘‘A Survey on Sensor Networks’’, publishedin IEEE Communication Magazine, in August 2002. Hereceived the 2003 ACM SIGMOBILE award for his significantcontributions to mobile computing and wireless networking.His current research interests are in Sensor Networks, Inter-PlaNetary Internet, Wireless Networks and Satellite Networks.

Dario Pompili received the ‘‘Laurea’’degree in Telecommunications Engi-neering in 2001, magna cum laude,from the University of Rome ‘‘LaSapienza’’. From June 2001 he hasbeen working at the same university onthe European Union IST Brahms andSatip6 projects. In 2004 he earned fromthe University of Rome ‘‘La Sapienza’’the Ph.D. degree in System Engineer-ing. In 2003 he worked on SensorNetworks at the Broadband andWireless Networking Laboratory,

Georgia Institute of Technology, Atlanta, as a visiting
researcher. Currently he is pursuing the Ph.D. degree in Elec-trical Engineering at the Georgia Institute of Technology. Hismain research interests are in Wireless Sensor Networks,Underwater Acoustic Sensor Networks, and Satellite Networks.
Tommaso Melodia received the ‘‘Lau-rea’’ degree in TelecommunicationsEngineering from the University ‘‘LaSapienza’’, Rome, Italy, in 2001. Hethen worked on a national researchproject on Mobile Networking andWireless Personal Area Networks atthe same University. He is currentlypursuing his Ph.D. and working as aresearch assistant at the Broadbandand Wireless Networking Laboratory,Georgia Institute of Technology,Atlanta. His main research interests are

in Wireless Ad Hoc and Sensor Networks, Wireless Sensor and
Actor Networks, Underwater Acoustic Sensor Networks, Per-sonal and Mobile Communications.

Documents

Underwater acoustic sensor networks: research challenges · 2005-11-26 · Underwater acoustic sensor networks: research challenges Ian F. Akyildiz *, Dario Pompili, Tommaso Melodia