A fast asymptotically efficient algorithm for blind separation of a linear mixture of block-wise stationary autoregressive processes

Petr Tichavský; Arie Yeredor; Zbyněk Koldovský

doi:10.1109/ICASSP.2009.4960288

A fast asymptotically efficient algorithm for blind separation of a linear mixture of block-wise stationary autoregressive processes

Petr Tichavský^*, Arie Yeredor, Zbyněk Koldovský

^*Corresponding author for this work

School of Electrical Engineering

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

Abstract

We propose a novel blind source separation algorithm called Block AutoRegressive Blind Identification (BARBI). The algorithm is asymptotically efficient in separation of instantaneous linear mixtures of blockwise stationary Gaussian autoregressive processes. A novel closed-form formula is derived for a Cramér Rao lower bound on elements of the corresponding Interference-to-Signal Ratio (ISR) matrix. This theoretical ISR matrix can serve as an estimate of the separation performance on the particular data. In simulations, the algorithm is shown to be applicable in blind separation of a linear mixture of speech signals.

Original language	English
Title of host publication	2009 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings, ICASSP 2009
Pages	3133-3136
Number of pages	4
DOIs	https://doi.org/10.1109/ICASSP.2009.4960288
State	Published - 2009
Event	2009 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2009 - Taipei, Taiwan, Province of China Duration: 19 Apr 2009 → 24 Apr 2009

Publication series

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

Conference

Conference	2009 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2009
Country/Territory	Taiwan, Province of China
City	Taipei
Period	19/04/09 → 24/04/09

Keywords

Approximate joint diagonalization
Autoregressive processes
Blind source separation
Second-order statistics

Access to Document

10.1109/ICASSP.2009.4960288

Cite this

Tichavský, P., Yeredor, A., & Koldovský, Z. (2009). A fast asymptotically efficient algorithm for blind separation of a linear mixture of block-wise stationary autoregressive processes. In 2009 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings, ICASSP 2009 (pp. 3133-3136). [4960288] (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings). https://doi.org/10.1109/ICASSP.2009.4960288

Tichavský, Petr ; Yeredor, Arie ; Koldovský, Zbyněk. / A fast asymptotically efficient algorithm for blind separation of a linear mixture of block-wise stationary autoregressive processes. 2009 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings, ICASSP 2009. 2009. pp. 3133-3136 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings).

@inproceedings{16f283ae717948da9ff6356abe31862c,

title = "A fast asymptotically efficient algorithm for blind separation of a linear mixture of block-wise stationary autoregressive processes",

abstract = "We propose a novel blind source separation algorithm called Block AutoRegressive Blind Identification (BARBI). The algorithm is asymptotically efficient in separation of instantaneous linear mixtures of blockwise stationary Gaussian autoregressive processes. A novel closed-form formula is derived for a Cram{\'e}r Rao lower bound on elements of the corresponding Interference-to-Signal Ratio (ISR) matrix. This theoretical ISR matrix can serve as an estimate of the separation performance on the particular data. In simulations, the algorithm is shown to be applicable in blind separation of a linear mixture of speech signals.",

keywords = "Approximate joint diagonalization, Autoregressive processes, Blind source separation, Second-order statistics",

author = "Petr Tichavsk{\'y} and Arie Yeredor and Zbyn{\v e}k Koldovsk{\'y}",

year = "2009",

doi = "10.1109/ICASSP.2009.4960288",

language = "אנגלית",

isbn = "9781424423545",

series = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",

pages = "3133--3136",

booktitle = "2009 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings, ICASSP 2009",

note = "null ; Conference date: 19-04-2009 Through 24-04-2009",

}

Tichavský, P, Yeredor, A & Koldovský, Z 2009, A fast asymptotically efficient algorithm for blind separation of a linear mixture of block-wise stationary autoregressive processes. in 2009 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings, ICASSP 2009., 4960288, ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, pp. 3133-3136, 2009 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2009, Taipei, Taiwan, Province of China, 19/04/09. https://doi.org/10.1109/ICASSP.2009.4960288

A fast asymptotically efficient algorithm for blind separation of a linear mixture of block-wise stationary autoregressive processes. / Tichavský, Petr; Yeredor, Arie; Koldovský, Zbyněk.

2009 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings, ICASSP 2009. 2009. p. 3133-3136 4960288 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings).

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

TY - GEN

T1 - A fast asymptotically efficient algorithm for blind separation of a linear mixture of block-wise stationary autoregressive processes

AU - Tichavský, Petr

AU - Yeredor, Arie

AU - Koldovský, Zbyněk

PY - 2009

Y1 - 2009

N2 - We propose a novel blind source separation algorithm called Block AutoRegressive Blind Identification (BARBI). The algorithm is asymptotically efficient in separation of instantaneous linear mixtures of blockwise stationary Gaussian autoregressive processes. A novel closed-form formula is derived for a Cramér Rao lower bound on elements of the corresponding Interference-to-Signal Ratio (ISR) matrix. This theoretical ISR matrix can serve as an estimate of the separation performance on the particular data. In simulations, the algorithm is shown to be applicable in blind separation of a linear mixture of speech signals.

AB - We propose a novel blind source separation algorithm called Block AutoRegressive Blind Identification (BARBI). The algorithm is asymptotically efficient in separation of instantaneous linear mixtures of blockwise stationary Gaussian autoregressive processes. A novel closed-form formula is derived for a Cramér Rao lower bound on elements of the corresponding Interference-to-Signal Ratio (ISR) matrix. This theoretical ISR matrix can serve as an estimate of the separation performance on the particular data. In simulations, the algorithm is shown to be applicable in blind separation of a linear mixture of speech signals.

KW - Approximate joint diagonalization

KW - Autoregressive processes

KW - Blind source separation

KW - Second-order statistics

UR - http://www.scopus.com/inward/record.url?scp=70349198991&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2009.4960288

DO - 10.1109/ICASSP.2009.4960288

M3 - ???researchoutput.researchoutputtypes.contributiontobookanthology.conference???

AN - SCOPUS:70349198991

SN - 9781424423545

T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

SP - 3133

EP - 3136

BT - 2009 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings, ICASSP 2009

Y2 - 19 April 2009 through 24 April 2009

ER -

Tichavský P, Yeredor A, Koldovský Z. A fast asymptotically efficient algorithm for blind separation of a linear mixture of block-wise stationary autoregressive processes. In 2009 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings, ICASSP 2009. 2009. p. 3133-3136. 4960288. (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings). doi: 10.1109/ICASSP.2009.4960288

A fast asymptotically efficient algorithm for blind separation of a linear mixture of block-wise stationary autoregressive processes

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this