D__published papers_DSTC2014_ORTHOGONAL FREQUENCY DIVITION MULTIPLEXING (UNDERWATER ACOUSTIC COMMUNICATIONS)

Abstract— The aim of this paper is to consider an encoding digitaltechnique concerning the use of orthogonal division multiplexing(OFDM) and studied its properties, The primary advantage ofOFDM over single-carrier schemes is its ability to cope with severechannel conditions (for example, attenuation of high frequencies ina long copper wire, narrowband interference andfrequency-selective fading due to multipath) without complexequalization filters, some of advantages and limitations. Alsoconsidered underwater acoustic communications UnderwaterAcoustic Sensor Networks will consist of sensors and vehiclesdeployed underwater and networked via acoustic links to performcollaborative monitoring tasks, The shallow water acousticcommunication channel exhibits a long delay spread because ofnumerous multipath arrivals resulting from surface and bottominteractions, also considered a direct sequence spread spectrum(DSSS) as one of the methods used in underwater acoustic networkcommunications (UAN).

Index Terms— AUV; DFE; DMT; DSSS; HR; ICI; OFDM;UAN; UAC.

I. INTRODUCTION

rthogonal frequency-divisionmultiplexing (OFDM) is amethod of encoding digital data on multiple carrier

frequencies. OFDM has developed into a popular scheme forwideband digital communication, whether wireless or overcopper wires, used in applications such as digital television andaudio broadcasting, DSL broadband internet access, wirelessnetworks, and 4G mobile communications [1].

OFDM is essentially identical to coded OFDM (COFDM),discrete multi-tone modulation (DMT), and is afrequency-division multiplexing (FDM) scheme used as adigital multi-carrier modulation method. A large number ofclosely spaced orthogonal sub-carrier signals are used to carrydata. The data is divided into several parallel data streamsorchannels, one

for each sub-carrier. Each sub-carrier is modulated with aconventional modulation scheme (such as quadrature amplitudemodulation or phase-shift keying) at a low symbol rate,maintaining total data rates similar to conventionalsingle-carrier modulation schemes in the same bandwidth.The use of direct sequence code division multiple access(DS-CDMA) has some benefits such as multi-user access andlow probability of detection (LPD). In Stojanovic and Freitag(2006), the authors explored the use of a DFE to combat ISI inDS-CDMA systems. A symbol decisions feedback (SDF) DFEuses the symbol decisions after de-spreading on the feedbackpath. As the symbols in a DS-CDMA system are relatively long,a DFE using SDF is not able to track rapidly varying channels.For such channels, a chip hypothesis feedback (CHF) can helptrack the channel at the chip rate rather than the symbol rate.For M-ary signal constellations, the complexity of the CHF is atleast M times higher than the SFD as M different hypotheseshave to be tracked. However, the advantage of a CHF wasclearly demonstrated at high, spreading factors during ashallow water experiment in Italy. The authors expect that theperformance difference would be more apparent in a mobileenvironment as one would expect the channel to change morerapidly. If the statistics of the errors in channel estimation areknown, DFE or linear equalizer performance can be estimated(Preisig 2005).

0 100 200 300 400-1

-0.5

0

0.5

1OFDM Symbol Input Data

0 100 200 300 400-1

-0.5

0

0.5

1OFDM Recovered Symbols

0 2000 4000 6000 8000-40

-20

0

20

40Transmitted OFDM

0 2000 4000 6000 8000-40

-20

0

20

40Received OFDM

Fig. 1. OFDM signal generation and recovering in Time and Frequencydomain respectively.

OC

Orthogonal Frequency Division Multiplexing

(Underwater Acoustic Communications)

Dheyauldeen Najm Abdulameer1,a, Prof. Dr. R. Badlishah Ahmad2,b

School of Computer and Communication Engineering (SCCE)University Malaysia Perlis (UniMAP)

Kuala Perlis, Perlis, [email protected]@unimap.edu.my

The primary advantage of OFDM over single-carrier schemesis its ability to cope with severe channel conditions (forexample, attenuation of high frequencies in a long copper wire,narrowband interference and frequency-selective fading due tomultipath) without complex equalization filters. Channelequalization is simplified because OFDM may be viewed asusing many slowly modulated narrowband signals rather thanone rapidly modulated wideband signal. The low symbol ratemakes the use of a guard interval between symbols affordable,making it possible to eliminate intersymbol interference (ISI)and utilize echoes and time-spreading to achieve a diversitygain, i.e., a signal-to-noise ratio improvement. This mechanismalso facilitates the design of single frequency networks (SFNs),where several adjacent transmitters send the same signalsimultaneously at the same frequency, as the signals frommultiple distant transmitters may be combined constructively,rather than interfering as would typically occur in a traditionalsingle-carrier system.

II. ORTHOGONALITY

In OFDM, the sub-carrier frequencies are chosen so that thesub-carriers are orthogonal to each other, meaning thatcrosstalk between the sub-channels is eliminated andinter-carrier guard bands are not required. This greatlysimplifies the design of both the transmitter and the receiver;unlike conventional FDM, a separate filter for each sub-channelis not required.

The orthogonality requires that the sub-carrier spacing islinespacing.

(1)

whereTU seconds are the useful symbol duration (the receiverside window size), andk is a positive integer, typically equal to1. Therefore, with N sub-carriers, the total passband bandwidthwill be:

B≈N.Δf (Hz) (2)

The orthogonality also allows high spectral efficiency, with atotal symbol rate near the Nyquist rate for the equivalentbaseband signal (i.e. near half the Nyquist rate for thedouble-side band physical passband signal). Almost the wholeavailable frequency band can be utilized. OFDM generally hasa nearly 'white' spectrum, giving it benign electromagneticinterference properties with respect to other co-channel users.

-5 0 5 10 15 20-0.4

-0.2

0

0.2

0.4

0.6

0.8

1

Frequency (K Hz)

S(f)

OFDM spectrum which consists of N equivalent distant sinc function

Fig. 2. The spectrum of an OFDM signal that shows its subcarriers areorthogonal to each other

OFDM requires very accurate frequency synchronizationbetween the receiver and the transmitter; with frequencydeviation the sub-carriers will no longer be orthogonal, causinginter-carrier interference (ICI) (i.e., Crosstalk between thesub-carriers). Frequency offsets are typically caused bymismatched transmitter and receiver oscillators, or by Dopplershift due to movement. While the Doppler shift alone may becompensated for by the receiver, the situation is worsenedwhen combined with multipath, as reflections will appear atvarious frequency offsets, which is much harder to correct.Thiseffect typically worsens as speed increases, and is an importantfactor limiting the use of OFDM in high-speed vehicles.Several techniques for ICI suppression are suggested, but theymay increase the receiver complexity.

III. U NDERWATERCOMMUNICATIONS

High-speed communication in the underwater acousticchannel has been challenging because of limited bandwidth,extended multipath, refractive properties of the medium, severefading, rapid time variation and large Doppler shifts. In theinitial years, rapid progress was made in deep watercommunication, but the shallow water channel was considereddifficult. In the past decade, significant advances have beenmade in shallow water communication [2].

0 2 4 6 8 10 12 140

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1Multipath Phenomenon in Underwater Acoustic Waves

Time (sec)

Am

plitu

de (

Vol

t)

Multipath in time domain

Fig. 3. Multipath Phenomenon in Underwater Acoustic Signals

The shallow water acoustic communication channel exhibitsalong delay spread because of numerous multipath arrivalsresulting from surface and bottom interactions. Movement oftransducers, ocean surface, internal waves lead to rapid timevariability and, consequently, a high Doppler spread in thechannel. Coherent modulation schemes such as phase shiftkeying (PSK) along with adaptive decision feedback equalizers(DFE) and spatial diversity combining have been shown to bean effective way of communication in such channels(Stojanovic et al., 1993). However, the long delay spread (oftenhundreds of symbols) and rapid time variation of the channeloften makes this approach computationally too complex forreal-time implementations.

Fig. 4. Underwater Acoustic Network (UAN)

IV. Channel Equalization

The error estimate can be split into the minimum achievableerror and the excess error. The excess error component isstrongly affected by rough sea conditions. Through a scatteringfunction analysis, it was also shown that the rate of change ofthe propagation path length of the surface bounced arrival is aprimary contributor to the error. This suggests that the ability toeffectively track the surface bounced arrival may provide animproved equalizer performance.

0.5 1 1.5 2 2.5 3 3.50

2000

4000

6000

8000

10000

Original Audio Signal

Time

Frequ

ency

(Hz)

-140

-120

-100

-80

-60

-40

0.5 1 1.5 2 2.5 3 3.50

2000

4000

6000

8000

10000

Audio Signal with Flanging

Time

Frequ

ency

(Hz)

-140

-120

-100

-80

-60

-40

-20

Fig. 4. The Acoustic signal that shows the effect of signal with/withoutFlanging

The demand for high quality underwater acousticcommunications (UAC) arises in many military, scientific andcivilian applications. Many of these involve the wirelesstransmission of controlling signals and commands toautonomous underwater vehicles (AUVs) and underwatersensors. On the other hand, AUV and sensor receivers have thelimited signal processing capability and only a fewhydrophones due to the size and power limitations. All thesemake reliable communications a challenging problem in thenaturally doubly-selective UAC channels. DSSS systems arerecently introduced to underwater communications becauseoftheir capability of resolving multipath and enabling thecollection of delay diversity and channel energy. Similar tothese existing schemes, our proposed approach also has verylow receiver complexity requiring only matched filter operation.However, different from them, we simultaneously transmitmultiple symbols during each sequence period. Compared withexisting underwater DSSS schemes, our proposed approachrequires shorter channel coherence time and is thus more robustagainst moderate channel variation that is inevitable inunderwater scenarios. In addition, our high reliability (HR-)DSSS scheme also facilitates higher and more flexible rates.More importantly, the high reliability and high data rate areachieved with negligible self- and co-channel interference [3].The research on Underwater Acoustic Networks (UANs) isattracting attention due to their important underwaterapplications for military and commercial purposes. More andmore research interest and efforts are shifting to this areainrecent years. The broad applications of UANs include, but notlimited to [4]:

1. Informationexchangeamongnodesthat arewithin therange of the network, or outside the network with thehelp of, e.g., a gateway, or a switching center. Theprimary design goal of communication networks is forexchanging information. In an UAN, exchanginginformation among nodes is one of its essentialapplications. An example is that underwater Internet,in which users can share information without a tether,will become realistic instead of just a dream, if UANsare deployed. Another important application is realtime communication with submarines andautonomous underwater vehicles in networkconfigurations.

2. Informationcollectionfor oceans,lakes,andrivers.Forexample, synoptic and cooperative adaptive samplingof 3D coastal ocean environment was performed byOdyssey-class AUVs. Such kind of activities couldimprove the human ability to observe and predict thecharacteristics of ocean/lake/river environment.

3. Surveillance.It includessurveillance,reconnaissance,targeting, and intrusion detection. By using differenttypes of sensors, an UAN can achieve more accurateand classification of low signature targets comparedwith traditional surveillance systems.

4. Underwaterexplorations.Underwaterexplorationsaredifficult for human beings due to the high waterpressure, unpredictable underwater activities and vastsize of unknown area. UANs can help us explore theunderwater world that we are not familiar with.

5. Disaster prevention. By deploying Acoustic SensorNetworks in remote locations to monitor underseaactivities, ocean-related disaster like a tsunami andsequence can be warned to coastal areas in real timewhen it happens.

6. Compared with the speed of electromagnetic waves,acoustic signal travels much slower in salty water(Approximately 1500 m/s, which is 2x105 lower thanthe electromagnetic wave counterpart [5]). Thiscauses another big issue: very long propagation delay.Additionally, Doppler shift has more significantimpacts subjects to the low velocity of acousticpropagation in water [6].

V. HIGH RELIABILITY (HR) -DSSS SCHEMEA. Transmitted SignalsIn traditional DSSS systems, a single symbol is modulated perspreading sequence. In a vector form, theith modulated signalblock is:

(3)where s(i) represents thei th transmitted symbol andc thespreading sequence.In such a DSSS system was introduced to underwatercommunications, where the spreading code is the maximumlength sequence (m-sequence). Here we also employ them-sequence as the spreading code.However, instead of transmitting one symbol per block as in (1),our HR-DSSS scheme simultaneously modulates multiplesymbols on circularly shifted versions of a m-sequence duringeach block. Define the circular shift matrix as:

which introduces a circular shift by 1 upon pre-multiplyingan(M × 1) vector. Accordingly, vectorTjc is the circularly shiftedm-sequence ofj chips. Note that a m-sequence and its circularlyshifted version have the following autocorrelation property:

(5)

Hence, in a flat-fading channel, distinct symbols riding onc andTmc will induce negligible interferences among themselves, aslong as the circular shiftm ≥ 1. However, UAC channels arewell known to have extensive multipath. Letτmax denote themaximum delay spread andTc as the chip duration.The multipath essentially spreads over (L + 1) chips, where

In order to separate the delayed multipath components ofneighboring symbols, the circular shift between the sequencesconveying adjacent symbols should be at least (L+1) chips.Hence, the transmitted signal block in our HRDSSS is given by:

c, (7)

wheres(i; j) is the jth transmitted symbol during theith block,and theJ is the number of superimposed sequences, at most

It is also worth mentioning thatJmax is the maximum number ofsymbols that can be simultaneously transmitted when thedistribution and strength of the actual (and possibly sparse)channel taps are not available at the transmitter. If thisinformation is also available, then it is possible to increaseJmax

by smartly scheduling the signals. In addition, in a very lowratesystem, one can also choose to transmitJ (2 ≤ J ≤ Jmax)symbols to further reduce the inter-symbol interference. Notethat in this design, the only information about the channelneeded at the transmitter is the channel delay spread or an upperbound on its.

The underwater environment differs from the terrestrial radioenvironment, both in terms of its energy costs and channelpropagation phenomena. The underwater channel ischaracterized by long propagation times andfrequency-dependent attenuation that is highly affected by thedistance between nodes as well as by the link orientation. Someof other issues in which UWSNs differ from terrestrial arelimited bandwidth, constrained battery power, more failure ofsensors because of fouling and corrosion, etc. This paperpresents several fundamental key aspects and architectures ofUWSNs, emerging research issues of underwater sensornetworks and exposes the researchers into the networking ofunderwater communication devices for exciting oceanmonitoring and exploration applications [7].

Multicarrier modulation: The idea of multicarrier modulationis to divide the available bandwidth into a large number ofoverlapping sub bands, so that the waveform duration for thesymbol at each sub band is long compared to the multipathspread of the channel [8], [9]. Consequently, inter-symbolinterference may be neglected in each sub band, greatlysimplifying the receiver complexity of channel equalization.Precisely due to this advantage, multicarrier modulation in theform of orthogonal frequency division multiplexing (OFDM)has prevailed in recent broadband wireless radio applications.However, underwater channels entail large Doppler spreadwhich introduces significant interference among OFDMsubcarriers. Lacking effective techniques to suppress theintercarrier interference (ICI), early attempts at applyingOFDM to underwater environments had a very limited success.

VI . Challenges in underwater acoustic sensor networksMajor challenges encountered in the design of underwateracoustic networks are as follows:1. The available bandwidth is severely limited.2. The underwater channel is impaired because of multi-pathand fading.

3. Propagation delay in underwater is five orders ofmagnitude higher than in Radio Frequency (RF) terrestrialchannels and variable.

4. High bit error rates and temporary losses of connectivity(shadow zones) can be experienced.

5. Underwater sensors are characterized by high cost becauseof extra protective sheaths needed for sensors and also

relatively small number of suppliers (i.e., not much economy ofscale) are available.

6. Battery power is limited and usually batteries cannot berecharged as solar energy cannot be exploited.7. Underwater sensors are more prone to failures because offouling and corrosion.

The differences between these are as follows:1. Communication method: Terrestrial sensor networks

employ electromagnetic waves, but in underwaternetworks because of the characteristic (large delay, thelong distance of communication) of the network, thecommunication relies on physical means like acousticsounds to transmit the signal. Traditional RF networksmight not work efficiently in underwater networks.

2. Protocols: Due to distinct network dynamics, existingcommunication protocols for pretrial networks are notsuitable for the underwater environment. Lowbandwidth and large latency result in long end to enddelays, and this brings in challenges in reliable datatransfer and traffic congestion control.

Cost: Terrestrial networks are becoming inexpensive due toadvancement in technology, but underwater sensors are stillexpensive devices. This is due to the extra protection requiredfor underwater environment and more complex transceiversneeded.

VII. Differences between underwater sensor networks andterrestrial networks

Underwater sensor networks are quite different from terrestrialsensor networks.

Table (I)Comparison of Acoustic, EM and Optical Waves in Sea EnvironmentParameter Acoustic EM OPTICALNominalspeed

1500 m/s 33,333,333 m/s 33,333,333 m/s

Power loss ˃0.1dB/m/Hz 28dB/1km/100MHz

Prop toturbidity

Bandwidth KHz MHz 10-150MHz

Frequencyband

KHz MHz 1014 -1013

HzAntennasize

0.1m 0.5m 0.1m

Effectiverange

Km 10m 10-100m

The table above represents the approximate values of nominalspeed, power loss, bandwidth, frequency band, antenna sizeand effective range [10].

VIII. Conclusion

In this paper, we considered OFDM as one of the mostimportant applications for military, scientific and public use;also we considered the underwater acoustic communicationswhich have high speed communication, but limited bandwidth,extended multipath, refractive properties of the medium, rapidtime variation and large Doppler shifts also considered directsequence (DSSS) as one of the methods used in underwatercommunication techniques. A brief comparison was given in atable between underwater acoustic network and terrestrialnetworks.

References

[1] Hermann Rohling, “Concept for future communication system",Springer-Verlag Berlin Heidelberg 2011.[2] Mandar Chitre, Shiraz Shahabudeen,"Underwater AcousticCommunications and Networking: Recent Advances and FutureChallenges",Marine Technology Society114 Society Journal, Spring 2008.[3] Fengzhong Qu, Liuqing Yang and T. C. Yang, "High ReliabilityDirect-Sequence Spread Spectrum for Underwater Acoustic Communication".[4] Zaihan Jiang, "Underwater Acoustic Networks – Issues and Solutions",INTERNATIONAL JOURNAL OF INTELLIGENT CONTROL ANDSYSTEMS VOL. 13, NO. 3, SEPTEMBER 2008,152-161[5] R. J. Urick, “Principles of Underwater Sound”, 3rd Edition, McGraw-HillPublishing Company, New York, NY, 1983.[6] Y. R. Zheng, “ Channel Estimation and Phase-Correction for RobustUnderwater Acoustic Communications,”IEEE Milcom 2007, pp. 1-6, Orlando,Fl., USA, Oct., 2007.[7] Manjula.R.B, Sunilkumar S. Manvi," Issues in Underwater Acoustic SensorNetworks", International Journal of Computer and Electrical Engineering,Vol.3, No.1, February, 2011.[8] J. A. C. Bingham, "Multicarrier modulation for data transmission: An ideawhose time has come", IEEE Communications Magazine, pp. 5–14, May 1990.[9] Z. Wang and G. B. Giannakis, "Wireless multicarrier communications:Where Fourier meets Shannon", IEEE Signal Processing Magazine, vol. 17, no.3, pp. 29-48, May 2000.[10] Manjula. R. B, Sunilkumar S. Manvi, "Issues in Underwater AcousticSensor Networks", International Journal of Computer and ElectricalEngineering, Vol.3, No.1, February, 2011 1793-8163.