Research On Underwater Acoustic Single Carrier Iterative Equalization Based On Expectation Propagation

Underwater acoustic single-carrier communication technology is an effective means to realize underwater information transmission.The multipath effect of underwater acoustic channels and the Doppler phenomenon caused by motion greatly restrict the improvement of the performance of underwater acoustic communication systems.In order to overcome the influence of inter-symbol interference(ISI)caused by multipath effect on underwater acoustic communication systems,this paper studies the iterative equalization technology based on expectation propagation(EP)and applies it to underwater acoustic communicationFirstly,in view of the long delay and frequency selective fading problems of underwater acoustic channels,this paper studies an improved decision feedback equalization algorithm and uses it for iterative equalization.The algorithm uses the minimum mean square error(MMSE)criterion to update the equalizer coefficients,which effectively reduces the computational complexity with a small performance loss.At the same time,the calculation method of the log-likelihood ratio(LLR)of the decision feedback equalizer(DFE)output is revised,that is,the decision error is considered in the calculation of the LLR,which can well suppress the error propagation phenomenon of traditional decision feedback equalization algorithm in iteration.Then this paper also uses Bellhop sound field simulation software for simulation experiments.The result shows that the improved algorithm can make the bit error rate reach the order of10^-5 when the underwater acoustic channel length is 34 symbols and the maximum frequency attenuation is 30d B.Secondly,this paper also studies the iterative equalizer based on EP algorithm and applies it to underwater acoustic iterative equalization.Based on the principle of EP,the paper derives a filter-type EP algorithm,which greatly reduces the complexity of the algorithm.The results of marine experimental data processing show that the filter-type EP algorithm has good performance and can obtain better equalization effects than that of the linear minimum mean square error(LMMSE)criterion.At the same time,this paper also combines DFE with EP algorithm and introduces it to underwater acoustic iterative equalization.Aiming at the error propagation phenomenon of classical DFE,this paper studies a bi-directional decision feedback equalization algorithm and an improved decision feedback equalization algorithm based on EP from the aspects of equalizer structure design and error propagation principle.Simulation and marine experimental data processing results show that both algorithms can effectively suppress the error propagation phenomenon and obtain a good equalization effect.