Delayed Combination of Adaptive Filters in Colored Noise

Research output: Journal Publications and Reviews (RGC: 21, 22, 62)21_Publication in refereed journalpeer-review

View graph of relations

Author(s)

Related Research Unit(s)

Detail(s)

Original languageEnglish
Pages (from-to)1918-1931
Journal / PublicationIEEE Transactions on Signal Processing
Volume70
Online published6 Apr 2022
Publication statusPublished - 2022

Abstract

In this work, we study the combination of adaptive filters in colored noise environments. First, a combination framework using delayed weights is introduced to tackle the colored noise. Based on this, delayed convex and affine combinations of two LMS filters are developed, resulting in the so-called Dcvx-LMS and Daff-LMS algorithms. Then, the convergence behaviors of the two algorithms are investigated using standard mean-square deviation analysis. In addition, to speed up the convergence and reduce the computational complexity, we propose delayed combination with periodic feedback, delayed combined-step-size and block implementation methods. Finally, simulation results demonstrate the superiority of our algorithms over previously reported techniques in the presence of colored noise.

Research Area(s)

  • Adaptive filter, Adaptive filters, colored noise, Colored noise, combination with delayed weights, Convergence, Filtering algorithms, mean-square analysis, Noise measurement, Signal processing algorithms, Steady-state