Research On Issues Related To OFDM Systems For Underwater Acoustic Communication

To the underwater acoustic(UWA)channel,there exits a serious intersymbol interference(ISI)for the long multipath time delay;there exits an obvious Doppler effect for the motion of the underwater equipments and the features of the channel,such as narrow bandwidth and fast time-varying;there exits a large noise for the complex medium composition,water movement form and environment.Serious distortion and fast attenuation to the signal are acquired for the above characteristics of the UWA channel.So how to solve the problems of the multipath phenomenon,the Doppler effect,and underwater noise is becoming an important subject to the UWA communication systems.Orthogonal frequency division multiplexing(OFDM)is a kind of multi-carrier parallel transmission technology.The technique has not only a good anti-multipath performance and higher spectrum efficiency but also a flexible modulation and power allocation based on sub-carrier channel conditions.Thus the OFDM technique has been widely applied in wireless electromagnetic communication on land and UWA communication systems.But compared with the wireless radio applications,there exists more problems in the UWA communication field.First,OFDM signal is easy to appear higher peak to average power ratio(PAPR),so it is very unfavorable to use low power wireless equipment under water normally.Secondly,OFDM system is more sensitive to frequency offset,and the Doppler effect is one of the main factors which would cause the deviation.In addition,due to the time-varying characteristic and large noise of UWA channel,not only better ability of resist noise and channel tracking,but also lower computational complexity are required for coherent demodulation.So there are more critical problems to solve for OFDM UWA communication system,namely higher PAPR,Doppler shift and channel estimation.Aiming at the problems about the higher PAPR,Doppler shift and channel estimation,further research is made in the paper for the OFDM UWA communication systems.Related improved algorithms are proposed and verified by simulation experiment in order to boost the effect of high speed and real-time performance to the systems.The paper's main work includes the following four parts:First,the characteristics of UWA channel are discussed from the bandwidth,noise,multipath and Doppler effect respectively.According to the characteristics,modeling and simulation are finished.And then,the advantages and disadvantages of OFDM UWA communication systems are summed up.Finally,problems of PAPR,Doppler shift and channel estimation are put forward related with the systems.Secondly,considering the low power characteristics of underwater wireless communication equipment,an algorithm with low complexity is proposed to reduce the clipping noise based on the traditional clipping methods for the higher PAPR of OFDM UWA communication systems.Different clipping threshold is used to limit the higher amplitude in the algorithm.After that,a vector is used to save the processing results for restoring the original value as far as possible at the receiving end.Theory analysis and simulation results all show that the algorithm can not only restrain the clipping noise but also improve the system real-time performance effectively.Thirdly,to the problem about the serious Doppler shift for OFDM UWA communication systems,a new algorithm is proposed about the Doppler shift factor(DSF),sampling rate conversion and carrier frequency shift(CFO)estimation.In the first place,DSF is estimated preliminary through the linear frequency modulation(LFM)signal to the OFDM packet at the receiving end.Then oversampling is carried out to the received data.Combining with the initial value of DSF and the oversampling data,the sampling rate conversion is finished through the linear interpolation.Finally,making the above DSF initial value as a data center and according to the system accuracy requirement,several candidate DSF values are listed,and the DSF is estimated accurately combined with energy cost function of the null carriers.So the estimation and compensation of the CFO is finished according to the more accurate factor.Through theory analyzing and computer simulating,it is proved that the improved algorithm has a good bit error rate(BER)performance and a low computational complexity compared to the traditional methods,and can effectively solve the problem of frequency offset causing by Doppler effect for OFDM UWA communication systems.Finally,aiming at the error problems of the least squares(LS)algorithm(influenced by noise)and the Discrete Fourier transform(DFT)algorithm(influenced by null carrier),an improved LS-DFT channel estimation algorithm is discussed combing with time-varying and noise feature of the UWA channel.First of all,the pilot subcarriers are estimated through LS method in frequency domain.Then the noise reduction is carried out through DFT algorithm.Thus another set of frequency domain pilot channel value are acquired.Next,the error comparison and the data substitution are implemented to the above two sets of channel data.So a new set of channel data is obtained.In the end,a linear interpolation is made to get data subcarriers channel value in frequency domain.Theory analysis and simulation results all demonstrate that the improved channel estimation algorithm can be a compromise between the LS algorithm and DFT algorithm,and the proposed algorithm can improve the system performance effectively.