Novel blind source separation algorithms using cumulants

Sergio Cruces, Luis Castedo, Andrzej Cichocki

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

23 Citations (Scopus)

Abstract

This paper investigates new algorithms for blind source separation that use cumulants instead of nonlinearities matched to the probability distribution of the sources. It is demonstrated that separation is a saddle point of a cumulant-based entropy cost function. To determine this point we propose two quasi-Newton algorithms whose convergence is isotropic and does not depend on the sources distribution. Moreover, convergence properties remain the same when there is Gaussian noise in the mixture.

Original languageEnglish
Title of host publicationCommunicationsSensor Array and Multichannel Signal Processing
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages3152-3155
Number of pages4
ISBN (Electronic)0780362934
DOIs
Publication statusPublished - 2000
Externally publishedYes
Event25th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2000 - Istanbul, Turkey
Duration: 5 Jun 20009 Jun 2000

Publication series

NameICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Volume5
ISSN (Print)1520-6149

Conference

Conference25th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2000
Country/TerritoryTurkey
CityIstanbul
Period5/06/009/06/00

Fingerprint

Dive into the research topics of 'Novel blind source separation algorithms using cumulants'. Together they form a unique fingerprint.

Cite this