A variable step size bias-compensated affine projection algorithm with noisy inputs

Chan Park

doi:10.1016/j.jfranklin.2025.107792

A variable step size bias-compensated affine projection algorithm with noisy inputs

2025-06-21

0

PRE

AI

DOI：10.1016/j.jfranklin.2025.107792

Original

Original request for help

Save

Abstract

En 中文

• Proposes a novel Variable Step Size Bias-Compensated APA. • First direct derivation of bias compensation vector in APA. • Uses MSD analysis to derive the optimal step size for each iteration. • Achieves improved convergence and reduced error under noisy conditions. • Enhances robustness in practical system identification tasks.

A variable step size bias-compensated affine projection algorithm with noisy inputs

Abstract

Journal

Organization