Adaptive RLS algorithm for blind source separation using a natural gradient

Author

Zhu, Xiao-Long ; Zhang, Xian-Da

Author_Institution

Key Lab. for Radar Signal Process., Xidian Univ., Xi´´an, China

Volume

9

Issue

12

fYear

2002

Firstpage

432

Lastpage

435

Abstract

By using the natural gradient on the Stiefel manifold to minimize a nonlinear principle component analysis criterion, this letter proposes a new adaptive recursive-least-squares (RLS) algorithm with prewhitening for blind source separation (BSS), which makes full use of the orthogonality constraint of the separating matrix. Simulations show that the new natural-gradient-based RLS algorithm has faster convergence than the existing least-mean-square algorithms and RLS algorithm for BSS.

Keywords

adaptive signal processing; blind source separation; convergence of numerical methods; least squares approximations; minimisation; principal component analysis; BSS; Stiefel manifold; adaptive RLS algorithm; adaptive recursive-least-squares algorithm; blind source separation; convergence; natural gradient; nonlinear principle component analysis criterion; orthogonality constraint; prewhitening; separating matrix; Algorithm design and analysis; Blind source separation; Convergence; Independent component analysis; Laboratories; Least squares methods; Principal component analysis; Resonance light scattering; Signal processing algorithms; Source separation;

fLanguage

English

Journal_Title

Signal Processing Letters, IEEE

Publisher

ieee

ISSN

1070-9908

Type

jour

DOI

10.1109/LSP.2002.806047

Filename

1159631