Performance analysis of adaptive filters for time-varying systems

Two typical adaptive algorithms, LMS filtering and RLS filtering, were introduced and compared in this paper. The convergence performance and tracking performance in non-stationary were analyzed by simulation. When the identification plant was time-varying, adaptive filters should have the ability of tracking the minimum point. Compared with LMS filters, one important feature RLS filters have is convergence rate, but the improvement of this performance is cost by the increasing calculation complexity of RLS filters. According to simulation analysis, in time-varying environment, LMS filters have better tracking performance than RLS filters.

1 Introduction

Adaptive filtering schemes have frequently employed in communications, signal processing, control and many other applications recently [1], [2]. Among these adaptive filtering algorithms, Least Mean Square (LMS) algorithm and Recursive Least Squares (RLS) algorithm have become the most popular adaptive filtering algorithms as a consequence of their simplicity and robustness.