一种自适应均衡器最优阶数确定算法

提出了一种基于信道估计与级数展开的均衡器阶数确定算法(equalizer order algorithm based on channel estimation andseries expansion,CESE-EOA)。自适应均衡器是相干水声通信系统的重要组成部分,其阶数参数的确定会对系统性能产生重大影响。传统的均衡器阶数确定方法有适应性较差、对训练序列的长度要求较高等缺点。从多重线性回归模型的角度讨论了自适应均衡器阶数选择的依据,并以此为基础给出了一种最优的阶数计算方法,减小了均衡器的自噪声。理论分析与仿真、试验结果证明了该算法的正确性。该算法可用于计算线性自适应均衡器、盲均衡器或判决反馈均衡器中前馈滤波器的最优阶数,对均衡器设计有重要借鉴意义。

Optimum order algorithm for adaptive equalizers

An equalizer order algorithm based on channel estimation and series expansion(CESE-EOA) is proposed in this paper.Adaptive equalizer is essential in coherent underwater acoustic communication system.Traditional methods of determining the order of an equalizer either lack flexibility or require long training sequence.The criterion of selecting equalization order based on the multiple linear regression model is discussed.CESE-EOA employs channel impulse response and Taylor’s series expansion.Theoretical analyses and simulation results show that CESE-EOA can ensure a desirable equalization order.The algorithm can be applied in linear adaptive equalizer,blind equalizer or the forward filter in a decision feedback equalizer to determine the optimum order,and has important reference value for equalizer design.