一种新的变步长LMS自适应滤波算法

A New Variable Step Size LMS Adaptive Filtering Algorithm

本文通过建立步长因子μ与误差信号之间的非线性函数关系,提出了一种新的变步长LMS(LeastMean Square)算法.该算法具有初始阶段和未知系统时变阶段步长自动增大而稳态时步长很小的特点,且克服了S函数变步长LMS算法(简称SVSLMS算法)在自适应稳态阶段μ(n)取值偏大的缺陷.理论分析和计算机仿真结果表明该算法的性能优于SVSLMS算法.

By building a nonlinear function relationship between μ and the error signal,this paper presents a novel variable step size LMS（ Least Mean Square） adaptive filtering algorithm. The step size of this algorithm increases automaticly at the beginning of this algorithm or when unknown system is changing with time, and it would be smaller during the steady state. This algorithm avoid the shortage of changing step size of SVSLMS, variable step size LMS based on Sigmoid function, in the process of the adaptive steady state. The performance of this paper algorithm is better than that of SVSLMS with the theoretical analysis and computer simulations.