A new proof of a contrast function for bounded component analysis and further analysis

Wei Gao; Shen Fan; Roberto Togneri; Victor Sreeram

doi:10.1016/j.csl.2018.03.002

A new proof of a contrast function for bounded component analysis and further analysis

Wei Gao, Shen Fan, Roberto Togneri, Victor Sreeram

School of Engineering

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

1 Citation (Scopus)

87 Downloads (Pure)

Abstract

Bounded Component Analysis (BCA) solves the Blind Source Separation (BSS) problem based on geometric assumptions. This paper introduces a new proof of a BCA contrast function, derived from elementary matrices, Gauss–Jordan elimination and convex geometry. The new proof and further analysis provide additional insight into a key assumption of BCA. In addition, an interpretation is presented to clarify one of the limitations of the instantaneous BCA algorithm. Experiments on audio sources support our analysis.

Original language	English
Journal	Computer Speech and Language
Volume	51
DOIs	https://doi.org/10.1016/j.csl.2018.03.002
Publication status	Published - 4 Apr 2018

Access to Document

10.1016/j.csl.2018.03.002

Gao et.al., 2018 A new proof of a contrast function for bounded component analysis and further analysisAccepted author manuscript, 602 KBLicence: CC BY-NC-ND

Cite this

@article{ef62faf326e84d0c98e9cd6651e3b0ca,

title = "A new proof of a contrast function for bounded component analysis and further analysis",

abstract = "Bounded Component Analysis (BCA) solves the Blind Source Separation (BSS) problem based on geometric assumptions. This paper introduces a new proof of a BCA contrast function, derived from elementary matrices, Gauss–Jordan elimination and convex geometry. The new proof and further analysis provide additional insight into a key assumption of BCA. In addition, an interpretation is presented to clarify one of the limitations of the instantaneous BCA algorithm. Experiments on audio sources support our analysis.",

keywords = "Audio signals, Blind source separation, Bounded component analysis, Convex geometry, Elementary matrix, Independent component analysis",

author = "Wei Gao and Shen Fan and Roberto Togneri and Victor Sreeram",

year = "2018",

month = apr,

day = "4",

doi = "10.1016/j.csl.2018.03.002",

language = "English",

volume = "51",

journal = "Computer Speech and Language",

issn = "0885-2308",

publisher = "Academic Press",

}

TY - JOUR

T1 - A new proof of a contrast function for bounded component analysis and further analysis

AU - Gao, Wei

AU - Fan, Shen

AU - Togneri, Roberto

AU - Sreeram, Victor

PY - 2018/4/4

Y1 - 2018/4/4

N2 - Bounded Component Analysis (BCA) solves the Blind Source Separation (BSS) problem based on geometric assumptions. This paper introduces a new proof of a BCA contrast function, derived from elementary matrices, Gauss–Jordan elimination and convex geometry. The new proof and further analysis provide additional insight into a key assumption of BCA. In addition, an interpretation is presented to clarify one of the limitations of the instantaneous BCA algorithm. Experiments on audio sources support our analysis.

AB - Bounded Component Analysis (BCA) solves the Blind Source Separation (BSS) problem based on geometric assumptions. This paper introduces a new proof of a BCA contrast function, derived from elementary matrices, Gauss–Jordan elimination and convex geometry. The new proof and further analysis provide additional insight into a key assumption of BCA. In addition, an interpretation is presented to clarify one of the limitations of the instantaneous BCA algorithm. Experiments on audio sources support our analysis.

KW - Audio signals

KW - Blind source separation

KW - Bounded component analysis

KW - Convex geometry

KW - Elementary matrix

KW - Independent component analysis

UR - http://www.scopus.com/inward/record.url?scp=85045638101&partnerID=8YFLogxK

U2 - 10.1016/j.csl.2018.03.002

DO - 10.1016/j.csl.2018.03.002

M3 - Article

AN - SCOPUS:85045638101

SN - 0885-2308

VL - 51

JO - Computer Speech and Language

JF - Computer Speech and Language

ER -

A new proof of a contrast function for bounded component analysis and further analysis

Abstract

Access to Document

Other files and links

Fingerprint

Cite this