基于SNR的变步长LMS算法
首发时间:2011-01-19
摘要:本文将信噪比(SNR)引入到变步长LMS(Least Mean Square)算法中,通过建立步长因子μ和SNR之间的非线性关系,提出了一种基于SNR的变步长LMS算法。对于各种不同的性噪比输入,该算法在初始阶段具有比较快的收敛速度,并且具有稳态误差较小的特点。采用Matlab进行计算机仿真,实验结果表明该算法的性能优于VSLMS 算法。
For information in English, please click here
SNR-based variable step size LMS algorithm
Abstract:This article introduces signal to noise ratio(SNR) into variable step size LMS algorithm. Through the establishment of nonlinear relationship between compensation factor μ and SNR, this article proposal s SNR-based variable step size LMS algorithm. For input signal of various signal to noise ratio, the algorithm has characteristics of faster convergence speed and smaller steady-state error. Using Matlab to do computer simulation, experimental result shows that the algorithm outperforms the S function variable step size LMS algorithm (SVSLMS algorithm).
Keywords: variable step LMS SNR
基金:
论文图表:
引用
No.4401979562020129****
同行评议
共计0人参与
勘误表
基于SNR的变步长LMS算法
评论
全部评论0/1000