A new fixed-point algorithm for independent component analysis

Zhenwei Shi; Huanwen Tang; Yiyuan Tang

doi:10.1016/j.neucom.2003.09.002

A new fixed-point algorithm for independent component analysis

Zhenwei Shi, Huanwen Tang, Yiyuan Tang

Research output: Contribution to journal › Article › peer-review

33 Scopus citations

Abstract

A new fixed-point algorithm for independent component analysis (ICA) is presented that is able blindly to separate mixed signals with sub- and super-Gaussian source distributions. The new fixed-point algorithm maximizes the likelihood of the ICA model under the constraint of decorrelation and uses the method of Lee et al. (Neural Comput. 11(2) (1999) 417) to switch between sub- and super-Gaussian regimes. The new fixed-point algorithm maximizes the likelihood very fast and reliably. The validity of this algorithm is confirmed by the simulations and experimental results.

Original language	English (US)
Pages (from-to)	467-473
Number of pages	7
Journal	Neurocomputing
Volume	56
Issue number	1-4
DOIs	https://doi.org/10.1016/j.neucom.2003.09.002
State	Published - Jan 2004
Externally published	Yes

Keywords

Blind source separation
Fixed-point algorithm
Independent component analysis

ASJC Scopus subject areas

Computer Science Applications
Cognitive Neuroscience
Artificial Intelligence

Access to Document

10.1016/j.neucom.2003.09.002

Cite this

@article{77873ac5d75a4b3087495236d461d025,

title = "A new fixed-point algorithm for independent component analysis",

abstract = "A new fixed-point algorithm for independent component analysis (ICA) is presented that is able blindly to separate mixed signals with sub- and super-Gaussian source distributions. The new fixed-point algorithm maximizes the likelihood of the ICA model under the constraint of decorrelation and uses the method of Lee et al. (Neural Comput. 11(2) (1999) 417) to switch between sub- and super-Gaussian regimes. The new fixed-point algorithm maximizes the likelihood very fast and reliably. The validity of this algorithm is confirmed by the simulations and experimental results.",

keywords = "Blind source separation, Fixed-point algorithm, Independent component analysis",

author = "Zhenwei Shi and Huanwen Tang and Yiyuan Tang",

note = "Funding Information: The work was supported by NSFC(30170321, 90103033), MOE(KP0302) and MOST(2001CCA00700). The authors would like to thank the reviewers for their useful comments and suggestions.",

year = "2004",

month = jan,

doi = "10.1016/j.neucom.2003.09.002",

language = "English (US)",

volume = "56",

pages = "467--473",

journal = "Neurocomputing",

issn = "0925-2312",

publisher = "Elsevier",

number = "1-4",

}

TY - JOUR

T1 - A new fixed-point algorithm for independent component analysis

AU - Shi, Zhenwei

AU - Tang, Huanwen

AU - Tang, Yiyuan

N1 - Funding Information: The work was supported by NSFC(30170321, 90103033), MOE(KP0302) and MOST(2001CCA00700). The authors would like to thank the reviewers for their useful comments and suggestions.

PY - 2004/1

Y1 - 2004/1

N2 - A new fixed-point algorithm for independent component analysis (ICA) is presented that is able blindly to separate mixed signals with sub- and super-Gaussian source distributions. The new fixed-point algorithm maximizes the likelihood of the ICA model under the constraint of decorrelation and uses the method of Lee et al. (Neural Comput. 11(2) (1999) 417) to switch between sub- and super-Gaussian regimes. The new fixed-point algorithm maximizes the likelihood very fast and reliably. The validity of this algorithm is confirmed by the simulations and experimental results.

AB - A new fixed-point algorithm for independent component analysis (ICA) is presented that is able blindly to separate mixed signals with sub- and super-Gaussian source distributions. The new fixed-point algorithm maximizes the likelihood of the ICA model under the constraint of decorrelation and uses the method of Lee et al. (Neural Comput. 11(2) (1999) 417) to switch between sub- and super-Gaussian regimes. The new fixed-point algorithm maximizes the likelihood very fast and reliably. The validity of this algorithm is confirmed by the simulations and experimental results.

KW - Blind source separation

KW - Fixed-point algorithm

KW - Independent component analysis

UR - http://www.scopus.com/inward/record.url?scp=0742324116&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0742324116&partnerID=8YFLogxK

U2 - 10.1016/j.neucom.2003.09.002

DO - 10.1016/j.neucom.2003.09.002

M3 - Article

AN - SCOPUS:0742324116

SN - 0925-2312

VL - 56

SP - 467

EP - 473

JO - Neurocomputing

JF - Neurocomputing

IS - 1-4

ER -

A new fixed-point algorithm for independent component analysis

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this