浅海水声信道自适应均衡算法

Experimental Studies on Shallow Water Acoustic Channel Equalization

浅海水声信道是一个典型的带宽受限、时变衰落信道,存在着时变的多径扩展、频率偏移以及随机相位起伏.在水声相干通信中,自适应均衡是克服多径效应、消除码间干扰的主要手段.针对在实际应用中自适应均衡算法对步长参数选择的敏感性,进行了浅海信道水声相干通信均衡算法研究,结合录制的水声信道数据,通过Matlab软件,从星座图、算法收敛特性以及系统的误码特性这3个方面对几种均衡算法的性能进行了分析比较.结果表明,最小均方(LMS)平行滤波器组算法具有自适应调整步长的性能,且能获得较快的收敛速度和较小的误码率,验证了该算法在水声信道中的实用性.

The shallow water acoustic channel is characterized as a band-limited time-dispersive fading channel,and exhibits the characters of time-varying multipath propagation,frequency offsets and phase fluctuations.Adaptive equalization is the main method to overcome multipath fading and eliminate inter symbol interference in underwater acoustic coherent communications.Considering the sensitivity in step-size parameter selection which causes performance degradation in practical applications,experiments were carried out to study the performance of some adaptive equalization algorithms.Matlab software was used to analyze the recorded data in practical underwater acoustic channel.And the performances of the algorithms were compared in several factors,such as the constellation,the convergence and the symbol error rate.Data processing results show that parallel filter banks algorithm has faster convergence speed along with low steady state error.And they verify the improvement of the algorithm in terms of step-size parameter tolerance.