一种改进的变步长NLMS算法及其仿真
首发时间:2009-12-21
摘要:本文通过对传统NLMS算法的分析,提出了一种新的变步长NLMS算法,该算法基于输入信号与误差的互相关函数去动态地调整步长。计算机仿真结果表明在白色高斯输入及相关输入条件下新算法跟传统NLMS算法相比在取得相同稳态失调的前提下具有更快的收敛速度,并且当未知系统发生突变时,新算法也具有较强的跟踪能力。
For information in English, please click here
An Improved NLMS Algorithm and Its Simulation
Abstract:By means of analysizing the traditional NLMS algorithm, an improved NLMS algorithm with variable step size is proposed in this paper. The step size is dynamically adjusted according to the coss-correlation function between the input signal and output error.Computer simulation results demonstrate that, with the same steady misadjustment, the new algorithm can achieve faster convergence rate than the traditional NLMS algorithm under the condition of white gussian input or correlated input. Also when the unknown system suddenly changes, the new algorithm has better track ability.
Keywords: NLMS Algorithm Variable Step Size Cross-correlation Function
基金:
论文图表:
引用
No.3784250723112613****
同行评议
勘误表
一种改进的变步长NLMS算法及其仿真
评论
全部评论0/1000