A Method for Filter Equalization in Convolutive Blind Source Separation

Radoslaw Mazur; Alfred Mertins

doi:10.1007/978-3-642-15995-4_41

A Method for Filter Equalization in Convolutive Blind Source Separation

Institute of Signal Processing

Abstract

The separation of convolutive mixed signals can be carried out in the time-frequency domain, where the task is reduced to multiple instantaneous problems. This direct approach leads to the permutation and scaling problems, but it is possible to introduce an objective function in the time-frequency domain and minimize it with respect to the time domain coefficients. While this approach allows for the elimination of the permutation problem, the unmixing filters can be quite distorted due the unsolved scaling problem. In this paper we propose a method for equalization of these filters by using the scaling ambiguity. The resulting filters have a characteristic of a Dirac pulse and introduce less distortion to the separated signals. The results are shown on a real-world example.

Original language	English
Title of host publication	Latent Variable Analysis and Signal Separation
Editors	Vincent Vigneron, Vicente Zarzoso, Eric Moreau, Rémi Gribonval, Emmanuel Vincent
Number of pages	9
Volume	6365 LNCS
Place of Publication	Berlin, Heidelberg
Publisher	Springer Berlin Heidelberg
Publication date	01.09.2010
Pages	328-336
ISBN (Print)	978-3-642-15994-7
ISBN (Electronic)	978-3-642-15995-4
DOIs	https://doi.org/10.1007/978-3-642-15995-4_41
Publication status	Published - 01.09.2010

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 9 Industry, Innovation, and Infrastructure

Access to Document

10.1007/978-3-642-15995-4_41

Cite this

Mazur, R., & Mertins, A. (2010). A Method for Filter Equalization in Convolutive Blind Source Separation. In V. Vigneron, V. Zarzoso, E. Moreau, R. Gribonval, & E. Vincent (Eds.), Latent Variable Analysis and Signal Separation (Vol. 6365 LNCS, pp. 328-336). ( Lecture Notes in Computer Science; Vol. 6365 LNCS). Springer Berlin Heidelberg. https://doi.org/10.1007/978-3-642-15995-4_41

Mazur, Radoslaw ; Mertins, Alfred. / A Method for Filter Equalization in Convolutive Blind Source Separation. Latent Variable Analysis and Signal Separation. editor / Vincent Vigneron ; Vicente Zarzoso ; Eric Moreau ; Rémi Gribonval ; Emmanuel Vincent. Vol. 6365 LNCS Berlin, Heidelberg : Springer Berlin Heidelberg, 2010. pp. 328-336 ( Lecture Notes in Computer Science).

@inproceedings{dfe8be1d7d2b48918757cf58ab58e5ed,

title = "A Method for Filter Equalization in Convolutive Blind Source Separation",

abstract = "The separation of convolutive mixed signals can be carried out in the time-frequency domain, where the task is reduced to multiple instantaneous problems. This direct approach leads to the permutation and scaling problems, but it is possible to introduce an objective function in the time-frequency domain and minimize it with respect to the time domain coefficients. While this approach allows for the elimination of the permutation problem, the unmixing filters can be quite distorted due the unsolved scaling problem. In this paper we propose a method for equalization of these filters by using the scaling ambiguity. The resulting filters have a characteristic of a Dirac pulse and introduce less distortion to the separated signals. The results are shown on a real-world example.",

author = "Radoslaw Mazur and Alfred Mertins",

year = "2010",

month = sep,

day = "1",

doi = "10.1007/978-3-642-15995-4\_41",

language = "English",

isbn = "978-3-642-15994-7",

volume = "6365 LNCS",

series = " Lecture Notes in Computer Science",

publisher = "Springer Berlin Heidelberg",

pages = "328--336",

editor = "Vincent Vigneron and Vicente Zarzoso and Eric Moreau and R{\'e}mi Gribonval and Emmanuel Vincent",

booktitle = "Latent Variable Analysis and Signal Separation",

}

Mazur, R & Mertins, A 2010, A Method for Filter Equalization in Convolutive Blind Source Separation. in V Vigneron, V Zarzoso, E Moreau, R Gribonval & E Vincent (eds), Latent Variable Analysis and Signal Separation. vol. 6365 LNCS, Lecture Notes in Computer Science, vol. 6365 LNCS, Springer Berlin Heidelberg, Berlin, Heidelberg, pp. 328-336. https://doi.org/10.1007/978-3-642-15995-4_41

A Method for Filter Equalization in Convolutive Blind Source Separation. / Mazur, Radoslaw; Mertins, Alfred.
Latent Variable Analysis and Signal Separation. ed. / Vincent Vigneron; Vicente Zarzoso; Eric Moreau; Rémi Gribonval; Emmanuel Vincent. Vol. 6365 LNCS Berlin, Heidelberg: Springer Berlin Heidelberg, 2010. p. 328-336 ( Lecture Notes in Computer Science; Vol. 6365 LNCS).

Research output: Chapters in Books/Reports/Conference Proceedings › Conference contribution › peer-review

TY - GEN

T1 - A Method for Filter Equalization in Convolutive Blind Source Separation

AU - Mazur, Radoslaw

AU - Mertins, Alfred

PY - 2010/9/1

Y1 - 2010/9/1

N2 - The separation of convolutive mixed signals can be carried out in the time-frequency domain, where the task is reduced to multiple instantaneous problems. This direct approach leads to the permutation and scaling problems, but it is possible to introduce an objective function in the time-frequency domain and minimize it with respect to the time domain coefficients. While this approach allows for the elimination of the permutation problem, the unmixing filters can be quite distorted due the unsolved scaling problem. In this paper we propose a method for equalization of these filters by using the scaling ambiguity. The resulting filters have a characteristic of a Dirac pulse and introduce less distortion to the separated signals. The results are shown on a real-world example.

AB - The separation of convolutive mixed signals can be carried out in the time-frequency domain, where the task is reduced to multiple instantaneous problems. This direct approach leads to the permutation and scaling problems, but it is possible to introduce an objective function in the time-frequency domain and minimize it with respect to the time domain coefficients. While this approach allows for the elimination of the permutation problem, the unmixing filters can be quite distorted due the unsolved scaling problem. In this paper we propose a method for equalization of these filters by using the scaling ambiguity. The resulting filters have a characteristic of a Dirac pulse and introduce less distortion to the separated signals. The results are shown on a real-world example.

UR - https://www.scopus.com/pages/publications/78349261606

U2 - 10.1007/978-3-642-15995-4_41

DO - 10.1007/978-3-642-15995-4_41

M3 - Conference contribution

SN - 978-3-642-15994-7

VL - 6365 LNCS

T3 - Lecture Notes in Computer Science

SP - 328

EP - 336

BT - Latent Variable Analysis and Signal Separation

A2 - Vigneron, Vincent

A2 - Zarzoso, Vicente

A2 - Moreau, Eric

A2 - Gribonval, Rémi

A2 - Vincent, Emmanuel

PB - Springer Berlin Heidelberg

CY - Berlin, Heidelberg

ER -

A Method for Filter Equalization in Convolutive Blind Source Separation

Abstract

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this