Research ArticleAn OFDM Receiver with Frequency Domain Diversity CombinedImpulsive Noise Canceller for Underwater Network

Rie Saotome1 Tran Minh Hai2 Yasuto Matsuda2 Taisaku Suzuki3 and Tomohisa Wada1

1Department of Information Engineering University of the Ryukyus 1 Senbaru Nishihara Okinawa 903-0213 Japan2Graduate School of Engineering and Science University of the Ryukyus 1 Senbaru Nishihara Okinawa 903-0213 Japan3Department of Media Information Engineering Okinawa National College of Technology 905 Aza-Henoko NagoOkinawa 905-2191 Japan

Correspondence should be addressed to Rie Saotome saotomelsiieu-ryukyuacjp

Received 7 April 2015 Revised 21 July 2015 Accepted 27 July 2015

Academic Editor Jianhua Zhang

Copyright copy 2015 Rie Saotome et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

In order to explore marine natural resources using remote robotic sensor or to enable rapid information exchange between ROV(remotely operated vehicles) AUV (autonomous underwater vehicle) divers and ships ultrasonic underwater communicationsystems are usedHowever if the communication system is applied to rich living creaturemarine environment such as shallow sea itsuffers fromgenerated ImpulsiveNoise so-called ShrimpNoise which is randomly generated in timedomain and seriously degradescommunication performance in underwater acoustic network With the purpose of supporting high performance underwatercommunication a robust digital communication method for Impulsive Noise environments is necessary In this paper we proposeOFDM ultrasonic communication system with diversity receiver The main feature of the receiver is a newly proposed FrequencyDomain Diversity Combined Impulsive Noise CancellerTheOFDM receiver utilizes 20ndash28KHz ultrasonic channel and subcarrierspacing of 46875Hz (MODE3) and 93750Hz (MODE2) OFDM modulations In addition the paper shows Impulsive Noisedistribution data measured at a fishing port in Okinawa and at a barge in Shizuoka prefectures and then proposed diversity OFDMtransceivers architecture and experimental results are described By the proposed Impulsive Noise Canceller frame bit error ratehas been decreased by 20ndash30

1 Introduction

Underwater wireless communication system [1] will offer awide variety of applications such as natural disaster warningremote control of offshore objects and discovery of newresources at the bottom of ocean Emerging underwaterwireless network devices can be equipped onto underwatervehicles or robots with sensors and video cameras Thensystem in ships can access those sensors and video camerasthrough underwater wireless network by acoustic wirelesslink Figure 1 shows an example to control a bottom seaROV (remotely operated vehicles) which engages exploringnatural resources by sensors from water surface operator inship through underwater network Since many living crea-tures in marine usually attach to communication devices andthey generate impulse type acoustic noise as an interferencea communication system for underwater application has to

be robust for this Impulsive Noise Then high bandwidthacoustic communication system which is robust for livingcreature generating Impulsive Noise so-called ShrimpNoiseis required

In this paper we have designed underwater acousticOrthogonal Frequency DivisionMultiplexing (OFDM) com-munication system with four receiving transducers diversity[2ndash4] The proposed system utilizes one TX (transmitting)transducer and four RX (receiving) transducers with centerfrequency of 24KHz 8 KHz bandwidth ultrasonic sound161 subcarriers in MODE3 and 81 subcarriers in MODE2are multiplexes in 8KHz bandwidth In order to increasethe robustness of the receiver under Impulsive and WhiteGaussian Noise circumstances MRC (maximum ration com-biner) Frequency Domain Diversity Combined ImpulsiveNoise Canceller is newly proposed Section 2 describes thestatistics analysis of the marine creaturesrsquo Impulsive Noise

Hindawi Publishing Corporatione Scientific World JournalVolume 2015 Article ID 841750 10 pageshttpdxdoiorg1011552015841750

2 The Scientific World Journal

ROV (remotelyoperated vehicle)

Sensing exploring

Underwateracoustic

communication

Figure 1 Application of underwater networking

Wha

rf

Wha

rf

Transmitter(transducer)

Receiver(4 transducers)

RX depth = 2mTX depth = 2m

Impulsive noise

Transmitter Receiver

(a)

Barge Barge

TX depth = 22m

RX depth =39m

Transmitter(transducer)

Receiver(4 transducers) Impulsive noise

Transmitter Receiver

(b)

Figure 2 (a) Okinawa experiment at OJIMA fishing port (b) Shizuoka experiment in UCHIURA barge

based on measurements at a fishing port in Okinawa andat a barge in Shizuoka Prefecture Section 3 shows thecommunication systemdesign details including the proposedFrequency Domain Diversity Combined Impulsive NoiseCanceller Then experimental results will be disclosed inSection 4 Finally conclusion is given in Section 5

2 Analysis of Impulsive Noise

We have performed ocean experiments at two sites One isOJIMA located at the south of Okinawa Island in OkinawaPrefecture in Japan The other is barge in UCHIURA-cove inNumazu city in Shizuoka which is roughly 400m offshoreand the sea depth is 30m Figure 2(a) shows a cross-sectionalview of OJIMA experiment site Both TX and RX transducers

are set in the depth of 2m and the distance between TX andRX is 715m The sea depth is roughly 5m At the wall of thewharf sea creatures are expected to generate Impulsive Noiseas shown in the figure Figure 2(b) shows a cross-sectionalview of UCHIURA barge experiment site TX transducer isset in the depth of 22m and four RX transducers are set inthe depth of 3m and 9m At the bottom of the barge seacreatures are expected to generate Impulsive Noise as shownin the figure

Figure 3 shows a measured received signal of one RXtransducer at UCHIURA barge experiment with RX depthof 3m The center portion of the signal is OFDM frameand another portion corresponds to No signal Many largeImpulsive Noises are observed Obviously a large impulsegives serious damage to the OFDM signal demodulation

The Scientific World Journal 3

015

01

005

0

minus005

minus0111 115 12 125 13 135 14

times106

Am

plitu

de

Time (sample)

Backgroundnoise

Impulsive noise

OFDMframe

Figure 3 Received signal with many Impulsive Noise Peaks

Then the background noise amplitude distributions are ana-lyzed This analysis only observes the background noise andno OFDM frame signal is included Figures 4(a) and 4(b) arebackground noise distribution at Shizuoka with depth = 3mand 9m respectively Figures 5(a) and 5(b) are accumulatedbackground noise from + infinity at Shizuoka site with depth= 3m 9m From the figures more than 99 of the signalscorrespond to (a) area which has relatively small signalamplitude that is low noise power However as shown infigures relatively large signal amplitude (large noise power)is observed in the areas (b) and (c) Comparing Figures 4(a)with 4(b) shallow depth of 3m shows larger signal amplitudeTherefore the source of the Impulsive Noise can be expectedat the bottom of the barge

Background noise distribution data at OJIMA site is alsoshown in Figure 4(c) Although the OJIMA site which locatesin semitropical area is a very much different condition asshown in Figure 2(a) comparing with the Shizuoka barge2(b) similar distribution data is obtained Figure 5(c) isaccumulated background noise from + infinity at OJIMAsite From the figure roughly 98 of the signals correspondto (a) area which has relatively small signal amplitude thatis low noise power Therefore in order to achieve reliabledigital communication in marine environment ImpulsiveNoisemitigation is required In the following section detail ofproposed OFDM receiver with Frequency Domain DiversityCombined Impulsive Noise Canceller will be described

3 Communication System Design withImpulsive Noise Canceller

31 Communication System Architecture [5 6] Figure 6shows a block diagram of OFDM communication systemwith Frequency Domain Diversity Combined ImpulsiveNoise Canceller The upper side is a transmitter (TX) Bitinformation ismapped toQPSK 16QAM or 64QAMconstel-lations Some BPSK modulated pilot patterns are inserted inthe stream in order to enable a channel estimation at receiverside (RX) After that OFDM modulation is performedthrough Inverse Fast Fourier Transform (IFFT) Then GuardInterval (GI) is added to each OFDM symbol By upconver-sion block the modulated baseband signal is upconvertedto passband such as 20ndash28KHz TX transducer outputsultrasonic sound OFDM signal into the sea water Throughthe underwater acoustic channel four RX transducers receivethe transmission signal Then the signals are processed inreverse order such as amplifying downconverting to base-band signal removing Guard Interval and FFT By usingthe inserted pilot signal channel estimation is performed Atthe equalizer block FFT output data is equalized The circle-A and circle-B in the figure correspond to Time Domainand Frequency Domain Impulsive Noise Cancellation signalprocessing block

Following the equalizer there are two stages of 4 inputsMaximum Ratio Combiner (MRC) blocks Equation (1)shows the MRC computation [7ndash9] MRC(119896) is the output ofeach MRC block

MRC (119896) =EQ (119896)1 sdot

1003816100381610038161003816119867 (119896)110038161003816100381610038162120590

21198991+EQ (119896)2 sdot

1003816100381610038161003816119867 (119896)210038161003816100381610038162120590

21198992 + EQ (119896)3 sdot

1003816100381610038161003816119867 (119896)310038161003816100381610038162120590

21198993 + EQ (119896)2 sdot

1003816100381610038161003816119867 (119896)410038161003816100381610038162120590

21198994

1003816100381610038161003816119867 (119896)11003816100381610038161003816212059021198991 +1003816100381610038161003816119867 (119896)21003816100381610038161003816212059021198992 +1003816100381610038161003816119867 (119896)31003816100381610038161003816212059021198993 +1003816100381610038161003816119867 (119896)41003816100381610038161003816212059021198994

(1)

Here EQ(119896)1ndash4 are 4 equalizer outputs119867(119896)1ndash4 are ChannelTransfer Functions (CTFs) of each underwater channel index

119896 corresponds to subcarrier number and 12059021198991minus4

are estimatedaverage noise power at each underwater channel Highlight

4 The Scientific World Journal

106

105

104

103

102

101

100

0 005 01 015 02 025

Noise amplitude

Dist

ribut

ion

(au

)

(A)

(B)

(C)

Shizuoka depth = 3m

(a)

106

105

104

103

102

101

100

Noise amplitude

Dist

ribut

ion

(au

) (A)

(B)

(C)

Shizuoka depth = 9m

0 001 002 003 004 005 006 007

(b)

0 0005 001 0015 002 0025

Noise amplitude

Dist

ribut

ion

(au

)

106

105

104

103

102

101

100

Okinawa

(A)

(B)

(C)

(c)

Figure 4 (a) Background noise distribution at Shizuoka (depth = 3m) (b) Background noise distribution at Shizuoka (depth = 9m) (c)Background noise distribution at Okinawa

of the system design is frequency domain Impulsive NoiseCanceller with two-stage MRC combiners although thereference paper [10] shows a frequency domain ImpulsiveNoise Canceller but no diversity combiners The 1st MRCcombinerrsquos outputs are utilized to suppress Impulsive Noiseby frequency domain Impulsive Noise Cancellation signalprocessing block circle-B

Table 1 summarizes detail system parameters Samplingfrequency of the system is 96KHz and two kinds of operationmodes are supported such as MODE2 and MODE3 whichcorrespond to 93750Hz and 46875Hz subcarrier spacingrespectively 8 KHz of communication bandwidth is dividedinto 81 subcarriers inMODE2 and 161 subcarriers inMODE3In order to simplify the design Guard Interval (GI) length ofa half size of OFDM symbol length 119879 is used

32 Time and Frequency Impulsive Noise Cancelling Becauseof the presence of Impulsive Noise so-called Shrimp Noise

Table 1 System parameters

Parameters Mode2 3

TX-RX elements 1 TX and 4RX transducerSampling frequency 96000HzTX center frequency 24000HzBand width 8000HzFFT size 1024 2048OFDM symbol length 119879 10667ms 21333msGI length 05119879 05119879Subcarrier spacing 9375Hz 46875HzNumber of subcarriers 81 161

time and frequency domain Impulsive Noise Cancelling(time-IMP and freq-IMP) methods [4 5] are employed

The Scientific World Journal 5

Noise amplitude0 005 01 015 02 025

(B)(A) (C)

Shizuoka depth = 3m

102

101

100

10minus1

10minus2

10minus3

10minus4

10minus5

Accu

mul

atio

nfro

m+

infin

ity(

)

(a)

0 001 002 003 004 005 006 007

(A)

Accu

mul

atio

nfro

m+

infin

ity(

)

Noise amplitude

(C)(B)

102

101

100

10minus1

10minus2

10minus3

10minus4

10minus5

Shizuoka depth = 9m

(b)

Noise amplitude0 0005 001 0015 002 0025

(B)(A) (C)

Okinawa

Accu

mul

atio

nfro

m+

infin

ity(

)

102

101

100

10minus1

10minus2

10minus3

10minus4

(c)

Figure 5 (a) Accumulated background noise from + infinity at Shizuoka (depth = 3m) (b) Accumulated background noise from + infinityat Shizuoka (depth = 9m) (c) Accumulated background noise from + infinity at Okinawa

and further combined with carrier diversity combiner Theblock diagram of time-IMP (circle-A in Figure 6) is shownin Figure 7(a) It detects Impulsive Noise samples in timedomain and then removes the Impulsive Noise samplesaccording to

119910 (119899) =

119903 (119899) if |119903 (119899)|2 le 120572 sdot 119875avg0 otherwise

(2)

Here 119903(119899) is 119899th time sample of received signal after pream-plifier 119875avg is the average power of an OFDM symbol and120572 is threshold parameter Obviously if the power of inputsignal 119903(119899) exceeds the threshold of 120572 sdot 119875avg the signal isreplaced by zero This method is called ldquoblankingrdquo [10 11]In order to determine the threshold parameter 120572 cut andtry experimentation was applied and finally 120572 = 20 isdeterminedThe system performance is not so sensitive to theparameter

In addition a frequency domain Impulsive Noise Cancel-lation is appliedThe detail of freq-IMP (circle-B in Figure 6)is shown in Figure 7(b) The 1st MRC block generates thecombined consternation of 4 branches equalizers Then thefollowing processing is applied in hard decision and pilotinsertion block

(1) Subcarriers which should be silent (no data or pilots)are set to zero

(2) Subcarriers which are used as pilots are replaced byknown pilot values

(3) Subcarriers which are used for data transmission aredemapped to nearest position in constellation plot

Then succeeding subtraction block computes frequencydomain noise components In the following blocks thefrequency domain noise components are converted to a TimeDomain Impulsive Noise The frequency domain noise is

6 The Scientific World Journal

Bitinformation Map Pilot

insertion IFFTGI

addition upconversion

TX transducer

Preamp-

4 RX transducers

Underwater channel

Down-

FFT

FFT

FFT

FFT

Equalize

Equalize

Equalize

Equalize

A

A

A

A

Poweramplification

Time DomainImpulsive Noise Canceller

Frequency DomainDiversity Combined

Impulsive Noise Canceller1stMRCwithnoiseest

2ndMRCwithnoiseest

De-map

Bitinformation

B

B

B

B

Re-move

GI

Re-move

GI

Re-move

GI

Re-move

GI

Pilotdata

Pilotdata

Data

Pilot

Pilot

data

Channelestimation

Channelestimation

Channelestimation

Channelestimation

lification

Preamp-lification

Preamp-lification

Preamp-lification

conversion

Down-conversion

Down-conversion

Down-conversion

Figure 6 Block diagram of OFDM communication system with Frequency Domain Diversity Combined Impulsive Noise Canceller

multiplied with Channel Transfer Function of branch 119894 (119867119894

119894 = 1 to 4) and is IFFTed If the output of IFFThas a large peakit can be considered as estimated Time Domain ImpulsiveNoise signal 119906(119899) according to

119906 (119899) =

119889 (119899) if 10038161003816100381610038161003816119889 (119899)10038161003816100381610038161003816

2ge 120573 sdot 120590

2

0 otherwise(3)

Here 119889(119899) is 119899th time sample of IFFT output 1205902 is the averagepower of the IFFT output and 120573 is threshold parameterObviously if the power of input signal 119889(119899) exceeds thethreshold of 120573 sdot 1205902 the signal is considered as an ImpulsiveNoise Then the estimated Time Domain Impulsive Noisesignal 119906(119899) is converted to frequency domain by FFT blockand division by 119867

119894 Finally the estimated frequency domain

Impulsive Noise is withdrawn from equalized signal ofbranch 119894 (119894 = 1 to 4) EQA

119894 Then the noise reduced outputs

EQB119894(119894 = 1 to 4) are obtained and then they are combined

in the 2nd MRC block to generate further noise reducedconstellation The parameter 120573 is determined by cut and tryexperimentations and 120573 = 8 is finally applied to the systemThe obtained noise cancel performance is not so sensitive tothe parameter 120573

4 Experimental Results

Table 2 summarizes ocean experiment parameters for bothOkinawa OJIMA fishing port site and Shizuoka UCHIURAbarge site In Okinawa site horizontal underwater com-munication is carried out while vertical communication isexperimented in Shizuoka site for The four RX transducersare placed as linear array with 20 cm pitch Figures 8(a) and8(b) are MODE3 and MODE2 constellations in Okinawasite respectively Upper four constellations correspond tobranches 1 to 4 2nd MRC inputs The bottom constellationis 2nd MRC output The horizontal axis of each figure showsreal part of digital modulation while the vertical axis corre-sponds to imaginary part According to the figures carrierdiversity combining MRC successfully reduces variance ofconstellation points

Figures 9(a) and 9(b) are the OFDM symbol by symbolbit error rate (BER) comparisons between frequency domainImpulsive Noise cancellation ON and OFF for MODE316QAM modulation in Okinawa site and MODE2 16QAMin Shizuoka site respectively The solid line corresponds tofreq-IMP OFF and the dashed line corresponds to ONHorizontal axis shows BER and vertical axis shows OFDMsymbol number According to the figures the proposedFrequency Domain Diversity Combined Impulsive NoiseCanceller effectively reduces symbol by symbol BER values

The Scientific World Journal 7

Preamplificationoutput

0 Yes

No

Downconverter

gtthreshold|and2|

(a) Time Domain Impulsive Noise Canceller

Harddecisionand pilotinsertion

+ IFFTPeak

detect FFT

1X

+

minus

Hminus1

1stMRCwithnoise

est

xx

Estimatedfreq

domainnoise

++

minus

EQAi EQBi

Hi

(b) Frequency Domain Impulsive Noise Canceller

Figure 7 Time and Frequency Domain Impulsive Noise Canceller

Table 2 Ocean experiment parameters

Parameters PlaceOkinawa Shizuoka

Type ofexperiment site Fishing port Barge

Number oftransducers 1 TX and 4RX 1 TX and 4RX

RX transducerarray Linear with 20 cm pitch Linear with 20 cm pitch

Modulation QPSK16QAM QPSK16QAMTX-RX distance 715m 13mTransmissiondirection Horizontal Vertical

Ocean depth 5m 30mTransducerdepth (TXRX) 2m2m 22m9m

Figure 10 summarizes BER comparisons for FrequencyDomain Diversity Combined Impulsive Noise Canceller forfreq-IMP ON and OFF In spite of place of experiment siteapproximately 20ndash30 of BER reduction has been obtainedby the proposed Frequency Domain Diversity CombinedImpulsive Noise Canceller

Finally Figures 11(a) and 11(b) showmeasured communi-cation system performance in Okinawa site with enabling theproposed Impulsive Noise Canceller In addition BER data

is also measured with moving the TX transducer speed suchas 00ms (stable) 03 and 06ms for QPSK modulationand 00ms (stable) and 03ms for 16QAM Since MODE2OFDM symbol length of 10667ms as shown in Table 1 is halfof MODE3 the Guard Interval length of MODE2 is shorterthan MODE3 Therefore it is considered that high levelmodulation of 16QAM of MODE2 is largely affected by InterSymbol Interference (ISI) As you can see from those figuresMODE2 shows higher robustness for moving cases such asstrangeness for Doppler Effect because of shorter OFDMsymbol length of MODE2 Once forward error correctionsuch as TURBO CODE with code rate 119877 = 13 is appliedto this system all error free communications (BER = 0) forQPSK modulation of all speeds and 16QAM stable case havebeen confirmed

5 Conclusion

In this paper an ultrasonic OFDM transceiver architecturewith four diverse receivers supporting 2 MODEs is proposedwith enhancement of Impulsive Noise insusceptibility Itutilizes 20ndash28KHz ultrasonic channel and subcarrier spacingof 46875Hz (MODE3) and 93750Hz (MODE2) OFDMmodulations

Two different environment ocean experiments were con-ducted to evaluate the statistics of Impulsive Noise so-called Shrimp Noise which is generated by living creaturesin marine environment To deal with challenges posed by

8 The Scientific World Journal

minus1

0

1

1

minus1 0 1

minus1

0

minus1 0 1

1

minus1

0

minus1 0 1

Branch 1

Branch 3

minus1

0

1

minus1 0 1

minus1

0

1

minus1 0 1

Branch 2

Branch 4

2nd MRC (MODE3)

(a)

minus1

0

1

minus1 0 1

Branch 1

minus1

0

1

minus1 0 1

minus1

0

1

minus1 0 1

Branch 3

minus1

0

1

minus1 0 1

minus1

0

1

minus1 0 1

Branch 2

Branch 4

2nd MRC (MODE2)

(b)

Figure 8 (a)MODE3QPSK constellations (branches 1 to 4 andMRCOkinawa) (b)MODE2QPSK constellations (branches 1 to 4 andMRCOkinawa)

The Scientific World Journal 9

5 10 15 20 25 30 35 40

Symbol number

BER

OFF F-IMP CAN

ON F-IMP CAN

100

10minus1

10minus2

10minus3

(a)

10 20 30 40 50 60 70 80

Symbol number

BER

OFF F-IMP CAN

ON F-IMP CAN

100

10minus1

10minus2

10minus3

(b)

Figure 9 (a) MODE3 BER comparison for 16QAM at Okinawa (b) MODE2 BER comparison for 16QAM at Shizuoka

0

002

004

006

008

01

012

014

QPSKOkinawa

BER

16QAMOkinawa

QPSKOkinawa

16QAMOkinawa

16QAMShizuoka

16QAMShizuoka

OFF F-IMP CANON F-IMP CAN

MODE3 MODE2 MODE2 MODE2MODE3 MODE3

Figure 10 Summary of BER comparison for Frequency Domain Diversity Combined Impulsive Noise Canceller

00001

0001

001

01

10

4 diversity RX transducers liner array with 20 cm pitchQPSK modulationDistance = 715mOkinawa sea experiment

00 03 06

MODE3

MODE2

TX moving speed (ms)

Bit e

rror

rate

(a)

00001

0001

001

01

10

00 03

MODE3

MODE2

16QAM modulationDistance = 715mOkinawa sea experiment

TX moving speed (ms)

Bit e

rror

rate

4 diversity RX transducers liner array with 20 cm pitch

(b)

Figure 11 (a) Communication system performance using QPSK modulation (b) Communication system performance using 16QAMmodulation

10 The Scientific World Journal

the Impulsive Noise interference we have employed a Fre-quency Domain Diversity Combined Impulsive Noise Can-celler

Measured data at a fishing port in Okinawa and ata barge in Shizuoka Prefecture are disclosed and in bothenvironment randomly generated Time Domain ImpulseNoise distribution is observed Although more than 90 to99 of background noise shows AWGN type distributionless than 1 to 10 noise components show large amplitudesuch as Impulsive Noise

By enabling the proposed MRC combined Frequencydomain Impulsive Noise Canceller 20ndash30 BER reductionhas been successfully obtained In addition 715m shallowfishing port horizontal underwater communication BERis also shown using QPSK and 16QAM modulations Byapplying TURBO 119877 = 13 forward error correction allerror free communications using QPSK and 16QAMwith notmoving case are also confirmed

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgment

This study has been carried out as part of the StrategicInformation andCommunications RampDPromotion Program(SCOPE) Project of the Ministry of Internal Affairs andCommunications Japan

References

[1] M Stojanovic ldquoRecent advances in high-speed underwateracoustic communicationsrdquo IEEE Journal of Oceanic Engineer-ing vol 21 no 2 pp 125ndash136 1996

[2] G Qiao W Wang R Guo R Khan and Y Wang ldquoFrequencydiversity for OFDM mobile communication via underwateracoustic channelsrdquo Journal of Marine Science and Applicationvol 11 no 1 pp 126ndash133 2012

[3] G Leus and P A van Walree ldquoMultiband OFDM for covertacoustic communicationsrdquo IEEE Journal on Selected Areas inCommunications vol 26 no 9 pp 1662ndash1673 2008

[4] H D Trung and N Van Duc ldquoAn analysis of MIMO-OFDMfor underwater communicationsrdquo in Proceedings of the 3rdInternational Congress on Ultra Modern Telecommunicationsand Control Systems and Workshops (ICUMT rsquo11) BudapestHungary October 2011

[5] T Suzuki H M Tran and T Wada ldquoAn underwater acousticOFDM communication system with shrimp (impulsive) noisecancellingrdquo in Proceedings of the IEEE International Conferenceon Computing Management and Telecommunications (Com-ManTel rsquo14) pp 152ndash156 Da Nang Vietnam April 2014

[6] T M Hai Y Matsuda T Suzuki and W A D A Tomo-hisa ldquoUltrasonic diversity OFDM transceiver architecture withimpulsive noise cancelling for shallow sea communicationrdquo inProceedings of the UA2014-2nd International Conference andExhibition on Underwater Acoustics Rodes Greece June 2014

[7] J Gao T IidaHMizutani et al ldquoThe implementation of a 248antennas configurable diversity OFDM receiver for mobile

HDTV applicationrdquo International Journal of Digital MultimediaBroadcasting vol 2009 Article ID 303627 9 pages 2009

[8] T Koujita T Wada T Iida et al ldquoEvaluation of diversitycombining systems for mobile reception in digital terrestrialtelevision broadcastingrdquo in Proceedings of the IEEE Inter-national Conference on Consumer Electronics SESSION 63Digital Receivers I Las Vegas Nev USA January 2009

[9] T Wada ldquoDiversity and error compensation in OFDMtransceivers principles and implementationrdquo in Digital Front-End in Wireless Communication and Broadcasting chapter 18Cambridge University Press 2011

[10] S V Zhidkov ldquoImpulsive noise suppression in OFDM basedcommunication systemsrdquo IEEE Transactions on Consumer Elec-tronics vol 49 no 4 pp 944ndash948 2003

[11] K Al-Mawali A Z Sadik and Z M Hussain ldquoJointtime-domainfrequency-domain impulsive noise reduction inOFDM-based power line communicationsrdquo in Proceedings ofthe Australasian Telecommunication Networks and ApplicationsConference (ATNAC rsquo08) pp 138ndash142 Adelaide AustralasianDecember 2008

Submit your manuscripts athttpwwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Mechanical Engineering

Advances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Distributed Sensor Networks

International Journal of

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Antennas andPropagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

The Scientific World Journal 3

015

01

005

0

minus005

minus0111 115 12 125 13 135 14

times106

Am

plitu

de

Time (sample)

Backgroundnoise

Impulsive noise

OFDMframe

Figure 3 Received signal with many Impulsive Noise Peaks

Then the background noise amplitude distributions are ana-lyzed This analysis only observes the background noise andno OFDM frame signal is included Figures 4(a) and 4(b) arebackground noise distribution at Shizuoka with depth = 3mand 9m respectively Figures 5(a) and 5(b) are accumulatedbackground noise from + infinity at Shizuoka site with depth= 3m 9m From the figures more than 99 of the signalscorrespond to (a) area which has relatively small signalamplitude that is low noise power However as shown infigures relatively large signal amplitude (large noise power)is observed in the areas (b) and (c) Comparing Figures 4(a)with 4(b) shallow depth of 3m shows larger signal amplitudeTherefore the source of the Impulsive Noise can be expectedat the bottom of the barge

Background noise distribution data at OJIMA site is alsoshown in Figure 4(c) Although the OJIMA site which locatesin semitropical area is a very much different condition asshown in Figure 2(a) comparing with the Shizuoka barge2(b) similar distribution data is obtained Figure 5(c) isaccumulated background noise from + infinity at OJIMAsite From the figure roughly 98 of the signals correspondto (a) area which has relatively small signal amplitude thatis low noise power Therefore in order to achieve reliabledigital communication in marine environment ImpulsiveNoisemitigation is required In the following section detail ofproposed OFDM receiver with Frequency Domain DiversityCombined Impulsive Noise Canceller will be described

3 Communication System Design withImpulsive Noise Canceller

31 Communication System Architecture [5 6] Figure 6shows a block diagram of OFDM communication systemwith Frequency Domain Diversity Combined ImpulsiveNoise Canceller The upper side is a transmitter (TX) Bitinformation ismapped toQPSK 16QAM or 64QAMconstel-lations Some BPSK modulated pilot patterns are inserted inthe stream in order to enable a channel estimation at receiverside (RX) After that OFDM modulation is performedthrough Inverse Fast Fourier Transform (IFFT) Then GuardInterval (GI) is added to each OFDM symbol By upconver-sion block the modulated baseband signal is upconvertedto passband such as 20ndash28KHz TX transducer outputsultrasonic sound OFDM signal into the sea water Throughthe underwater acoustic channel four RX transducers receivethe transmission signal Then the signals are processed inreverse order such as amplifying downconverting to base-band signal removing Guard Interval and FFT By usingthe inserted pilot signal channel estimation is performed Atthe equalizer block FFT output data is equalized The circle-A and circle-B in the figure correspond to Time Domainand Frequency Domain Impulsive Noise Cancellation signalprocessing block

Following the equalizer there are two stages of 4 inputsMaximum Ratio Combiner (MRC) blocks Equation (1)shows the MRC computation [7ndash9] MRC(119896) is the output ofeach MRC block

MRC (119896) =EQ (119896)1 sdot

1003816100381610038161003816119867 (119896)110038161003816100381610038162120590

21198991+EQ (119896)2 sdot

1003816100381610038161003816119867 (119896)210038161003816100381610038162120590

21198992 + EQ (119896)3 sdot

1003816100381610038161003816119867 (119896)310038161003816100381610038162120590

21198993 + EQ (119896)2 sdot

1003816100381610038161003816119867 (119896)410038161003816100381610038162120590

21198994

1003816100381610038161003816119867 (119896)11003816100381610038161003816212059021198991 +1003816100381610038161003816119867 (119896)21003816100381610038161003816212059021198992 +1003816100381610038161003816119867 (119896)31003816100381610038161003816212059021198993 +1003816100381610038161003816119867 (119896)41003816100381610038161003816212059021198994

(1)

Here EQ(119896)1ndash4 are 4 equalizer outputs119867(119896)1ndash4 are ChannelTransfer Functions (CTFs) of each underwater channel index

119896 corresponds to subcarrier number and 12059021198991minus4

are estimatedaverage noise power at each underwater channel Highlight

4 The Scientific World Journal

106

105

104

103

102

101

100

0 005 01 015 02 025

Noise amplitude

Dist

ribut

ion

(au

)

(A)

(B)

(C)

Shizuoka depth = 3m

(a)

106

105

104

103

102

101

100

Noise amplitude

Dist

ribut

ion

(au

) (A)

(B)

(C)

Shizuoka depth = 9m

0 001 002 003 004 005 006 007

(b)

0 0005 001 0015 002 0025

Noise amplitude

Dist

ribut

ion

(au

)

106

105

104

103

102

101

100

Okinawa

(A)

(B)

(C)

(c)

Figure 4 (a) Background noise distribution at Shizuoka (depth = 3m) (b) Background noise distribution at Shizuoka (depth = 9m) (c)Background noise distribution at Okinawa

of the system design is frequency domain Impulsive NoiseCanceller with two-stage MRC combiners although thereference paper [10] shows a frequency domain ImpulsiveNoise Canceller but no diversity combiners The 1st MRCcombinerrsquos outputs are utilized to suppress Impulsive Noiseby frequency domain Impulsive Noise Cancellation signalprocessing block circle-B

Table 1 summarizes detail system parameters Samplingfrequency of the system is 96KHz and two kinds of operationmodes are supported such as MODE2 and MODE3 whichcorrespond to 93750Hz and 46875Hz subcarrier spacingrespectively 8 KHz of communication bandwidth is dividedinto 81 subcarriers inMODE2 and 161 subcarriers inMODE3In order to simplify the design Guard Interval (GI) length ofa half size of OFDM symbol length 119879 is used

32 Time and Frequency Impulsive Noise Cancelling Becauseof the presence of Impulsive Noise so-called Shrimp Noise

Table 1 System parameters

Parameters Mode2 3

TX-RX elements 1 TX and 4RX transducerSampling frequency 96000HzTX center frequency 24000HzBand width 8000HzFFT size 1024 2048OFDM symbol length 119879 10667ms 21333msGI length 05119879 05119879Subcarrier spacing 9375Hz 46875HzNumber of subcarriers 81 161

time and frequency domain Impulsive Noise Cancelling(time-IMP and freq-IMP) methods [4 5] are employed

The Scientific World Journal 5

Noise amplitude0 005 01 015 02 025

(B)(A) (C)

Shizuoka depth = 3m

102

101

100

10minus1

10minus2

10minus3

10minus4

10minus5

Accu

mul

atio

nfro

m+

infin

ity(

)

(a)

0 001 002 003 004 005 006 007

(A)

Accu

mul

atio

nfro

m+

infin

ity(

)

Noise amplitude

(C)(B)

102

101

100

10minus1

10minus2

10minus3

10minus4

10minus5

Shizuoka depth = 9m

(b)

Noise amplitude0 0005 001 0015 002 0025

(B)(A) (C)

Okinawa

Accu

mul

atio

nfro

m+

infin

ity(

)

102

101

100

10minus1

10minus2

10minus3

10minus4

(c)

Figure 5 (a) Accumulated background noise from + infinity at Shizuoka (depth = 3m) (b) Accumulated background noise from + infinityat Shizuoka (depth = 9m) (c) Accumulated background noise from + infinity at Okinawa

and further combined with carrier diversity combiner Theblock diagram of time-IMP (circle-A in Figure 6) is shownin Figure 7(a) It detects Impulsive Noise samples in timedomain and then removes the Impulsive Noise samplesaccording to

119910 (119899) =

119903 (119899) if |119903 (119899)|2 le 120572 sdot 119875avg0 otherwise

(2)

Here 119903(119899) is 119899th time sample of received signal after pream-plifier 119875avg is the average power of an OFDM symbol and120572 is threshold parameter Obviously if the power of inputsignal 119903(119899) exceeds the threshold of 120572 sdot 119875avg the signal isreplaced by zero This method is called ldquoblankingrdquo [10 11]In order to determine the threshold parameter 120572 cut andtry experimentation was applied and finally 120572 = 20 isdeterminedThe system performance is not so sensitive to theparameter

In addition a frequency domain Impulsive Noise Cancel-lation is appliedThe detail of freq-IMP (circle-B in Figure 6)is shown in Figure 7(b) The 1st MRC block generates thecombined consternation of 4 branches equalizers Then thefollowing processing is applied in hard decision and pilotinsertion block

(1) Subcarriers which should be silent (no data or pilots)are set to zero

(2) Subcarriers which are used as pilots are replaced byknown pilot values

(3) Subcarriers which are used for data transmission aredemapped to nearest position in constellation plot

Then succeeding subtraction block computes frequencydomain noise components In the following blocks thefrequency domain noise components are converted to a TimeDomain Impulsive Noise The frequency domain noise is

6 The Scientific World Journal

Bitinformation Map Pilot

insertion IFFTGI

addition upconversion

TX transducer

Preamp-

4 RX transducers

Underwater channel

Down-

FFT

FFT

FFT

FFT

Equalize

Equalize

Equalize

Equalize

A

A

A

A

Poweramplification

Time DomainImpulsive Noise Canceller

Frequency DomainDiversity Combined

Impulsive Noise Canceller1stMRCwithnoiseest

2ndMRCwithnoiseest

De-map

Bitinformation

B

B

B

B

Re-move

GI

Re-move

GI

Re-move

GI

Re-move

GI

Pilotdata

Pilotdata

Data

Pilot

Pilot

data

Channelestimation

Channelestimation

Channelestimation

Channelestimation

lification

Preamp-lification

Preamp-lification

Preamp-lification

conversion

Down-conversion

Down-conversion

Down-conversion

Figure 6 Block diagram of OFDM communication system with Frequency Domain Diversity Combined Impulsive Noise Canceller

multiplied with Channel Transfer Function of branch 119894 (119867119894

119894 = 1 to 4) and is IFFTed If the output of IFFThas a large peakit can be considered as estimated Time Domain ImpulsiveNoise signal 119906(119899) according to

119906 (119899) =

119889 (119899) if 10038161003816100381610038161003816119889 (119899)10038161003816100381610038161003816

2ge 120573 sdot 120590

2

0 otherwise(3)

Here 119889(119899) is 119899th time sample of IFFT output 1205902 is the averagepower of the IFFT output and 120573 is threshold parameterObviously if the power of input signal 119889(119899) exceeds thethreshold of 120573 sdot 1205902 the signal is considered as an ImpulsiveNoise Then the estimated Time Domain Impulsive Noisesignal 119906(119899) is converted to frequency domain by FFT blockand division by 119867

119894 Finally the estimated frequency domain

Impulsive Noise is withdrawn from equalized signal ofbranch 119894 (119894 = 1 to 4) EQA

119894 Then the noise reduced outputs

EQB119894(119894 = 1 to 4) are obtained and then they are combined

in the 2nd MRC block to generate further noise reducedconstellation The parameter 120573 is determined by cut and tryexperimentations and 120573 = 8 is finally applied to the systemThe obtained noise cancel performance is not so sensitive tothe parameter 120573

4 Experimental Results

Table 2 summarizes ocean experiment parameters for bothOkinawa OJIMA fishing port site and Shizuoka UCHIURAbarge site In Okinawa site horizontal underwater com-munication is carried out while vertical communication isexperimented in Shizuoka site for The four RX transducersare placed as linear array with 20 cm pitch Figures 8(a) and8(b) are MODE3 and MODE2 constellations in Okinawasite respectively Upper four constellations correspond tobranches 1 to 4 2nd MRC inputs The bottom constellationis 2nd MRC output The horizontal axis of each figure showsreal part of digital modulation while the vertical axis corre-sponds to imaginary part According to the figures carrierdiversity combining MRC successfully reduces variance ofconstellation points

Figures 9(a) and 9(b) are the OFDM symbol by symbolbit error rate (BER) comparisons between frequency domainImpulsive Noise cancellation ON and OFF for MODE316QAM modulation in Okinawa site and MODE2 16QAMin Shizuoka site respectively The solid line corresponds tofreq-IMP OFF and the dashed line corresponds to ONHorizontal axis shows BER and vertical axis shows OFDMsymbol number According to the figures the proposedFrequency Domain Diversity Combined Impulsive NoiseCanceller effectively reduces symbol by symbol BER values

The Scientific World Journal 7

Preamplificationoutput

0 Yes

No

Downconverter

gtthreshold|and2|

(a) Time Domain Impulsive Noise Canceller

Harddecisionand pilotinsertion

+ IFFTPeak

detect FFT

1X

+

minus

Hminus1

1stMRCwithnoise

est

xx

Estimatedfreq

domainnoise

++

minus

EQAi EQBi

Hi

(b) Frequency Domain Impulsive Noise Canceller

Figure 7 Time and Frequency Domain Impulsive Noise Canceller

Table 2 Ocean experiment parameters

Parameters PlaceOkinawa Shizuoka

Type ofexperiment site Fishing port Barge

Number oftransducers 1 TX and 4RX 1 TX and 4RX

RX transducerarray Linear with 20 cm pitch Linear with 20 cm pitch

Modulation QPSK16QAM QPSK16QAMTX-RX distance 715m 13mTransmissiondirection Horizontal Vertical

Ocean depth 5m 30mTransducerdepth (TXRX) 2m2m 22m9m

Figure 10 summarizes BER comparisons for FrequencyDomain Diversity Combined Impulsive Noise Canceller forfreq-IMP ON and OFF In spite of place of experiment siteapproximately 20ndash30 of BER reduction has been obtainedby the proposed Frequency Domain Diversity CombinedImpulsive Noise Canceller

Finally Figures 11(a) and 11(b) showmeasured communi-cation system performance in Okinawa site with enabling theproposed Impulsive Noise Canceller In addition BER data

is also measured with moving the TX transducer speed suchas 00ms (stable) 03 and 06ms for QPSK modulationand 00ms (stable) and 03ms for 16QAM Since MODE2OFDM symbol length of 10667ms as shown in Table 1 is halfof MODE3 the Guard Interval length of MODE2 is shorterthan MODE3 Therefore it is considered that high levelmodulation of 16QAM of MODE2 is largely affected by InterSymbol Interference (ISI) As you can see from those figuresMODE2 shows higher robustness for moving cases such asstrangeness for Doppler Effect because of shorter OFDMsymbol length of MODE2 Once forward error correctionsuch as TURBO CODE with code rate 119877 = 13 is appliedto this system all error free communications (BER = 0) forQPSK modulation of all speeds and 16QAM stable case havebeen confirmed

5 Conclusion

In this paper an ultrasonic OFDM transceiver architecturewith four diverse receivers supporting 2 MODEs is proposedwith enhancement of Impulsive Noise insusceptibility Itutilizes 20ndash28KHz ultrasonic channel and subcarrier spacingof 46875Hz (MODE3) and 93750Hz (MODE2) OFDMmodulations

Two different environment ocean experiments were con-ducted to evaluate the statistics of Impulsive Noise so-called Shrimp Noise which is generated by living creaturesin marine environment To deal with challenges posed by

8 The Scientific World Journal

minus1

0

1

1

minus1 0 1

minus1

0

minus1 0 1

1

minus1

0

minus1 0 1

Branch 1

Branch 3

minus1

0

1

minus1 0 1

minus1

0

1

minus1 0 1

Branch 2

Branch 4

2nd MRC (MODE3)

(a)

minus1

0

1

minus1 0 1

Branch 1

minus1

0

1

minus1 0 1

minus1

0

1

minus1 0 1

Branch 3

minus1

0

1

minus1 0 1

minus1

0

1

minus1 0 1

Branch 2

Branch 4

2nd MRC (MODE2)

(b)

Figure 8 (a)MODE3QPSK constellations (branches 1 to 4 andMRCOkinawa) (b)MODE2QPSK constellations (branches 1 to 4 andMRCOkinawa)

The Scientific World Journal 9

5 10 15 20 25 30 35 40

Symbol number

BER

OFF F-IMP CAN

ON F-IMP CAN

100

10minus1

10minus2

10minus3

(a)

10 20 30 40 50 60 70 80

Symbol number

BER

OFF F-IMP CAN

ON F-IMP CAN

100

10minus1

10minus2

10minus3

(b)

Figure 9 (a) MODE3 BER comparison for 16QAM at Okinawa (b) MODE2 BER comparison for 16QAM at Shizuoka

0

002

004

006

008

01

012

014

QPSKOkinawa

BER

16QAMOkinawa

QPSKOkinawa

16QAMOkinawa

16QAMShizuoka

16QAMShizuoka

OFF F-IMP CANON F-IMP CAN

MODE3 MODE2 MODE2 MODE2MODE3 MODE3

Figure 10 Summary of BER comparison for Frequency Domain Diversity Combined Impulsive Noise Canceller

00001

0001

001

01

10

4 diversity RX transducers liner array with 20 cm pitchQPSK modulationDistance = 715mOkinawa sea experiment

00 03 06

MODE3

MODE2

TX moving speed (ms)

Bit e

rror

rate

(a)

00001

0001

001

01

10

00 03

MODE3

MODE2

16QAM modulationDistance = 715mOkinawa sea experiment

TX moving speed (ms)

Bit e

rror

rate

4 diversity RX transducers liner array with 20 cm pitch

(b)

Figure 11 (a) Communication system performance using QPSK modulation (b) Communication system performance using 16QAMmodulation

10 The Scientific World Journal

the Impulsive Noise interference we have employed a Fre-quency Domain Diversity Combined Impulsive Noise Can-celler

Measured data at a fishing port in Okinawa and ata barge in Shizuoka Prefecture are disclosed and in bothenvironment randomly generated Time Domain ImpulseNoise distribution is observed Although more than 90 to99 of background noise shows AWGN type distributionless than 1 to 10 noise components show large amplitudesuch as Impulsive Noise

By enabling the proposed MRC combined Frequencydomain Impulsive Noise Canceller 20ndash30 BER reductionhas been successfully obtained In addition 715m shallowfishing port horizontal underwater communication BERis also shown using QPSK and 16QAM modulations Byapplying TURBO 119877 = 13 forward error correction allerror free communications using QPSK and 16QAMwith notmoving case are also confirmed

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgment

This study has been carried out as part of the StrategicInformation andCommunications RampDPromotion Program(SCOPE) Project of the Ministry of Internal Affairs andCommunications Japan

References

[1] M Stojanovic ldquoRecent advances in high-speed underwateracoustic communicationsrdquo IEEE Journal of Oceanic Engineer-ing vol 21 no 2 pp 125ndash136 1996

[2] G Qiao W Wang R Guo R Khan and Y Wang ldquoFrequencydiversity for OFDM mobile communication via underwateracoustic channelsrdquo Journal of Marine Science and Applicationvol 11 no 1 pp 126ndash133 2012

[3] G Leus and P A van Walree ldquoMultiband OFDM for covertacoustic communicationsrdquo IEEE Journal on Selected Areas inCommunications vol 26 no 9 pp 1662ndash1673 2008

[4] H D Trung and N Van Duc ldquoAn analysis of MIMO-OFDMfor underwater communicationsrdquo in Proceedings of the 3rdInternational Congress on Ultra Modern Telecommunicationsand Control Systems and Workshops (ICUMT rsquo11) BudapestHungary October 2011

[5] T Suzuki H M Tran and T Wada ldquoAn underwater acousticOFDM communication system with shrimp (impulsive) noisecancellingrdquo in Proceedings of the IEEE International Conferenceon Computing Management and Telecommunications (Com-ManTel rsquo14) pp 152ndash156 Da Nang Vietnam April 2014

[6] T M Hai Y Matsuda T Suzuki and W A D A Tomo-hisa ldquoUltrasonic diversity OFDM transceiver architecture withimpulsive noise cancelling for shallow sea communicationrdquo inProceedings of the UA2014-2nd International Conference andExhibition on Underwater Acoustics Rodes Greece June 2014

[7] J Gao T IidaHMizutani et al ldquoThe implementation of a 248antennas configurable diversity OFDM receiver for mobile

HDTV applicationrdquo International Journal of Digital MultimediaBroadcasting vol 2009 Article ID 303627 9 pages 2009

[8] T Koujita T Wada T Iida et al ldquoEvaluation of diversitycombining systems for mobile reception in digital terrestrialtelevision broadcastingrdquo in Proceedings of the IEEE Inter-national Conference on Consumer Electronics SESSION 63Digital Receivers I Las Vegas Nev USA January 2009

[9] T Wada ldquoDiversity and error compensation in OFDMtransceivers principles and implementationrdquo in Digital Front-End in Wireless Communication and Broadcasting chapter 18Cambridge University Press 2011

[10] S V Zhidkov ldquoImpulsive noise suppression in OFDM basedcommunication systemsrdquo IEEE Transactions on Consumer Elec-tronics vol 49 no 4 pp 944ndash948 2003

[11] K Al-Mawali A Z Sadik and Z M Hussain ldquoJointtime-domainfrequency-domain impulsive noise reduction inOFDM-based power line communicationsrdquo in Proceedings ofthe Australasian Telecommunication Networks and ApplicationsConference (ATNAC rsquo08) pp 138ndash142 Adelaide AustralasianDecember 2008

Submit your manuscripts athttpwwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Mechanical Engineering

Advances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Distributed Sensor Networks

International Journal of

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Antennas andPropagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

4 The Scientific World Journal

106

105

104

103

102

101

100

0 005 01 015 02 025

Noise amplitude

Dist

ribut

ion

(au

)

(A)

(B)

(C)

Shizuoka depth = 3m

(a)

106

105

104

103

102

101

100

Noise amplitude

Dist

ribut

ion

(au

) (A)

(B)

(C)

Shizuoka depth = 9m

0 001 002 003 004 005 006 007

(b)

0 0005 001 0015 002 0025

Noise amplitude

Dist

ribut

ion

(au

)

106

105

104

103

102

101

100

Okinawa

(A)

(B)

(C)

(c)

Figure 4 (a) Background noise distribution at Shizuoka (depth = 3m) (b) Background noise distribution at Shizuoka (depth = 9m) (c)Background noise distribution at Okinawa

of the system design is frequency domain Impulsive NoiseCanceller with two-stage MRC combiners although thereference paper [10] shows a frequency domain ImpulsiveNoise Canceller but no diversity combiners The 1st MRCcombinerrsquos outputs are utilized to suppress Impulsive Noiseby frequency domain Impulsive Noise Cancellation signalprocessing block circle-B

Table 1 summarizes detail system parameters Samplingfrequency of the system is 96KHz and two kinds of operationmodes are supported such as MODE2 and MODE3 whichcorrespond to 93750Hz and 46875Hz subcarrier spacingrespectively 8 KHz of communication bandwidth is dividedinto 81 subcarriers inMODE2 and 161 subcarriers inMODE3In order to simplify the design Guard Interval (GI) length ofa half size of OFDM symbol length 119879 is used

32 Time and Frequency Impulsive Noise Cancelling Becauseof the presence of Impulsive Noise so-called Shrimp Noise

Table 1 System parameters

Parameters Mode2 3

TX-RX elements 1 TX and 4RX transducerSampling frequency 96000HzTX center frequency 24000HzBand width 8000HzFFT size 1024 2048OFDM symbol length 119879 10667ms 21333msGI length 05119879 05119879Subcarrier spacing 9375Hz 46875HzNumber of subcarriers 81 161

time and frequency domain Impulsive Noise Cancelling(time-IMP and freq-IMP) methods [4 5] are employed

The Scientific World Journal 5

Noise amplitude0 005 01 015 02 025

(B)(A) (C)

Shizuoka depth = 3m

102

101

100

10minus1

10minus2

10minus3

10minus4

10minus5

Accu

mul

atio

nfro

m+

infin

ity(

)

(a)

0 001 002 003 004 005 006 007

(A)

Accu

mul

atio

nfro

m+

infin

ity(

)

Noise amplitude

(C)(B)

102

101

100

10minus1

10minus2

10minus3

10minus4

10minus5

Shizuoka depth = 9m

(b)

Noise amplitude0 0005 001 0015 002 0025

(B)(A) (C)

Okinawa

Accu

mul

atio

nfro

m+

infin

ity(

)

102

101

100

10minus1

10minus2

10minus3

10minus4

(c)

Figure 5 (a) Accumulated background noise from + infinity at Shizuoka (depth = 3m) (b) Accumulated background noise from + infinityat Shizuoka (depth = 9m) (c) Accumulated background noise from + infinity at Okinawa

and further combined with carrier diversity combiner Theblock diagram of time-IMP (circle-A in Figure 6) is shownin Figure 7(a) It detects Impulsive Noise samples in timedomain and then removes the Impulsive Noise samplesaccording to

119910 (119899) =

119903 (119899) if |119903 (119899)|2 le 120572 sdot 119875avg0 otherwise

(2)

Here 119903(119899) is 119899th time sample of received signal after pream-plifier 119875avg is the average power of an OFDM symbol and120572 is threshold parameter Obviously if the power of inputsignal 119903(119899) exceeds the threshold of 120572 sdot 119875avg the signal isreplaced by zero This method is called ldquoblankingrdquo [10 11]In order to determine the threshold parameter 120572 cut andtry experimentation was applied and finally 120572 = 20 isdeterminedThe system performance is not so sensitive to theparameter

In addition a frequency domain Impulsive Noise Cancel-lation is appliedThe detail of freq-IMP (circle-B in Figure 6)is shown in Figure 7(b) The 1st MRC block generates thecombined consternation of 4 branches equalizers Then thefollowing processing is applied in hard decision and pilotinsertion block

(1) Subcarriers which should be silent (no data or pilots)are set to zero

(2) Subcarriers which are used as pilots are replaced byknown pilot values

(3) Subcarriers which are used for data transmission aredemapped to nearest position in constellation plot

Then succeeding subtraction block computes frequencydomain noise components In the following blocks thefrequency domain noise components are converted to a TimeDomain Impulsive Noise The frequency domain noise is

6 The Scientific World Journal

Bitinformation Map Pilot

insertion IFFTGI

addition upconversion

TX transducer

Preamp-

4 RX transducers

Underwater channel

Down-

FFT

FFT

FFT

FFT

Equalize

Equalize

Equalize

Equalize

A

A

A

A

Poweramplification

Time DomainImpulsive Noise Canceller

Frequency DomainDiversity Combined

Impulsive Noise Canceller1stMRCwithnoiseest

2ndMRCwithnoiseest

De-map

Bitinformation

B

B

B

B

Re-move

GI

Re-move

GI

Re-move

GI

Re-move

GI

Pilotdata

Pilotdata

Data

Pilot

Pilot

data

Channelestimation

Channelestimation

Channelestimation

Channelestimation

lification

Preamp-lification

Preamp-lification

Preamp-lification

conversion

Down-conversion

Down-conversion

Down-conversion

Figure 6 Block diagram of OFDM communication system with Frequency Domain Diversity Combined Impulsive Noise Canceller

multiplied with Channel Transfer Function of branch 119894 (119867119894

119894 = 1 to 4) and is IFFTed If the output of IFFThas a large peakit can be considered as estimated Time Domain ImpulsiveNoise signal 119906(119899) according to

119906 (119899) =

119889 (119899) if 10038161003816100381610038161003816119889 (119899)10038161003816100381610038161003816

2ge 120573 sdot 120590

2

0 otherwise(3)

Here 119889(119899) is 119899th time sample of IFFT output 1205902 is the averagepower of the IFFT output and 120573 is threshold parameterObviously if the power of input signal 119889(119899) exceeds thethreshold of 120573 sdot 1205902 the signal is considered as an ImpulsiveNoise Then the estimated Time Domain Impulsive Noisesignal 119906(119899) is converted to frequency domain by FFT blockand division by 119867

119894 Finally the estimated frequency domain

Impulsive Noise is withdrawn from equalized signal ofbranch 119894 (119894 = 1 to 4) EQA

119894 Then the noise reduced outputs

EQB119894(119894 = 1 to 4) are obtained and then they are combined

in the 2nd MRC block to generate further noise reducedconstellation The parameter 120573 is determined by cut and tryexperimentations and 120573 = 8 is finally applied to the systemThe obtained noise cancel performance is not so sensitive tothe parameter 120573

4 Experimental Results

Table 2 summarizes ocean experiment parameters for bothOkinawa OJIMA fishing port site and Shizuoka UCHIURAbarge site In Okinawa site horizontal underwater com-munication is carried out while vertical communication isexperimented in Shizuoka site for The four RX transducersare placed as linear array with 20 cm pitch Figures 8(a) and8(b) are MODE3 and MODE2 constellations in Okinawasite respectively Upper four constellations correspond tobranches 1 to 4 2nd MRC inputs The bottom constellationis 2nd MRC output The horizontal axis of each figure showsreal part of digital modulation while the vertical axis corre-sponds to imaginary part According to the figures carrierdiversity combining MRC successfully reduces variance ofconstellation points

Figures 9(a) and 9(b) are the OFDM symbol by symbolbit error rate (BER) comparisons between frequency domainImpulsive Noise cancellation ON and OFF for MODE316QAM modulation in Okinawa site and MODE2 16QAMin Shizuoka site respectively The solid line corresponds tofreq-IMP OFF and the dashed line corresponds to ONHorizontal axis shows BER and vertical axis shows OFDMsymbol number According to the figures the proposedFrequency Domain Diversity Combined Impulsive NoiseCanceller effectively reduces symbol by symbol BER values

The Scientific World Journal 7

Preamplificationoutput

0 Yes

No

Downconverter

gtthreshold|and2|

(a) Time Domain Impulsive Noise Canceller

Harddecisionand pilotinsertion

+ IFFTPeak

detect FFT

1X

+

minus

Hminus1

1stMRCwithnoise

est

xx

Estimatedfreq

domainnoise

++

minus

EQAi EQBi

Hi

(b) Frequency Domain Impulsive Noise Canceller

Figure 7 Time and Frequency Domain Impulsive Noise Canceller

Table 2 Ocean experiment parameters

Parameters PlaceOkinawa Shizuoka

Type ofexperiment site Fishing port Barge

Number oftransducers 1 TX and 4RX 1 TX and 4RX

RX transducerarray Linear with 20 cm pitch Linear with 20 cm pitch

Modulation QPSK16QAM QPSK16QAMTX-RX distance 715m 13mTransmissiondirection Horizontal Vertical

Ocean depth 5m 30mTransducerdepth (TXRX) 2m2m 22m9m

Figure 10 summarizes BER comparisons for FrequencyDomain Diversity Combined Impulsive Noise Canceller forfreq-IMP ON and OFF In spite of place of experiment siteapproximately 20ndash30 of BER reduction has been obtainedby the proposed Frequency Domain Diversity CombinedImpulsive Noise Canceller

Finally Figures 11(a) and 11(b) showmeasured communi-cation system performance in Okinawa site with enabling theproposed Impulsive Noise Canceller In addition BER data

is also measured with moving the TX transducer speed suchas 00ms (stable) 03 and 06ms for QPSK modulationand 00ms (stable) and 03ms for 16QAM Since MODE2OFDM symbol length of 10667ms as shown in Table 1 is halfof MODE3 the Guard Interval length of MODE2 is shorterthan MODE3 Therefore it is considered that high levelmodulation of 16QAM of MODE2 is largely affected by InterSymbol Interference (ISI) As you can see from those figuresMODE2 shows higher robustness for moving cases such asstrangeness for Doppler Effect because of shorter OFDMsymbol length of MODE2 Once forward error correctionsuch as TURBO CODE with code rate 119877 = 13 is appliedto this system all error free communications (BER = 0) forQPSK modulation of all speeds and 16QAM stable case havebeen confirmed

5 Conclusion

In this paper an ultrasonic OFDM transceiver architecturewith four diverse receivers supporting 2 MODEs is proposedwith enhancement of Impulsive Noise insusceptibility Itutilizes 20ndash28KHz ultrasonic channel and subcarrier spacingof 46875Hz (MODE3) and 93750Hz (MODE2) OFDMmodulations

Two different environment ocean experiments were con-ducted to evaluate the statistics of Impulsive Noise so-called Shrimp Noise which is generated by living creaturesin marine environment To deal with challenges posed by

8 The Scientific World Journal

minus1

0

1

1

minus1 0 1

minus1

0

minus1 0 1

1

minus1

0

minus1 0 1

Branch 1

Branch 3

minus1

0

1

minus1 0 1

minus1

0

1

minus1 0 1

Branch 2

Branch 4

2nd MRC (MODE3)

(a)

minus1

0

1

minus1 0 1

Branch 1

minus1

0

1

minus1 0 1

minus1

0

1

minus1 0 1

Branch 3

minus1

0

1

minus1 0 1

minus1

0

1

minus1 0 1

Branch 2

Branch 4

2nd MRC (MODE2)

(b)

Figure 8 (a)MODE3QPSK constellations (branches 1 to 4 andMRCOkinawa) (b)MODE2QPSK constellations (branches 1 to 4 andMRCOkinawa)

The Scientific World Journal 9

5 10 15 20 25 30 35 40

Symbol number

BER

OFF F-IMP CAN

ON F-IMP CAN

100

10minus1

10minus2

10minus3

(a)

10 20 30 40 50 60 70 80

Symbol number

BER

OFF F-IMP CAN

ON F-IMP CAN

100

10minus1

10minus2

10minus3

(b)

Figure 9 (a) MODE3 BER comparison for 16QAM at Okinawa (b) MODE2 BER comparison for 16QAM at Shizuoka

0

002

004

006

008

01

012

014

QPSKOkinawa

BER

16QAMOkinawa

QPSKOkinawa

16QAMOkinawa

16QAMShizuoka

16QAMShizuoka

OFF F-IMP CANON F-IMP CAN

MODE3 MODE2 MODE2 MODE2MODE3 MODE3

Figure 10 Summary of BER comparison for Frequency Domain Diversity Combined Impulsive Noise Canceller

00001

0001

001

01

10

4 diversity RX transducers liner array with 20 cm pitchQPSK modulationDistance = 715mOkinawa sea experiment

00 03 06

MODE3

MODE2

TX moving speed (ms)

Bit e

rror

rate

(a)

00001

0001

001

01

10

00 03

MODE3

MODE2

16QAM modulationDistance = 715mOkinawa sea experiment

TX moving speed (ms)

Bit e

rror

rate

4 diversity RX transducers liner array with 20 cm pitch

(b)

Figure 11 (a) Communication system performance using QPSK modulation (b) Communication system performance using 16QAMmodulation

10 The Scientific World Journal

the Impulsive Noise interference we have employed a Fre-quency Domain Diversity Combined Impulsive Noise Can-celler

Measured data at a fishing port in Okinawa and ata barge in Shizuoka Prefecture are disclosed and in bothenvironment randomly generated Time Domain ImpulseNoise distribution is observed Although more than 90 to99 of background noise shows AWGN type distributionless than 1 to 10 noise components show large amplitudesuch as Impulsive Noise

By enabling the proposed MRC combined Frequencydomain Impulsive Noise Canceller 20ndash30 BER reductionhas been successfully obtained In addition 715m shallowfishing port horizontal underwater communication BERis also shown using QPSK and 16QAM modulations Byapplying TURBO 119877 = 13 forward error correction allerror free communications using QPSK and 16QAMwith notmoving case are also confirmed

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgment

This study has been carried out as part of the StrategicInformation andCommunications RampDPromotion Program(SCOPE) Project of the Ministry of Internal Affairs andCommunications Japan

References

[1] M Stojanovic ldquoRecent advances in high-speed underwateracoustic communicationsrdquo IEEE Journal of Oceanic Engineer-ing vol 21 no 2 pp 125ndash136 1996

[2] G Qiao W Wang R Guo R Khan and Y Wang ldquoFrequencydiversity for OFDM mobile communication via underwateracoustic channelsrdquo Journal of Marine Science and Applicationvol 11 no 1 pp 126ndash133 2012

[3] G Leus and P A van Walree ldquoMultiband OFDM for covertacoustic communicationsrdquo IEEE Journal on Selected Areas inCommunications vol 26 no 9 pp 1662ndash1673 2008

[4] H D Trung and N Van Duc ldquoAn analysis of MIMO-OFDMfor underwater communicationsrdquo in Proceedings of the 3rdInternational Congress on Ultra Modern Telecommunicationsand Control Systems and Workshops (ICUMT rsquo11) BudapestHungary October 2011

[5] T Suzuki H M Tran and T Wada ldquoAn underwater acousticOFDM communication system with shrimp (impulsive) noisecancellingrdquo in Proceedings of the IEEE International Conferenceon Computing Management and Telecommunications (Com-ManTel rsquo14) pp 152ndash156 Da Nang Vietnam April 2014

[6] T M Hai Y Matsuda T Suzuki and W A D A Tomo-hisa ldquoUltrasonic diversity OFDM transceiver architecture withimpulsive noise cancelling for shallow sea communicationrdquo inProceedings of the UA2014-2nd International Conference andExhibition on Underwater Acoustics Rodes Greece June 2014

[7] J Gao T IidaHMizutani et al ldquoThe implementation of a 248antennas configurable diversity OFDM receiver for mobile

HDTV applicationrdquo International Journal of Digital MultimediaBroadcasting vol 2009 Article ID 303627 9 pages 2009

[8] T Koujita T Wada T Iida et al ldquoEvaluation of diversitycombining systems for mobile reception in digital terrestrialtelevision broadcastingrdquo in Proceedings of the IEEE Inter-national Conference on Consumer Electronics SESSION 63Digital Receivers I Las Vegas Nev USA January 2009

[9] T Wada ldquoDiversity and error compensation in OFDMtransceivers principles and implementationrdquo in Digital Front-End in Wireless Communication and Broadcasting chapter 18Cambridge University Press 2011

[10] S V Zhidkov ldquoImpulsive noise suppression in OFDM basedcommunication systemsrdquo IEEE Transactions on Consumer Elec-tronics vol 49 no 4 pp 944ndash948 2003

[11] K Al-Mawali A Z Sadik and Z M Hussain ldquoJointtime-domainfrequency-domain impulsive noise reduction inOFDM-based power line communicationsrdquo in Proceedings ofthe Australasian Telecommunication Networks and ApplicationsConference (ATNAC rsquo08) pp 138ndash142 Adelaide AustralasianDecember 2008

Submit your manuscripts athttpwwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Mechanical Engineering

Advances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Distributed Sensor Networks

International Journal of

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Antennas andPropagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

The Scientific World Journal 5

Noise amplitude0 005 01 015 02 025

(B)(A) (C)

Shizuoka depth = 3m

102

101

100

10minus1

10minus2

10minus3

10minus4

10minus5

Accu

mul

atio

nfro

m+

infin

ity(

)

(a)

0 001 002 003 004 005 006 007

(A)

Accu

mul

atio

nfro

m+

infin

ity(

)

Noise amplitude

(C)(B)

102

101

100

10minus1

10minus2

10minus3

10minus4

10minus5

Shizuoka depth = 9m

(b)

Noise amplitude0 0005 001 0015 002 0025

(B)(A) (C)

Okinawa

Accu

mul

atio

nfro

m+

infin

ity(

)

102

101

100

10minus1

10minus2

10minus3

10minus4

(c)

Figure 5 (a) Accumulated background noise from + infinity at Shizuoka (depth = 3m) (b) Accumulated background noise from + infinityat Shizuoka (depth = 9m) (c) Accumulated background noise from + infinity at Okinawa

and further combined with carrier diversity combiner Theblock diagram of time-IMP (circle-A in Figure 6) is shownin Figure 7(a) It detects Impulsive Noise samples in timedomain and then removes the Impulsive Noise samplesaccording to

119910 (119899) =

119903 (119899) if |119903 (119899)|2 le 120572 sdot 119875avg0 otherwise

(2)

Here 119903(119899) is 119899th time sample of received signal after pream-plifier 119875avg is the average power of an OFDM symbol and120572 is threshold parameter Obviously if the power of inputsignal 119903(119899) exceeds the threshold of 120572 sdot 119875avg the signal isreplaced by zero This method is called ldquoblankingrdquo [10 11]In order to determine the threshold parameter 120572 cut andtry experimentation was applied and finally 120572 = 20 isdeterminedThe system performance is not so sensitive to theparameter

In addition a frequency domain Impulsive Noise Cancel-lation is appliedThe detail of freq-IMP (circle-B in Figure 6)is shown in Figure 7(b) The 1st MRC block generates thecombined consternation of 4 branches equalizers Then thefollowing processing is applied in hard decision and pilotinsertion block

(1) Subcarriers which should be silent (no data or pilots)are set to zero

(2) Subcarriers which are used as pilots are replaced byknown pilot values

(3) Subcarriers which are used for data transmission aredemapped to nearest position in constellation plot

Then succeeding subtraction block computes frequencydomain noise components In the following blocks thefrequency domain noise components are converted to a TimeDomain Impulsive Noise The frequency domain noise is

6 The Scientific World Journal

Bitinformation Map Pilot

insertion IFFTGI

addition upconversion

TX transducer

Preamp-

4 RX transducers

Underwater channel

Down-

FFT

FFT

FFT

FFT

Equalize

Equalize

Equalize

Equalize

A

A

A

A

Poweramplification

Time DomainImpulsive Noise Canceller

Frequency DomainDiversity Combined

Impulsive Noise Canceller1stMRCwithnoiseest

2ndMRCwithnoiseest

De-map

Bitinformation

B

B

B

B

Re-move

GI

Re-move

GI

Re-move

GI

Re-move

GI

Pilotdata

Pilotdata

Data

Pilot

Pilot

data

Channelestimation

Channelestimation

Channelestimation

Channelestimation

lification

Preamp-lification

Preamp-lification

Preamp-lification

conversion

Down-conversion

Down-conversion

Down-conversion

Figure 6 Block diagram of OFDM communication system with Frequency Domain Diversity Combined Impulsive Noise Canceller

multiplied with Channel Transfer Function of branch 119894 (119867119894

119894 = 1 to 4) and is IFFTed If the output of IFFThas a large peakit can be considered as estimated Time Domain ImpulsiveNoise signal 119906(119899) according to

119906 (119899) =

119889 (119899) if 10038161003816100381610038161003816119889 (119899)10038161003816100381610038161003816

2ge 120573 sdot 120590

2

0 otherwise(3)

Here 119889(119899) is 119899th time sample of IFFT output 1205902 is the averagepower of the IFFT output and 120573 is threshold parameterObviously if the power of input signal 119889(119899) exceeds thethreshold of 120573 sdot 1205902 the signal is considered as an ImpulsiveNoise Then the estimated Time Domain Impulsive Noisesignal 119906(119899) is converted to frequency domain by FFT blockand division by 119867

119894 Finally the estimated frequency domain

Impulsive Noise is withdrawn from equalized signal ofbranch 119894 (119894 = 1 to 4) EQA

119894 Then the noise reduced outputs

EQB119894(119894 = 1 to 4) are obtained and then they are combined

in the 2nd MRC block to generate further noise reducedconstellation The parameter 120573 is determined by cut and tryexperimentations and 120573 = 8 is finally applied to the systemThe obtained noise cancel performance is not so sensitive tothe parameter 120573

4 Experimental Results

Table 2 summarizes ocean experiment parameters for bothOkinawa OJIMA fishing port site and Shizuoka UCHIURAbarge site In Okinawa site horizontal underwater com-munication is carried out while vertical communication isexperimented in Shizuoka site for The four RX transducersare placed as linear array with 20 cm pitch Figures 8(a) and8(b) are MODE3 and MODE2 constellations in Okinawasite respectively Upper four constellations correspond tobranches 1 to 4 2nd MRC inputs The bottom constellationis 2nd MRC output The horizontal axis of each figure showsreal part of digital modulation while the vertical axis corre-sponds to imaginary part According to the figures carrierdiversity combining MRC successfully reduces variance ofconstellation points

Figures 9(a) and 9(b) are the OFDM symbol by symbolbit error rate (BER) comparisons between frequency domainImpulsive Noise cancellation ON and OFF for MODE316QAM modulation in Okinawa site and MODE2 16QAMin Shizuoka site respectively The solid line corresponds tofreq-IMP OFF and the dashed line corresponds to ONHorizontal axis shows BER and vertical axis shows OFDMsymbol number According to the figures the proposedFrequency Domain Diversity Combined Impulsive NoiseCanceller effectively reduces symbol by symbol BER values

The Scientific World Journal 7

Preamplificationoutput

0 Yes

No

Downconverter

gtthreshold|and2|

(a) Time Domain Impulsive Noise Canceller

Harddecisionand pilotinsertion

+ IFFTPeak

detect FFT

1X

+

minus

Hminus1

1stMRCwithnoise

est

xx

Estimatedfreq

domainnoise

++

minus

EQAi EQBi

Hi

(b) Frequency Domain Impulsive Noise Canceller

Figure 7 Time and Frequency Domain Impulsive Noise Canceller

Table 2 Ocean experiment parameters

Parameters PlaceOkinawa Shizuoka

Type ofexperiment site Fishing port Barge

Number oftransducers 1 TX and 4RX 1 TX and 4RX

RX transducerarray Linear with 20 cm pitch Linear with 20 cm pitch

Modulation QPSK16QAM QPSK16QAMTX-RX distance 715m 13mTransmissiondirection Horizontal Vertical

Ocean depth 5m 30mTransducerdepth (TXRX) 2m2m 22m9m

Figure 10 summarizes BER comparisons for FrequencyDomain Diversity Combined Impulsive Noise Canceller forfreq-IMP ON and OFF In spite of place of experiment siteapproximately 20ndash30 of BER reduction has been obtainedby the proposed Frequency Domain Diversity CombinedImpulsive Noise Canceller

Finally Figures 11(a) and 11(b) showmeasured communi-cation system performance in Okinawa site with enabling theproposed Impulsive Noise Canceller In addition BER data

is also measured with moving the TX transducer speed suchas 00ms (stable) 03 and 06ms for QPSK modulationand 00ms (stable) and 03ms for 16QAM Since MODE2OFDM symbol length of 10667ms as shown in Table 1 is halfof MODE3 the Guard Interval length of MODE2 is shorterthan MODE3 Therefore it is considered that high levelmodulation of 16QAM of MODE2 is largely affected by InterSymbol Interference (ISI) As you can see from those figuresMODE2 shows higher robustness for moving cases such asstrangeness for Doppler Effect because of shorter OFDMsymbol length of MODE2 Once forward error correctionsuch as TURBO CODE with code rate 119877 = 13 is appliedto this system all error free communications (BER = 0) forQPSK modulation of all speeds and 16QAM stable case havebeen confirmed

5 Conclusion

In this paper an ultrasonic OFDM transceiver architecturewith four diverse receivers supporting 2 MODEs is proposedwith enhancement of Impulsive Noise insusceptibility Itutilizes 20ndash28KHz ultrasonic channel and subcarrier spacingof 46875Hz (MODE3) and 93750Hz (MODE2) OFDMmodulations

Two different environment ocean experiments were con-ducted to evaluate the statistics of Impulsive Noise so-called Shrimp Noise which is generated by living creaturesin marine environment To deal with challenges posed by

8 The Scientific World Journal

minus1

0

1

1

minus1 0 1

minus1

0

minus1 0 1

1

minus1

0

minus1 0 1

Branch 1

Branch 3

minus1

0

1

minus1 0 1

minus1

0

1

minus1 0 1

Branch 2

Branch 4

2nd MRC (MODE3)

(a)

minus1

0

1

minus1 0 1

Branch 1

minus1

0

1

minus1 0 1

minus1

0

1

minus1 0 1

Branch 3

minus1

0

1

minus1 0 1

minus1

0

1

minus1 0 1

Branch 2

Branch 4

2nd MRC (MODE2)

(b)

Figure 8 (a)MODE3QPSK constellations (branches 1 to 4 andMRCOkinawa) (b)MODE2QPSK constellations (branches 1 to 4 andMRCOkinawa)

The Scientific World Journal 9

5 10 15 20 25 30 35 40

Symbol number

BER

OFF F-IMP CAN

ON F-IMP CAN

100

10minus1

10minus2

10minus3

(a)

10 20 30 40 50 60 70 80

Symbol number

BER

OFF F-IMP CAN

ON F-IMP CAN

100

10minus1

10minus2

10minus3

(b)

Figure 9 (a) MODE3 BER comparison for 16QAM at Okinawa (b) MODE2 BER comparison for 16QAM at Shizuoka

0

002

004

006

008

01

012

014

QPSKOkinawa

BER

16QAMOkinawa

QPSKOkinawa

16QAMOkinawa

16QAMShizuoka

16QAMShizuoka

OFF F-IMP CANON F-IMP CAN

MODE3 MODE2 MODE2 MODE2MODE3 MODE3

Figure 10 Summary of BER comparison for Frequency Domain Diversity Combined Impulsive Noise Canceller

00001

0001

001

01

10

4 diversity RX transducers liner array with 20 cm pitchQPSK modulationDistance = 715mOkinawa sea experiment

00 03 06

MODE3

MODE2

TX moving speed (ms)

Bit e

rror

rate

(a)

00001

0001

001

01

10

00 03

MODE3

MODE2

16QAM modulationDistance = 715mOkinawa sea experiment

TX moving speed (ms)

Bit e

rror

rate

4 diversity RX transducers liner array with 20 cm pitch

(b)

Figure 11 (a) Communication system performance using QPSK modulation (b) Communication system performance using 16QAMmodulation

10 The Scientific World Journal

the Impulsive Noise interference we have employed a Fre-quency Domain Diversity Combined Impulsive Noise Can-celler

Measured data at a fishing port in Okinawa and ata barge in Shizuoka Prefecture are disclosed and in bothenvironment randomly generated Time Domain ImpulseNoise distribution is observed Although more than 90 to99 of background noise shows AWGN type distributionless than 1 to 10 noise components show large amplitudesuch as Impulsive Noise

By enabling the proposed MRC combined Frequencydomain Impulsive Noise Canceller 20ndash30 BER reductionhas been successfully obtained In addition 715m shallowfishing port horizontal underwater communication BERis also shown using QPSK and 16QAM modulations Byapplying TURBO 119877 = 13 forward error correction allerror free communications using QPSK and 16QAMwith notmoving case are also confirmed

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgment

This study has been carried out as part of the StrategicInformation andCommunications RampDPromotion Program(SCOPE) Project of the Ministry of Internal Affairs andCommunications Japan

References

[1] M Stojanovic ldquoRecent advances in high-speed underwateracoustic communicationsrdquo IEEE Journal of Oceanic Engineer-ing vol 21 no 2 pp 125ndash136 1996

[2] G Qiao W Wang R Guo R Khan and Y Wang ldquoFrequencydiversity for OFDM mobile communication via underwateracoustic channelsrdquo Journal of Marine Science and Applicationvol 11 no 1 pp 126ndash133 2012

[3] G Leus and P A van Walree ldquoMultiband OFDM for covertacoustic communicationsrdquo IEEE Journal on Selected Areas inCommunications vol 26 no 9 pp 1662ndash1673 2008

[4] H D Trung and N Van Duc ldquoAn analysis of MIMO-OFDMfor underwater communicationsrdquo in Proceedings of the 3rdInternational Congress on Ultra Modern Telecommunicationsand Control Systems and Workshops (ICUMT rsquo11) BudapestHungary October 2011

[5] T Suzuki H M Tran and T Wada ldquoAn underwater acousticOFDM communication system with shrimp (impulsive) noisecancellingrdquo in Proceedings of the IEEE International Conferenceon Computing Management and Telecommunications (Com-ManTel rsquo14) pp 152ndash156 Da Nang Vietnam April 2014

[6] T M Hai Y Matsuda T Suzuki and W A D A Tomo-hisa ldquoUltrasonic diversity OFDM transceiver architecture withimpulsive noise cancelling for shallow sea communicationrdquo inProceedings of the UA2014-2nd International Conference andExhibition on Underwater Acoustics Rodes Greece June 2014

[7] J Gao T IidaHMizutani et al ldquoThe implementation of a 248antennas configurable diversity OFDM receiver for mobile

HDTV applicationrdquo International Journal of Digital MultimediaBroadcasting vol 2009 Article ID 303627 9 pages 2009

[8] T Koujita T Wada T Iida et al ldquoEvaluation of diversitycombining systems for mobile reception in digital terrestrialtelevision broadcastingrdquo in Proceedings of the IEEE Inter-national Conference on Consumer Electronics SESSION 63Digital Receivers I Las Vegas Nev USA January 2009

[9] T Wada ldquoDiversity and error compensation in OFDMtransceivers principles and implementationrdquo in Digital Front-End in Wireless Communication and Broadcasting chapter 18Cambridge University Press 2011

[10] S V Zhidkov ldquoImpulsive noise suppression in OFDM basedcommunication systemsrdquo IEEE Transactions on Consumer Elec-tronics vol 49 no 4 pp 944ndash948 2003

[11] K Al-Mawali A Z Sadik and Z M Hussain ldquoJointtime-domainfrequency-domain impulsive noise reduction inOFDM-based power line communicationsrdquo in Proceedings ofthe Australasian Telecommunication Networks and ApplicationsConference (ATNAC rsquo08) pp 138ndash142 Adelaide AustralasianDecember 2008

Submit your manuscripts athttpwwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Mechanical Engineering

Advances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Distributed Sensor Networks

International Journal of

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Antennas andPropagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

6 The Scientific World Journal

Bitinformation Map Pilot

insertion IFFTGI

addition upconversion

TX transducer

Preamp-

4 RX transducers

Underwater channel

Down-

FFT

FFT

FFT

FFT

Equalize

Equalize

Equalize

Equalize

A

A

A

A

Poweramplification

Time DomainImpulsive Noise Canceller

Frequency DomainDiversity Combined

Impulsive Noise Canceller1stMRCwithnoiseest

2ndMRCwithnoiseest

De-map

Bitinformation

B

B

B

B

Re-move

GI

Re-move

GI

Re-move

GI

Re-move

GI

Pilotdata

Pilotdata

Data

Pilot

Pilot

data

Channelestimation

Channelestimation

Channelestimation

Channelestimation

lification

Preamp-lification

Preamp-lification

Preamp-lification

conversion

Down-conversion

Down-conversion

Down-conversion

Figure 6 Block diagram of OFDM communication system with Frequency Domain Diversity Combined Impulsive Noise Canceller

multiplied with Channel Transfer Function of branch 119894 (119867119894

119894 = 1 to 4) and is IFFTed If the output of IFFThas a large peakit can be considered as estimated Time Domain ImpulsiveNoise signal 119906(119899) according to

119906 (119899) =

119889 (119899) if 10038161003816100381610038161003816119889 (119899)10038161003816100381610038161003816

2ge 120573 sdot 120590

2

0 otherwise(3)

Here 119889(119899) is 119899th time sample of IFFT output 1205902 is the averagepower of the IFFT output and 120573 is threshold parameterObviously if the power of input signal 119889(119899) exceeds thethreshold of 120573 sdot 1205902 the signal is considered as an ImpulsiveNoise Then the estimated Time Domain Impulsive Noisesignal 119906(119899) is converted to frequency domain by FFT blockand division by 119867

119894 Finally the estimated frequency domain

Impulsive Noise is withdrawn from equalized signal ofbranch 119894 (119894 = 1 to 4) EQA

119894 Then the noise reduced outputs

EQB119894(119894 = 1 to 4) are obtained and then they are combined

in the 2nd MRC block to generate further noise reducedconstellation The parameter 120573 is determined by cut and tryexperimentations and 120573 = 8 is finally applied to the systemThe obtained noise cancel performance is not so sensitive tothe parameter 120573

4 Experimental Results

Table 2 summarizes ocean experiment parameters for bothOkinawa OJIMA fishing port site and Shizuoka UCHIURAbarge site In Okinawa site horizontal underwater com-munication is carried out while vertical communication isexperimented in Shizuoka site for The four RX transducersare placed as linear array with 20 cm pitch Figures 8(a) and8(b) are MODE3 and MODE2 constellations in Okinawasite respectively Upper four constellations correspond tobranches 1 to 4 2nd MRC inputs The bottom constellationis 2nd MRC output The horizontal axis of each figure showsreal part of digital modulation while the vertical axis corre-sponds to imaginary part According to the figures carrierdiversity combining MRC successfully reduces variance ofconstellation points

Figures 9(a) and 9(b) are the OFDM symbol by symbolbit error rate (BER) comparisons between frequency domainImpulsive Noise cancellation ON and OFF for MODE316QAM modulation in Okinawa site and MODE2 16QAMin Shizuoka site respectively The solid line corresponds tofreq-IMP OFF and the dashed line corresponds to ONHorizontal axis shows BER and vertical axis shows OFDMsymbol number According to the figures the proposedFrequency Domain Diversity Combined Impulsive NoiseCanceller effectively reduces symbol by symbol BER values

The Scientific World Journal 7

Preamplificationoutput

0 Yes

No

Downconverter

gtthreshold|and2|

(a) Time Domain Impulsive Noise Canceller

Harddecisionand pilotinsertion

+ IFFTPeak

detect FFT

1X

+

minus

Hminus1

1stMRCwithnoise

est

xx

Estimatedfreq

domainnoise

++

minus

EQAi EQBi

Hi

(b) Frequency Domain Impulsive Noise Canceller

Figure 7 Time and Frequency Domain Impulsive Noise Canceller

Table 2 Ocean experiment parameters

Parameters PlaceOkinawa Shizuoka

Type ofexperiment site Fishing port Barge

Number oftransducers 1 TX and 4RX 1 TX and 4RX

RX transducerarray Linear with 20 cm pitch Linear with 20 cm pitch

Modulation QPSK16QAM QPSK16QAMTX-RX distance 715m 13mTransmissiondirection Horizontal Vertical

Ocean depth 5m 30mTransducerdepth (TXRX) 2m2m 22m9m

Figure 10 summarizes BER comparisons for FrequencyDomain Diversity Combined Impulsive Noise Canceller forfreq-IMP ON and OFF In spite of place of experiment siteapproximately 20ndash30 of BER reduction has been obtainedby the proposed Frequency Domain Diversity CombinedImpulsive Noise Canceller

Finally Figures 11(a) and 11(b) showmeasured communi-cation system performance in Okinawa site with enabling theproposed Impulsive Noise Canceller In addition BER data

is also measured with moving the TX transducer speed suchas 00ms (stable) 03 and 06ms for QPSK modulationand 00ms (stable) and 03ms for 16QAM Since MODE2OFDM symbol length of 10667ms as shown in Table 1 is halfof MODE3 the Guard Interval length of MODE2 is shorterthan MODE3 Therefore it is considered that high levelmodulation of 16QAM of MODE2 is largely affected by InterSymbol Interference (ISI) As you can see from those figuresMODE2 shows higher robustness for moving cases such asstrangeness for Doppler Effect because of shorter OFDMsymbol length of MODE2 Once forward error correctionsuch as TURBO CODE with code rate 119877 = 13 is appliedto this system all error free communications (BER = 0) forQPSK modulation of all speeds and 16QAM stable case havebeen confirmed

5 Conclusion

In this paper an ultrasonic OFDM transceiver architecturewith four diverse receivers supporting 2 MODEs is proposedwith enhancement of Impulsive Noise insusceptibility Itutilizes 20ndash28KHz ultrasonic channel and subcarrier spacingof 46875Hz (MODE3) and 93750Hz (MODE2) OFDMmodulations

Two different environment ocean experiments were con-ducted to evaluate the statistics of Impulsive Noise so-called Shrimp Noise which is generated by living creaturesin marine environment To deal with challenges posed by

8 The Scientific World Journal

minus1

0

1

1

minus1 0 1

minus1

0

minus1 0 1

1

minus1

0

minus1 0 1

Branch 1

Branch 3

minus1

0

1

minus1 0 1

minus1

0

1

minus1 0 1

Branch 2

Branch 4

2nd MRC (MODE3)

(a)

minus1

0

1

minus1 0 1

Branch 1

minus1

0

1

minus1 0 1

minus1

0

1

minus1 0 1

Branch 3

minus1

0

1

minus1 0 1

minus1

0

1

minus1 0 1

Branch 2

Branch 4

2nd MRC (MODE2)

(b)

Figure 8 (a)MODE3QPSK constellations (branches 1 to 4 andMRCOkinawa) (b)MODE2QPSK constellations (branches 1 to 4 andMRCOkinawa)

The Scientific World Journal 9

5 10 15 20 25 30 35 40

Symbol number

BER

OFF F-IMP CAN

ON F-IMP CAN

100

10minus1

10minus2

10minus3

(a)

10 20 30 40 50 60 70 80

Symbol number

BER

OFF F-IMP CAN

ON F-IMP CAN

100

10minus1

10minus2

10minus3

(b)

Figure 9 (a) MODE3 BER comparison for 16QAM at Okinawa (b) MODE2 BER comparison for 16QAM at Shizuoka

0

002

004

006

008

01

012

014

QPSKOkinawa

BER

16QAMOkinawa

QPSKOkinawa

16QAMOkinawa

16QAMShizuoka

16QAMShizuoka

OFF F-IMP CANON F-IMP CAN

MODE3 MODE2 MODE2 MODE2MODE3 MODE3

Figure 10 Summary of BER comparison for Frequency Domain Diversity Combined Impulsive Noise Canceller

00001

0001

001

01

10

4 diversity RX transducers liner array with 20 cm pitchQPSK modulationDistance = 715mOkinawa sea experiment

00 03 06

MODE3

MODE2

TX moving speed (ms)

Bit e

rror

rate

(a)

00001

0001

001

01

10

00 03

MODE3

MODE2

16QAM modulationDistance = 715mOkinawa sea experiment

TX moving speed (ms)

Bit e

rror

rate

4 diversity RX transducers liner array with 20 cm pitch

(b)

Figure 11 (a) Communication system performance using QPSK modulation (b) Communication system performance using 16QAMmodulation

10 The Scientific World Journal

the Impulsive Noise interference we have employed a Fre-quency Domain Diversity Combined Impulsive Noise Can-celler

Measured data at a fishing port in Okinawa and ata barge in Shizuoka Prefecture are disclosed and in bothenvironment randomly generated Time Domain ImpulseNoise distribution is observed Although more than 90 to99 of background noise shows AWGN type distributionless than 1 to 10 noise components show large amplitudesuch as Impulsive Noise

By enabling the proposed MRC combined Frequencydomain Impulsive Noise Canceller 20ndash30 BER reductionhas been successfully obtained In addition 715m shallowfishing port horizontal underwater communication BERis also shown using QPSK and 16QAM modulations Byapplying TURBO 119877 = 13 forward error correction allerror free communications using QPSK and 16QAMwith notmoving case are also confirmed

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgment

This study has been carried out as part of the StrategicInformation andCommunications RampDPromotion Program(SCOPE) Project of the Ministry of Internal Affairs andCommunications Japan

References

[1] M Stojanovic ldquoRecent advances in high-speed underwateracoustic communicationsrdquo IEEE Journal of Oceanic Engineer-ing vol 21 no 2 pp 125ndash136 1996

[2] G Qiao W Wang R Guo R Khan and Y Wang ldquoFrequencydiversity for OFDM mobile communication via underwateracoustic channelsrdquo Journal of Marine Science and Applicationvol 11 no 1 pp 126ndash133 2012

[3] G Leus and P A van Walree ldquoMultiband OFDM for covertacoustic communicationsrdquo IEEE Journal on Selected Areas inCommunications vol 26 no 9 pp 1662ndash1673 2008

[4] H D Trung and N Van Duc ldquoAn analysis of MIMO-OFDMfor underwater communicationsrdquo in Proceedings of the 3rdInternational Congress on Ultra Modern Telecommunicationsand Control Systems and Workshops (ICUMT rsquo11) BudapestHungary October 2011

[5] T Suzuki H M Tran and T Wada ldquoAn underwater acousticOFDM communication system with shrimp (impulsive) noisecancellingrdquo in Proceedings of the IEEE International Conferenceon Computing Management and Telecommunications (Com-ManTel rsquo14) pp 152ndash156 Da Nang Vietnam April 2014

[6] T M Hai Y Matsuda T Suzuki and W A D A Tomo-hisa ldquoUltrasonic diversity OFDM transceiver architecture withimpulsive noise cancelling for shallow sea communicationrdquo inProceedings of the UA2014-2nd International Conference andExhibition on Underwater Acoustics Rodes Greece June 2014

[7] J Gao T IidaHMizutani et al ldquoThe implementation of a 248antennas configurable diversity OFDM receiver for mobile

HDTV applicationrdquo International Journal of Digital MultimediaBroadcasting vol 2009 Article ID 303627 9 pages 2009

[8] T Koujita T Wada T Iida et al ldquoEvaluation of diversitycombining systems for mobile reception in digital terrestrialtelevision broadcastingrdquo in Proceedings of the IEEE Inter-national Conference on Consumer Electronics SESSION 63Digital Receivers I Las Vegas Nev USA January 2009

[9] T Wada ldquoDiversity and error compensation in OFDMtransceivers principles and implementationrdquo in Digital Front-End in Wireless Communication and Broadcasting chapter 18Cambridge University Press 2011

[10] S V Zhidkov ldquoImpulsive noise suppression in OFDM basedcommunication systemsrdquo IEEE Transactions on Consumer Elec-tronics vol 49 no 4 pp 944ndash948 2003

[11] K Al-Mawali A Z Sadik and Z M Hussain ldquoJointtime-domainfrequency-domain impulsive noise reduction inOFDM-based power line communicationsrdquo in Proceedings ofthe Australasian Telecommunication Networks and ApplicationsConference (ATNAC rsquo08) pp 138ndash142 Adelaide AustralasianDecember 2008

Submit your manuscripts athttpwwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Mechanical Engineering

Advances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Distributed Sensor Networks

International Journal of

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Antennas andPropagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

The Scientific World Journal 7

Preamplificationoutput

0 Yes

No

Downconverter

gtthreshold|and2|

(a) Time Domain Impulsive Noise Canceller

Harddecisionand pilotinsertion

+ IFFTPeak

detect FFT

1X

+

minus

Hminus1

1stMRCwithnoise

est

xx

Estimatedfreq

domainnoise

++

minus

EQAi EQBi

Hi

(b) Frequency Domain Impulsive Noise Canceller

Figure 7 Time and Frequency Domain Impulsive Noise Canceller

Table 2 Ocean experiment parameters

Parameters PlaceOkinawa Shizuoka

Type ofexperiment site Fishing port Barge

Number oftransducers 1 TX and 4RX 1 TX and 4RX

RX transducerarray Linear with 20 cm pitch Linear with 20 cm pitch

Modulation QPSK16QAM QPSK16QAMTX-RX distance 715m 13mTransmissiondirection Horizontal Vertical

Ocean depth 5m 30mTransducerdepth (TXRX) 2m2m 22m9m

Figure 10 summarizes BER comparisons for FrequencyDomain Diversity Combined Impulsive Noise Canceller forfreq-IMP ON and OFF In spite of place of experiment siteapproximately 20ndash30 of BER reduction has been obtainedby the proposed Frequency Domain Diversity CombinedImpulsive Noise Canceller

Finally Figures 11(a) and 11(b) showmeasured communi-cation system performance in Okinawa site with enabling theproposed Impulsive Noise Canceller In addition BER data

is also measured with moving the TX transducer speed suchas 00ms (stable) 03 and 06ms for QPSK modulationand 00ms (stable) and 03ms for 16QAM Since MODE2OFDM symbol length of 10667ms as shown in Table 1 is halfof MODE3 the Guard Interval length of MODE2 is shorterthan MODE3 Therefore it is considered that high levelmodulation of 16QAM of MODE2 is largely affected by InterSymbol Interference (ISI) As you can see from those figuresMODE2 shows higher robustness for moving cases such asstrangeness for Doppler Effect because of shorter OFDMsymbol length of MODE2 Once forward error correctionsuch as TURBO CODE with code rate 119877 = 13 is appliedto this system all error free communications (BER = 0) forQPSK modulation of all speeds and 16QAM stable case havebeen confirmed

5 Conclusion

In this paper an ultrasonic OFDM transceiver architecturewith four diverse receivers supporting 2 MODEs is proposedwith enhancement of Impulsive Noise insusceptibility Itutilizes 20ndash28KHz ultrasonic channel and subcarrier spacingof 46875Hz (MODE3) and 93750Hz (MODE2) OFDMmodulations

Two different environment ocean experiments were con-ducted to evaluate the statistics of Impulsive Noise so-called Shrimp Noise which is generated by living creaturesin marine environment To deal with challenges posed by

8 The Scientific World Journal

minus1

0

1

1

minus1 0 1

minus1

0

minus1 0 1

1

minus1

0

minus1 0 1

Branch 1

Branch 3

minus1

0

1

minus1 0 1

minus1

0

1

minus1 0 1

Branch 2

Branch 4

2nd MRC (MODE3)

(a)

minus1

0

1

minus1 0 1

Branch 1

minus1

0

1

minus1 0 1

minus1

0

1

minus1 0 1

Branch 3

minus1

0

1

minus1 0 1

minus1

0

1

minus1 0 1

Branch 2

Branch 4

2nd MRC (MODE2)

(b)

Figure 8 (a)MODE3QPSK constellations (branches 1 to 4 andMRCOkinawa) (b)MODE2QPSK constellations (branches 1 to 4 andMRCOkinawa)

The Scientific World Journal 9

5 10 15 20 25 30 35 40

Symbol number

BER

OFF F-IMP CAN

ON F-IMP CAN

100

10minus1

10minus2

10minus3

(a)

10 20 30 40 50 60 70 80

Symbol number

BER

OFF F-IMP CAN

ON F-IMP CAN

100

10minus1

10minus2

10minus3

(b)

Figure 9 (a) MODE3 BER comparison for 16QAM at Okinawa (b) MODE2 BER comparison for 16QAM at Shizuoka

0

002

004

006

008

01

012

014

QPSKOkinawa

BER

16QAMOkinawa

QPSKOkinawa

16QAMOkinawa

16QAMShizuoka

16QAMShizuoka

OFF F-IMP CANON F-IMP CAN

MODE3 MODE2 MODE2 MODE2MODE3 MODE3

Figure 10 Summary of BER comparison for Frequency Domain Diversity Combined Impulsive Noise Canceller

00001

0001

001

01

10

4 diversity RX transducers liner array with 20 cm pitchQPSK modulationDistance = 715mOkinawa sea experiment

00 03 06

MODE3

MODE2

TX moving speed (ms)

Bit e

rror

rate

(a)

00001

0001

001

01

10

00 03

MODE3

MODE2

16QAM modulationDistance = 715mOkinawa sea experiment

TX moving speed (ms)

Bit e

rror

rate

4 diversity RX transducers liner array with 20 cm pitch

(b)

Figure 11 (a) Communication system performance using QPSK modulation (b) Communication system performance using 16QAMmodulation

10 The Scientific World Journal

the Impulsive Noise interference we have employed a Fre-quency Domain Diversity Combined Impulsive Noise Can-celler

Measured data at a fishing port in Okinawa and ata barge in Shizuoka Prefecture are disclosed and in bothenvironment randomly generated Time Domain ImpulseNoise distribution is observed Although more than 90 to99 of background noise shows AWGN type distributionless than 1 to 10 noise components show large amplitudesuch as Impulsive Noise

By enabling the proposed MRC combined Frequencydomain Impulsive Noise Canceller 20ndash30 BER reductionhas been successfully obtained In addition 715m shallowfishing port horizontal underwater communication BERis also shown using QPSK and 16QAM modulations Byapplying TURBO 119877 = 13 forward error correction allerror free communications using QPSK and 16QAMwith notmoving case are also confirmed

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgment

This study has been carried out as part of the StrategicInformation andCommunications RampDPromotion Program(SCOPE) Project of the Ministry of Internal Affairs andCommunications Japan

References

[1] M Stojanovic ldquoRecent advances in high-speed underwateracoustic communicationsrdquo IEEE Journal of Oceanic Engineer-ing vol 21 no 2 pp 125ndash136 1996

[2] G Qiao W Wang R Guo R Khan and Y Wang ldquoFrequencydiversity for OFDM mobile communication via underwateracoustic channelsrdquo Journal of Marine Science and Applicationvol 11 no 1 pp 126ndash133 2012

[3] G Leus and P A van Walree ldquoMultiband OFDM for covertacoustic communicationsrdquo IEEE Journal on Selected Areas inCommunications vol 26 no 9 pp 1662ndash1673 2008

[4] H D Trung and N Van Duc ldquoAn analysis of MIMO-OFDMfor underwater communicationsrdquo in Proceedings of the 3rdInternational Congress on Ultra Modern Telecommunicationsand Control Systems and Workshops (ICUMT rsquo11) BudapestHungary October 2011

[5] T Suzuki H M Tran and T Wada ldquoAn underwater acousticOFDM communication system with shrimp (impulsive) noisecancellingrdquo in Proceedings of the IEEE International Conferenceon Computing Management and Telecommunications (Com-ManTel rsquo14) pp 152ndash156 Da Nang Vietnam April 2014

[6] T M Hai Y Matsuda T Suzuki and W A D A Tomo-hisa ldquoUltrasonic diversity OFDM transceiver architecture withimpulsive noise cancelling for shallow sea communicationrdquo inProceedings of the UA2014-2nd International Conference andExhibition on Underwater Acoustics Rodes Greece June 2014

[7] J Gao T IidaHMizutani et al ldquoThe implementation of a 248antennas configurable diversity OFDM receiver for mobile

HDTV applicationrdquo International Journal of Digital MultimediaBroadcasting vol 2009 Article ID 303627 9 pages 2009

[8] T Koujita T Wada T Iida et al ldquoEvaluation of diversitycombining systems for mobile reception in digital terrestrialtelevision broadcastingrdquo in Proceedings of the IEEE Inter-national Conference on Consumer Electronics SESSION 63Digital Receivers I Las Vegas Nev USA January 2009

[9] T Wada ldquoDiversity and error compensation in OFDMtransceivers principles and implementationrdquo in Digital Front-End in Wireless Communication and Broadcasting chapter 18Cambridge University Press 2011

[10] S V Zhidkov ldquoImpulsive noise suppression in OFDM basedcommunication systemsrdquo IEEE Transactions on Consumer Elec-tronics vol 49 no 4 pp 944ndash948 2003

[11] K Al-Mawali A Z Sadik and Z M Hussain ldquoJointtime-domainfrequency-domain impulsive noise reduction inOFDM-based power line communicationsrdquo in Proceedings ofthe Australasian Telecommunication Networks and ApplicationsConference (ATNAC rsquo08) pp 138ndash142 Adelaide AustralasianDecember 2008

Submit your manuscripts athttpwwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Mechanical Engineering

Advances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Distributed Sensor Networks

International Journal of

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Antennas andPropagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

8 The Scientific World Journal

minus1

0

1

1

minus1 0 1

minus1

0

minus1 0 1

1

minus1

0

minus1 0 1

Branch 1

Branch 3

minus1

0

1

minus1 0 1

minus1

0

1

minus1 0 1

Branch 2

Branch 4

2nd MRC (MODE3)

(a)

minus1

0

1

minus1 0 1

Branch 1

minus1

0

1

minus1 0 1

minus1

0

1

minus1 0 1

Branch 3

minus1

0

1

minus1 0 1

minus1

0

1

minus1 0 1

Branch 2

Branch 4

2nd MRC (MODE2)

(b)

Figure 8 (a)MODE3QPSK constellations (branches 1 to 4 andMRCOkinawa) (b)MODE2QPSK constellations (branches 1 to 4 andMRCOkinawa)

The Scientific World Journal 9

5 10 15 20 25 30 35 40

Symbol number

BER

OFF F-IMP CAN

ON F-IMP CAN

100

10minus1

10minus2

10minus3

(a)

10 20 30 40 50 60 70 80

Symbol number

BER

OFF F-IMP CAN

ON F-IMP CAN

100

10minus1

10minus2

10minus3

(b)

Figure 9 (a) MODE3 BER comparison for 16QAM at Okinawa (b) MODE2 BER comparison for 16QAM at Shizuoka

0

002

004

006

008

01

012

014

QPSKOkinawa

BER

16QAMOkinawa

QPSKOkinawa

16QAMOkinawa

16QAMShizuoka

16QAMShizuoka

OFF F-IMP CANON F-IMP CAN

MODE3 MODE2 MODE2 MODE2MODE3 MODE3

Figure 10 Summary of BER comparison for Frequency Domain Diversity Combined Impulsive Noise Canceller

00001

0001

001

01

10

4 diversity RX transducers liner array with 20 cm pitchQPSK modulationDistance = 715mOkinawa sea experiment

00 03 06

MODE3

MODE2

TX moving speed (ms)

Bit e

rror

rate

(a)

00001

0001

001

01

10

00 03

MODE3

MODE2

16QAM modulationDistance = 715mOkinawa sea experiment

TX moving speed (ms)

Bit e

rror

rate

4 diversity RX transducers liner array with 20 cm pitch

(b)

Figure 11 (a) Communication system performance using QPSK modulation (b) Communication system performance using 16QAMmodulation

10 The Scientific World Journal

the Impulsive Noise interference we have employed a Fre-quency Domain Diversity Combined Impulsive Noise Can-celler

Measured data at a fishing port in Okinawa and ata barge in Shizuoka Prefecture are disclosed and in bothenvironment randomly generated Time Domain ImpulseNoise distribution is observed Although more than 90 to99 of background noise shows AWGN type distributionless than 1 to 10 noise components show large amplitudesuch as Impulsive Noise

By enabling the proposed MRC combined Frequencydomain Impulsive Noise Canceller 20ndash30 BER reductionhas been successfully obtained In addition 715m shallowfishing port horizontal underwater communication BERis also shown using QPSK and 16QAM modulations Byapplying TURBO 119877 = 13 forward error correction allerror free communications using QPSK and 16QAMwith notmoving case are also confirmed

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgment

This study has been carried out as part of the StrategicInformation andCommunications RampDPromotion Program(SCOPE) Project of the Ministry of Internal Affairs andCommunications Japan

References

[1] M Stojanovic ldquoRecent advances in high-speed underwateracoustic communicationsrdquo IEEE Journal of Oceanic Engineer-ing vol 21 no 2 pp 125ndash136 1996

[2] G Qiao W Wang R Guo R Khan and Y Wang ldquoFrequencydiversity for OFDM mobile communication via underwateracoustic channelsrdquo Journal of Marine Science and Applicationvol 11 no 1 pp 126ndash133 2012

[3] G Leus and P A van Walree ldquoMultiband OFDM for covertacoustic communicationsrdquo IEEE Journal on Selected Areas inCommunications vol 26 no 9 pp 1662ndash1673 2008

[4] H D Trung and N Van Duc ldquoAn analysis of MIMO-OFDMfor underwater communicationsrdquo in Proceedings of the 3rdInternational Congress on Ultra Modern Telecommunicationsand Control Systems and Workshops (ICUMT rsquo11) BudapestHungary October 2011

[5] T Suzuki H M Tran and T Wada ldquoAn underwater acousticOFDM communication system with shrimp (impulsive) noisecancellingrdquo in Proceedings of the IEEE International Conferenceon Computing Management and Telecommunications (Com-ManTel rsquo14) pp 152ndash156 Da Nang Vietnam April 2014

[6] T M Hai Y Matsuda T Suzuki and W A D A Tomo-hisa ldquoUltrasonic diversity OFDM transceiver architecture withimpulsive noise cancelling for shallow sea communicationrdquo inProceedings of the UA2014-2nd International Conference andExhibition on Underwater Acoustics Rodes Greece June 2014

[7] J Gao T IidaHMizutani et al ldquoThe implementation of a 248antennas configurable diversity OFDM receiver for mobile

HDTV applicationrdquo International Journal of Digital MultimediaBroadcasting vol 2009 Article ID 303627 9 pages 2009

[8] T Koujita T Wada T Iida et al ldquoEvaluation of diversitycombining systems for mobile reception in digital terrestrialtelevision broadcastingrdquo in Proceedings of the IEEE Inter-national Conference on Consumer Electronics SESSION 63Digital Receivers I Las Vegas Nev USA January 2009

[9] T Wada ldquoDiversity and error compensation in OFDMtransceivers principles and implementationrdquo in Digital Front-End in Wireless Communication and Broadcasting chapter 18Cambridge University Press 2011

[10] S V Zhidkov ldquoImpulsive noise suppression in OFDM basedcommunication systemsrdquo IEEE Transactions on Consumer Elec-tronics vol 49 no 4 pp 944ndash948 2003

[11] K Al-Mawali A Z Sadik and Z M Hussain ldquoJointtime-domainfrequency-domain impulsive noise reduction inOFDM-based power line communicationsrdquo in Proceedings ofthe Australasian Telecommunication Networks and ApplicationsConference (ATNAC rsquo08) pp 138ndash142 Adelaide AustralasianDecember 2008

Submit your manuscripts athttpwwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Mechanical Engineering

Advances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Distributed Sensor Networks

International Journal of

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Antennas andPropagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

The Scientific World Journal 9

5 10 15 20 25 30 35 40

Symbol number

BER

OFF F-IMP CAN

ON F-IMP CAN

100

10minus1

10minus2

10minus3

(a)

10 20 30 40 50 60 70 80

Symbol number

BER

OFF F-IMP CAN

ON F-IMP CAN

100

10minus1

10minus2

10minus3

(b)

Figure 9 (a) MODE3 BER comparison for 16QAM at Okinawa (b) MODE2 BER comparison for 16QAM at Shizuoka

0

002

004

006

008

01

012

014

QPSKOkinawa

BER

16QAMOkinawa

QPSKOkinawa

16QAMOkinawa

16QAMShizuoka

16QAMShizuoka

OFF F-IMP CANON F-IMP CAN

MODE3 MODE2 MODE2 MODE2MODE3 MODE3

Figure 10 Summary of BER comparison for Frequency Domain Diversity Combined Impulsive Noise Canceller

00001

0001

001

01

10

4 diversity RX transducers liner array with 20 cm pitchQPSK modulationDistance = 715mOkinawa sea experiment

00 03 06

MODE3

MODE2

TX moving speed (ms)

Bit e

rror

rate

(a)

00001

0001

001

01

10

00 03

MODE3

MODE2

16QAM modulationDistance = 715mOkinawa sea experiment

TX moving speed (ms)

Bit e

rror

rate

4 diversity RX transducers liner array with 20 cm pitch

(b)

Figure 11 (a) Communication system performance using QPSK modulation (b) Communication system performance using 16QAMmodulation

10 The Scientific World Journal

the Impulsive Noise interference we have employed a Fre-quency Domain Diversity Combined Impulsive Noise Can-celler

Measured data at a fishing port in Okinawa and ata barge in Shizuoka Prefecture are disclosed and in bothenvironment randomly generated Time Domain ImpulseNoise distribution is observed Although more than 90 to99 of background noise shows AWGN type distributionless than 1 to 10 noise components show large amplitudesuch as Impulsive Noise

By enabling the proposed MRC combined Frequencydomain Impulsive Noise Canceller 20ndash30 BER reductionhas been successfully obtained In addition 715m shallowfishing port horizontal underwater communication BERis also shown using QPSK and 16QAM modulations Byapplying TURBO 119877 = 13 forward error correction allerror free communications using QPSK and 16QAMwith notmoving case are also confirmed

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgment

This study has been carried out as part of the StrategicInformation andCommunications RampDPromotion Program(SCOPE) Project of the Ministry of Internal Affairs andCommunications Japan

References

[1] M Stojanovic ldquoRecent advances in high-speed underwateracoustic communicationsrdquo IEEE Journal of Oceanic Engineer-ing vol 21 no 2 pp 125ndash136 1996

[2] G Qiao W Wang R Guo R Khan and Y Wang ldquoFrequencydiversity for OFDM mobile communication via underwateracoustic channelsrdquo Journal of Marine Science and Applicationvol 11 no 1 pp 126ndash133 2012

[3] G Leus and P A van Walree ldquoMultiband OFDM for covertacoustic communicationsrdquo IEEE Journal on Selected Areas inCommunications vol 26 no 9 pp 1662ndash1673 2008

[4] H D Trung and N Van Duc ldquoAn analysis of MIMO-OFDMfor underwater communicationsrdquo in Proceedings of the 3rdInternational Congress on Ultra Modern Telecommunicationsand Control Systems and Workshops (ICUMT rsquo11) BudapestHungary October 2011

[5] T Suzuki H M Tran and T Wada ldquoAn underwater acousticOFDM communication system with shrimp (impulsive) noisecancellingrdquo in Proceedings of the IEEE International Conferenceon Computing Management and Telecommunications (Com-ManTel rsquo14) pp 152ndash156 Da Nang Vietnam April 2014

[6] T M Hai Y Matsuda T Suzuki and W A D A Tomo-hisa ldquoUltrasonic diversity OFDM transceiver architecture withimpulsive noise cancelling for shallow sea communicationrdquo inProceedings of the UA2014-2nd International Conference andExhibition on Underwater Acoustics Rodes Greece June 2014

[7] J Gao T IidaHMizutani et al ldquoThe implementation of a 248antennas configurable diversity OFDM receiver for mobile

HDTV applicationrdquo International Journal of Digital MultimediaBroadcasting vol 2009 Article ID 303627 9 pages 2009

[8] T Koujita T Wada T Iida et al ldquoEvaluation of diversitycombining systems for mobile reception in digital terrestrialtelevision broadcastingrdquo in Proceedings of the IEEE Inter-national Conference on Consumer Electronics SESSION 63Digital Receivers I Las Vegas Nev USA January 2009

[9] T Wada ldquoDiversity and error compensation in OFDMtransceivers principles and implementationrdquo in Digital Front-End in Wireless Communication and Broadcasting chapter 18Cambridge University Press 2011

[10] S V Zhidkov ldquoImpulsive noise suppression in OFDM basedcommunication systemsrdquo IEEE Transactions on Consumer Elec-tronics vol 49 no 4 pp 944ndash948 2003

[11] K Al-Mawali A Z Sadik and Z M Hussain ldquoJointtime-domainfrequency-domain impulsive noise reduction inOFDM-based power line communicationsrdquo in Proceedings ofthe Australasian Telecommunication Networks and ApplicationsConference (ATNAC rsquo08) pp 138ndash142 Adelaide AustralasianDecember 2008

Submit your manuscripts athttpwwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Mechanical Engineering

Advances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Distributed Sensor Networks

International Journal of

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Antennas andPropagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

10 The Scientific World Journal

the Impulsive Noise interference we have employed a Fre-quency Domain Diversity Combined Impulsive Noise Can-celler

Measured data at a fishing port in Okinawa and ata barge in Shizuoka Prefecture are disclosed and in bothenvironment randomly generated Time Domain ImpulseNoise distribution is observed Although more than 90 to99 of background noise shows AWGN type distributionless than 1 to 10 noise components show large amplitudesuch as Impulsive Noise

By enabling the proposed MRC combined Frequencydomain Impulsive Noise Canceller 20ndash30 BER reductionhas been successfully obtained In addition 715m shallowfishing port horizontal underwater communication BERis also shown using QPSK and 16QAM modulations Byapplying TURBO 119877 = 13 forward error correction allerror free communications using QPSK and 16QAMwith notmoving case are also confirmed

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

Acknowledgment

This study has been carried out as part of the StrategicInformation andCommunications RampDPromotion Program(SCOPE) Project of the Ministry of Internal Affairs andCommunications Japan

References

[1] M Stojanovic ldquoRecent advances in high-speed underwateracoustic communicationsrdquo IEEE Journal of Oceanic Engineer-ing vol 21 no 2 pp 125ndash136 1996

[2] G Qiao W Wang R Guo R Khan and Y Wang ldquoFrequencydiversity for OFDM mobile communication via underwateracoustic channelsrdquo Journal of Marine Science and Applicationvol 11 no 1 pp 126ndash133 2012

[3] G Leus and P A van Walree ldquoMultiband OFDM for covertacoustic communicationsrdquo IEEE Journal on Selected Areas inCommunications vol 26 no 9 pp 1662ndash1673 2008

[4] H D Trung and N Van Duc ldquoAn analysis of MIMO-OFDMfor underwater communicationsrdquo in Proceedings of the 3rdInternational Congress on Ultra Modern Telecommunicationsand Control Systems and Workshops (ICUMT rsquo11) BudapestHungary October 2011

[5] T Suzuki H M Tran and T Wada ldquoAn underwater acousticOFDM communication system with shrimp (impulsive) noisecancellingrdquo in Proceedings of the IEEE International Conferenceon Computing Management and Telecommunications (Com-ManTel rsquo14) pp 152ndash156 Da Nang Vietnam April 2014

[6] T M Hai Y Matsuda T Suzuki and W A D A Tomo-hisa ldquoUltrasonic diversity OFDM transceiver architecture withimpulsive noise cancelling for shallow sea communicationrdquo inProceedings of the UA2014-2nd International Conference andExhibition on Underwater Acoustics Rodes Greece June 2014

[7] J Gao T IidaHMizutani et al ldquoThe implementation of a 248antennas configurable diversity OFDM receiver for mobile

HDTV applicationrdquo International Journal of Digital MultimediaBroadcasting vol 2009 Article ID 303627 9 pages 2009

[8] T Koujita T Wada T Iida et al ldquoEvaluation of diversitycombining systems for mobile reception in digital terrestrialtelevision broadcastingrdquo in Proceedings of the IEEE Inter-national Conference on Consumer Electronics SESSION 63Digital Receivers I Las Vegas Nev USA January 2009

[9] T Wada ldquoDiversity and error compensation in OFDMtransceivers principles and implementationrdquo in Digital Front-End in Wireless Communication and Broadcasting chapter 18Cambridge University Press 2011

[10] S V Zhidkov ldquoImpulsive noise suppression in OFDM basedcommunication systemsrdquo IEEE Transactions on Consumer Elec-tronics vol 49 no 4 pp 944ndash948 2003

[11] K Al-Mawali A Z Sadik and Z M Hussain ldquoJointtime-domainfrequency-domain impulsive noise reduction inOFDM-based power line communicationsrdquo in Proceedings ofthe Australasian Telecommunication Networks and ApplicationsConference (ATNAC rsquo08) pp 138ndash142 Adelaide AustralasianDecember 2008

Submit your manuscripts athttpwwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Mechanical Engineering

Advances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Distributed Sensor Networks

International Journal of

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Antennas andPropagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

International Journal of

RotatingMachinery

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Shock and Vibration

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Mechanical Engineering

Advances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Civil EngineeringAdvances in

Acoustics and VibrationAdvances in

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Electrical and Computer Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Distributed Sensor Networks

International Journal of

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Active and Passive Electronic Components

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Chemical EngineeringInternational Journal of

Control Scienceand Engineering

Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Antennas andPropagation

International Journal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Navigation and Observation

International Journal of

Advances inOptoElectronics

Hindawi Publishing Corporation httpwwwhindawicom

Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014