Conjugate Gradient Adaptive Learning with Tukey's Biweight M-Estimate

TL;DR

This paper introduces the TbMCG algorithm, combining conjugate gradient methods with Tukey's biweight M-estimate, to improve system identification in impulsive noise environments with faster convergence and lower complexity.

Contribution

It presents a novel M-estimate conjugate gradient algorithm that effectively handles impulsive noise, offering improved convergence and computational efficiency over traditional methods.

Findings

TbMCG achieves faster convergence than RLS.

It effectively suppresses impulsive noise in system identification.

Simulation confirms superior performance in noise control applications.

Abstract

We propose a novel M-estimate conjugate gradient (CG) algorithm, termed Tukey's biweight M-estimate CG (TbMCG), for system identification in impulsive noise environments. In particular, the TbMCG algorithm can achieve a faster convergence while retaining a reduced computational complexity as compared to the recursive least-squares (RLS) algorithm. Specifically, the Tukey's biweight M-estimate incorporates a constraint into the CG filter to tackle impulsive noise environments. Moreover, the convergence behavior of the TbMCG algorithm is analyzed. Simulation results confirm the excellent performance of the proposed TbMCG algorithm for system identification and active noise control applications.