Research On The Key Technology Of Single Carrier Underwater Acoustic Communication

Single carrier underwater acoustic communication is a kind of communication modulation method with high spectral efficiency.However,it is still difficult to realize reliable underwater wireless information transmission in the underwater acoustic channel.The main obstacles are as follows:?1?Serious multi-path extensions cause severe Inter-Symbol Interference?ISI?,which leads to severe selective frequency fading.Moreover,the serious multi-path extension of the underwater acoustic channel brings great challenges to computing complexity and convergence rate of channel equalization in the coherent underwater acoustic communication;?2?Serious Doppler effect,which is brought by the low sound speed of the underwater acoustic communication,makes the coherent underwater acoustic communication more difficult.In order to overcome the time-varying characteristic,sparse characteristic,severe multi-path expansion and broadband Doppler spread on the underwater acoustic single carrier communication system,a novel adaptive algorithm with adaptive closed loop Doppler estimation and compensation and channel shortening method is proposed in this paper.Firstly,the technical challenges faced by the single carrier underwater acoustic communication technology are discussed?i.e.,severe multi-path extension brings severe ISI,leading to severe selective frequency fading.In addition,the low sound speed in underwater acoustic communication leads to severe Doppler effect,which makes the demodulation of the single carrier underwater acoustic communication more and more difficult?.The history and development of existing technologies of single carrier underwater acoustic communication are summarized.Faced with the problems in the single carrier underwater acoustic communication system,the key technologies of single carrier underwater acoustic communication are proposed,and the research direction is pointed out.Secondly,facing with the problem of severe multi-path extension which brings severe ISI,leads to severe selective frequency fading.In addition,the time-varying characteristic and the sparse charactertic of the underwater acoustic channel,theoretical research are studied as follows:?1?Facing with the fast time-varying characteristic of underwater acoustic channel,a novel Recursive Least Squares?RLS?algorithm with variable forgetting factor is proposed to solve this problem,since the conventional RLS adaptive with fixed forgetting factor can not adapt the fast time-varying underwater acoustic channel.?2?Facing with the non-circular characteristic of signals in the single carrier underwater acoustic communication system,a novel widely linear RLS adaptive algorithm is proposed.By using the widely linear model of the signal,the conventional RLS adaptive algorithm can achieve a lower Mean Square Error?MSE?.?3?Facing with the sparsity characteristic of the underwater acoustic channel,the conventional sparse RLS algorithm,the group sparse RLS algorithm and the mixed sparse RLS algorithm are proposed.By using the sparse constraints,the noise enhancement property of conventional RLS adaptive algorithm is improved.?4?Facing with the problem of large computing complexity of RLS algorithm,the novel RLS algorithm based on Dichotomous Coordinate Descent?DCD?iteration is proposed.The computing complexity of conventional RLS algorithm can be decreased from O?N²?to O?N??where,N is the order of the adaptive filter?by using DCD iteration.Simulation results show that compared with the conventional RLS algorithm,the proposed RLS algorithm with variable forgetting factor,the proposed widely linear RLS adaptive algorithm and the proposed sparse RLS algorithm can achieve 5 dB,3 dB and 8 dB gain in terms of steady-state MSE,respectively.Moreover,the proposed RLS adaptive algorithm based on DCD iteration can reduce the computing complexity and improve the convergence rate.Experimental results show that the proposed widely linear sparse DCD-RLS algorithm with variable forgetting factor can reduce the steady-state MSE by about 18 dB,and reduce the Symbol Error Rate?SER?by about 28%in the single carrier underwater acoustic communication system.The widely linear sparse DCD-RLS algorithm with variable forgetting factor can enhance the receiver's tracking capability and adaptability to sparse time-varying underwater acoustic channel and achieve reliable and effective single carrier underwater acoustic communication.Thirdly,faced with the problem of broadband Doppler effect of single carrier underwater acoustic communication system,theoretical researches are studied as follows:?1?For the case of slow time-varying characteristic within the data frame,the open-loop Doppler estimation and compensation algorithm is proposed,which can estimate the average Doppler effect within the data frame.?2?For the case of broadband Doppler effect within the data frame,the adaptive closed-loop Doppler estimation and compensation algorithm is proposed.This algorithm can estimate and compensate the time-varying broadband Doppler within the data frame.Simulation results show that the proposed equalizer coped with adaptive closed-loop Doppler estimation and compensation algorithm is better than the conventional equalizer coped with open-loop Doppler estimation and compensation algorithm and a second-order digital phase-locked loop with the gain of the output signal to noise ratio to at least 1 dB,reducing the bit error rate?BER?from 0.056%to 0.011%in the single carrier underwater acoustic communication system.Experimental results show that compared with the conventional equalizer coped with open-loop Doppler estimation and compensation algorithm with a second-order phase-locked loop,the equalizer coped with adaptive closed-loop Doppler estimation and compensation algorithm can achieve a gain of at least 1 dB in terms of output signal to noise ratio with a symbol rate of 1k sym/s in the shallow water.The proposed algorithm can also reduce the BER from 8.73%to 0.11%.Faced with the broadband Doppler effect,the proposed equalizer coped with adaptive closed-loop Doppler estimation and compensation algorithm can significantly improve the performance of the single carrier underwater acoustic communication system.Finally,faced with the severe ISI caused by the severe multi-path extension,resulting in severe selective frequency fading,time-varying characteristic,sparse characteristic and the broadband Doppler effect of the underwater acoustic channel in the single carrier adaptive Turbo equalizer system,the theoretical work includes:?1?Aiming at the suppression of ISI in underwater acoustic channel,the Turbo equalizer based on MMSE criterion of prior information is studied.Using the soft feedback information from decoder,the estimated value of the symbol is transmitted and the soft information about the transmitted symbol is obtained by the mapping as soft input such as a soft-decoder.The soft information on the transmitted symbol is the output of the decoder and feedback to the equalizer,forming a loop.The performance of the equalizer will get better and better by increasing the number of iterations.?2?Faced with the problem that the underwater acoustic channel has a long multi-path extension,the Turbo equalizer coped with channel shortening method?i.e.,passive Time Reversal?pTR?method or Minimum Mean Square Error?MMSE?method?is proposed to reduce the computing complexity of the equalizer by reducing the length of the multi-path extension of the underwater acoustic channel.?3?Faced with the serious ISI leading to severe selective frequency fading,time-varying characteristic and sparse characteristic of underwater acoustic channel,a novel widely linear sparse DCD-RLS algorithm with variable forgetting factor is proposed to improve the performance of the Turbo equalizer.?4?Faced with the broadband Doppler effect,the Turbo equalizer coped with adaptive closed-loop Doppler estimation and compensation algorithm is proposed.Coping with the adaptive closed-loop Doppler estimation and compensation algorithm,the Turbo equalizer can achieve symbol-by-symbol Doppler estimation and compensation.Simulation results show that the proposed algorithm based on the widely linear sparse DCD-RLS with varable forget factor algorithm has an 8dB performance improvement over the conventional RLS algorithm in terms of the steady-state MSE.The proposed RLS algorithm also has a performance improvement of 4.2 dB when SER is 10^-4.Experimental results show that the Turbo equalizer coped with the widely linear sparse DCD-RLS algorithm with variable forgetting factor,adaptive closed-loop Doppler estimation and compensation algorithm and the channel shortening method can significantly improve the performance in terms of steady-state MSE and SER with a lower computing complexity in the single carrier underwater acoustic communication system?about 17 dB improvement in steady-state MSE and 28%improvement in SER?.