Adaptive sparse factorization for even-determined and over-determined blind source separation

Fuxiang Wang; Jun Zhang

doi:10.1109/CISE.2009.5366379

Adaptive sparse factorization for even-determined and over-determined blind source separation

Fuxiang Wang^*, Jun Zhang

^*Corresponding author for this work

Beihang University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

In this paper, we present an adaptive sparse factorization method for even-determined and over-determined blind source separation, where the sources are assumed to be sparse. The objective of our method is to find a demixing matrix to make the output signal as sparse as possible. First, a cost function measuring the sparsity of the output signals is introduced. Then an adaptive algorithm for the learning of the demixing matrix is proposed by the nature gradient. Compared to Independent Component Analysis, the new method can deal with the mixing cases that the sources are mutually correlated. Simulation results show the effectiveness of the new method.

Original language	English
Title of host publication	Proceedings - 2009 International Conference on Computational Intelligence and Software Engineering, CiSE 2009
DOIs	https://doi.org/10.1109/CISE.2009.5366379
Publication status	Published - 2009
Externally published	Yes
Event	2009 International Conference on Computational Intelligence and Software Engineering, CiSE 2009 - Wuhan, China Duration: 11 Dec 2009 → 13 Dec 2009

Publication series

Name	Proceedings - 2009 International Conference on Computational Intelligence and Software Engineering, CiSE 2009

Conference

Conference	2009 International Conference on Computational Intelligence and Software Engineering, CiSE 2009
Country/Territory	China
City	Wuhan
Period	11/12/09 → 13/12/09

Keywords

Blind source separation
Independent component analysis
Penalty function
Sparse representation

Access to Document

10.1109/CISE.2009.5366379

Cite this

Wang, F., & Zhang, J. (2009). Adaptive sparse factorization for even-determined and over-determined blind source separation. In Proceedings - 2009 International Conference on Computational Intelligence and Software Engineering, CiSE 2009 Article 5366379 (Proceedings - 2009 International Conference on Computational Intelligence and Software Engineering, CiSE 2009). https://doi.org/10.1109/CISE.2009.5366379

@inproceedings{785eebe45e97414082700ed6921e23e5,

title = "Adaptive sparse factorization for even-determined and over-determined blind source separation",

abstract = "In this paper, we present an adaptive sparse factorization method for even-determined and over-determined blind source separation, where the sources are assumed to be sparse. The objective of our method is to find a demixing matrix to make the output signal as sparse as possible. First, a cost function measuring the sparsity of the output signals is introduced. Then an adaptive algorithm for the learning of the demixing matrix is proposed by the nature gradient. Compared to Independent Component Analysis, the new method can deal with the mixing cases that the sources are mutually correlated. Simulation results show the effectiveness of the new method.",

keywords = "Blind source separation, Independent component analysis, Penalty function, Sparse representation",

author = "Fuxiang Wang and Jun Zhang",

year = "2009",

doi = "10.1109/CISE.2009.5366379",

language = "English",

isbn = "9781424445073",

series = "Proceedings - 2009 International Conference on Computational Intelligence and Software Engineering, CiSE 2009",

booktitle = "Proceedings - 2009 International Conference on Computational Intelligence and Software Engineering, CiSE 2009",

note = "2009 International Conference on Computational Intelligence and Software Engineering, CiSE 2009 ; Conference date: 11-12-2009 Through 13-12-2009",

}

Wang, F & Zhang, J 2009, Adaptive sparse factorization for even-determined and over-determined blind source separation. in Proceedings - 2009 International Conference on Computational Intelligence and Software Engineering, CiSE 2009., 5366379, Proceedings - 2009 International Conference on Computational Intelligence and Software Engineering, CiSE 2009, 2009 International Conference on Computational Intelligence and Software Engineering, CiSE 2009, Wuhan, China, 11/12/09. https://doi.org/10.1109/CISE.2009.5366379

Adaptive sparse factorization for even-determined and over-determined blind source separation. / Wang, Fuxiang; Zhang, Jun.
Proceedings - 2009 International Conference on Computational Intelligence and Software Engineering, CiSE 2009. 2009. 5366379 (Proceedings - 2009 International Conference on Computational Intelligence and Software Engineering, CiSE 2009).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Adaptive sparse factorization for even-determined and over-determined blind source separation

AU - Wang, Fuxiang

AU - Zhang, Jun

PY - 2009

Y1 - 2009

N2 - In this paper, we present an adaptive sparse factorization method for even-determined and over-determined blind source separation, where the sources are assumed to be sparse. The objective of our method is to find a demixing matrix to make the output signal as sparse as possible. First, a cost function measuring the sparsity of the output signals is introduced. Then an adaptive algorithm for the learning of the demixing matrix is proposed by the nature gradient. Compared to Independent Component Analysis, the new method can deal with the mixing cases that the sources are mutually correlated. Simulation results show the effectiveness of the new method.

AB - In this paper, we present an adaptive sparse factorization method for even-determined and over-determined blind source separation, where the sources are assumed to be sparse. The objective of our method is to find a demixing matrix to make the output signal as sparse as possible. First, a cost function measuring the sparsity of the output signals is introduced. Then an adaptive algorithm for the learning of the demixing matrix is proposed by the nature gradient. Compared to Independent Component Analysis, the new method can deal with the mixing cases that the sources are mutually correlated. Simulation results show the effectiveness of the new method.

KW - Blind source separation

KW - Independent component analysis

KW - Penalty function

KW - Sparse representation

UR - http://www.scopus.com/inward/record.url?scp=77949705080&partnerID=8YFLogxK

U2 - 10.1109/CISE.2009.5366379

DO - 10.1109/CISE.2009.5366379

M3 - Conference contribution

AN - SCOPUS:77949705080

SN - 9781424445073

T3 - Proceedings - 2009 International Conference on Computational Intelligence and Software Engineering, CiSE 2009

BT - Proceedings - 2009 International Conference on Computational Intelligence and Software Engineering, CiSE 2009

T2 - 2009 International Conference on Computational Intelligence and Software Engineering, CiSE 2009

Y2 - 11 December 2009 through 13 December 2009

ER -

Wang F, Zhang J. Adaptive sparse factorization for even-determined and over-determined blind source separation. In Proceedings - 2009 International Conference on Computational Intelligence and Software Engineering, CiSE 2009. 2009. 5366379. (Proceedings - 2009 International Conference on Computational Intelligence and Software Engineering, CiSE 2009). doi: 10.1109/CISE.2009.5366379

Adaptive sparse factorization for even-determined and over-determined blind source separation

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this