快速收敛的指数线性均衡器长度更新算法  

Adaptive Length Updating Algorithm Based on Exponential for Equalizers with High Convergence Rate

作　　者：付少忠[1] 葛建华[1] 王勇[1] 

机构地区：[1]西安电子科技大学综合业务网ISN国家重点实验室,西安710071

出　　处：《系统仿真学报》2009年第16期5018-5021,共4页Journal of System Simulation

基　　金：国家自然科学基金重点项目(60332030);国家自然科学基金重大项目(60496316);国家高技术发展863计划专项课题(2006Aa01Z270)

摘　　要：提出了基于基2及其改进的可变基指数运算的自适应均衡器长度更新算法,算法设置均衡器为较小的初始长度以提高均衡器收敛速度,通过增加或去除长度按指数运算规则自适应变化的子均衡器的方法更新均衡器长度,减少均衡器长度调整次数,提高均衡器的收敛速度和性能。分析表明指数自适应算法能够使均衡器从任意当前长度调整到任意最佳长度。仿真表明基于指数运算的自适应长度均衡器能够获得更低的MSE,有效减小计算复杂度,提高信道跟踪速度,从而可以跟踪时变信道。Adaptive linear equalizers are commonly used in wireless systems to reduce the distorting effects of the channel. One of the key parameters to be determined is the number of taps of the equalizer. Two adaptive algorithms for updating the length of the equalizer based on exponential function with base 2 and adaptive base were proposed. Its main idea is to adjust the length of the equalizer from few initial taps by adding or removing a subequalizer whose length is chosen according to exponential function, which increases the convergence rate and improves the system performace by reducing adjust times. Anaysis indicts that the adaptive length equalizer based on exponential function can reach any optimal length from arbitrary initial length. Simulation results show that the algorithms proposed are low in complexity, high in the rate of tracing the channel and can track enven a very fast changing channel.

关 键 词：自适应长度 分段均衡器 指数运算 MSE 

分 类 号：TN929.5[电子电信—通信与信息系统]