A new ICA algorithm for blind source separation

Author

Ku, Chin-Jen ; Fine, Terrence L.

Author_Institution

Dept. of Electr. & Comput. Eng., Cornell Univ., Ithaca, NY, USA

Volume

5

fYear

2003

fDate

6-10 April 2003

Abstract

We propose a nonparametric independent component analysis (ICA) algorithm for the case of instantaneous and linear mixtures. Our algorithm combines minimization of correlation among nonlinear expansions of the output signals with good initialization derived from a search guided by statistical tests for independence. Simulation results obtained from both synthetic and real-life data show that our method provides consistent results and compares favorably to existing ICA algorithms.

Keywords

blind source separation; independent component analysis; iterative methods; matrix algebra; minimisation; ICA algorithm; blind source separation; initialization; instantaneous mixtures; iterative correlation minimization; linear mixtures; mixing matrix; nonlinear expansion; nonparametric independent component analysis; Algorithm design and analysis; Blind source separation; Covariance matrix; Data mining; Independent component analysis; Noise measurement; Probability; Signal processing; Source separation; Testing;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech, and Signal Processing, 2003. Proceedings. (ICASSP '03). 2003 IEEE International Conference on

ISSN

1520-6149

Print_ISBN

0-7803-7663-3

Type

conf

DOI

10.1109/ICASSP.2003.1199861

Filename

1199861