一种变步长Proportionate NLMS自适应滤波算法及其在网络回声消除中的应用

A Variable Step-Size Proportionate NLMS Adaptive Filtering Algorithm and Its Application in Network Echo Cancellation

Proportionate自适应算法利用稀疏冲激响应的结构特征,极大地加速了算法的收敛速度.但是快速收敛与低稳态失调是一对矛盾的需求,固定步长算法必需折中选择一个步长参数来满足应用的要求.本文提出了一种适用于Proportionate算法的变步长方法,有效解决了收敛速度和稳态失调之间的矛盾.所提的算法首先利用最小干扰原理,得到了一个Proportionate NLMS算法的推导;进而将干扰信号考虑进算法的系数更新过程,通过在每一步迭代中用后验误差去补偿干扰信号的负面作用,得到一个新的优化准则;最后利用这个准侧,推导出了一个适用于Proportionate算法的步长调节方法.仿真实验验证了本文方法的有效性.

Proportionate adaptive algorithms exploit structure characteristic of sparse impulse response to considerably improve the convergence speed.However,the requirements of fast convergence and low steady-state misalignment are conflict for constant step-size adaptive algorithms,whose step size parameter has to be selected by compromising these two conflict requirements.In this article,a novel variable step-size method is proposed for proportionate adaptive algorithm to solve this problem.By using principle of minimal disturbance to proportionate adaptive algorithm,a derivation of proportionate NLMS is provided first.Then by taking into account the disturbance signal,forcing the a posterior error to cancel negative effect of disturbance signal,a new optimization criterion is obtaind.At last,using this criterion,a step size control approach for proportionate NLMS algorithm is proposed.Simulation results verify the effectiveness of the proposed algorithm.