水声通信中基于信道估计的稀疏均衡算法研究  

Simulation of Multichannel Sparse Equalization Algorithm Based on Channel Estimation for Underwater Acoustic Communications

作　　者：冉茂华[1] 黄建国[1] 韩晶[1] 

机构地区：[1]西北工业大学航海学院,西安710072

出　　处：《系统仿真学报》2009年第14期4479-4482,共4页Journal of System Simulation

基　　金：国家自然科学基金(60572098)

摘　　要：为实现相位相干水声通信系统的实时处理,降低由信道大时延多径结构所导致的接收算法复杂度与稳定性问题,提出了一种基于信道估计的多通道稀疏均衡算法。该算法针对水声信道的稀疏特性,采用稀疏算法减小均衡器的阶数,降低了系统复杂度;并且将变步长LMS算法与基于信道估计的完全反馈均衡器相结合,提高了均衡器性能;同时采用多通道可变步长算法,各个通道的步长可以独立的自适应调节,增加了均衡器的稳定性与灵活性。在实际湖试信道条件下的性能仿真表明:在10km的距离上,与传统四通道DFE-LMS算法相比,此算法在降低稳态误差的同时,参与迭代计算的抽头数减少了78%,从而计算量大为减小,同时系统性能也得到了改善。In order to reduce complexity of the receiver and improve performance of data detection, an improved multichannel sparse equalizer algorithm was proposed according to the sparse characteristic of high rate underwater acoustic communications. The algorithm combined an adaptive step size DFE-LMS algorithm and channel estimation with a variation of the decision feedback equalizer, namely the complete feedback equalizer （CFE）. It could not only achieve better equalization of each channel but also reduce the order of equalizers and calculate effectively. The state mean squared error of equalizers was decreased and its flexibility was increased when using the adaptive step size DFE-LMS algorithm. Simulation tests were conducted using 10Km real underwater acoustic channel. Results show that the proposed algorithm has the good performance of robustness, lower MSE, calculation-efficient , and that is better than conventional DFE-LMS algorithm.

关 键 词：高速水声通信 信道估计 完全反馈均衡器 变步长 稀疏均衡 

分 类 号：TN911[电子电信—通信与信息系统]