On separation performance enhancement in convolutive blind source separation

R. Mazur; A. Mertins

doi:10.1109/ACSSC.2008.5074719

On separation performance enhancement in convolutive blind source separation

Institute of Signal Processing

Abstract

In this paper we present a method for estimating the bin-wise separation performance in convolutive blind source separation. A common way to separate convolutive mixtures is the transformation to the time-frequency domain and the separation of the single bins using an instantaneous ICA algorithm. This approach reduces the complexity but leads to the so-called permutation problem which has been widely studied. Another problem arises when the single bins are only poorly separable or even not separable at all. These bins can significantly reduce the overall performance. In this paper we propose a method for detecting such bins based on properties of the unmixing matrices.

Original language	English
Title of host publication	2008 42nd Asilomar Conference on Signals, Systems and Computers
Number of pages	4
Publisher	IEEE
Publication date	01.10.2008
Pages	1718-1721
Article number	5074719
ISBN (Print)	978-1-4244-2940-0
ISBN (Electronic)	978-1-4244-2941-7
DOIs	https://doi.org/10.1109/ACSSC.2008.5074719
Publication status	Published - 01.10.2008
Event	2008 42nd Asilomar Conference on Signals, Systems and Computers - Pacific Grove, United States Duration: 26.10.2008 → 29.10.2008 Conference number: 77519

UN SDGs

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

Access to Document

10.1109/ACSSC.2008.5074719

Cite this

@inproceedings{71f6634c1a75431b8a354a2cc9ab4339,

title = "On separation performance enhancement in convolutive blind source separation",

abstract = "In this paper we present a method for estimating the bin-wise separation performance in convolutive blind source separation. A common way to separate convolutive mixtures is the transformation to the time-frequency domain and the separation of the single bins using an instantaneous ICA algorithm. This approach reduces the complexity but leads to the so-called permutation problem which has been widely studied. Another problem arises when the single bins are only poorly separable or even not separable at all. These bins can significantly reduce the overall performance. In this paper we propose a method for detecting such bins based on properties of the unmixing matrices.",

author = "R. Mazur and A. Mertins",

year = "2008",

month = oct,

day = "1",

doi = "10.1109/ACSSC.2008.5074719",

language = "English",

isbn = "978-1-4244-2940-0",

series = "Conference Record - Asilomar Conference on Signals, Systems and Computers",

publisher = "IEEE",

pages = "1718--1721",

booktitle = "2008 42nd Asilomar Conference on Signals, Systems and Computers",

note = "2008 42nd Asilomar Conference on Signals, Systems and Computers, ASILOMAR 2008 ; Conference date: 26-10-2008 Through 29-10-2008",

}

Mazur, R & Mertins, A 2008, On separation performance enhancement in convolutive blind source separation. in 2008 42nd Asilomar Conference on Signals, Systems and Computers., 5074719, Conference Record - Asilomar Conference on Signals, Systems and Computers, IEEE, pp. 1718-1721, 2008 42nd Asilomar Conference on Signals, Systems and Computers, Pacific Grove, California, United States, 26.10.08. https://doi.org/10.1109/ACSSC.2008.5074719

On separation performance enhancement in convolutive blind source separation. / Mazur, R.; Mertins, A.
2008 42nd Asilomar Conference on Signals, Systems and Computers. IEEE, 2008. p. 1718-1721 5074719 (Conference Record - Asilomar Conference on Signals, Systems and Computers).

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

TY - GEN

T1 - On separation performance enhancement in convolutive blind source separation

AU - Mazur, R.

AU - Mertins, A.

N1 - Conference code: 77519

PY - 2008/10/1

Y1 - 2008/10/1

N2 - In this paper we present a method for estimating the bin-wise separation performance in convolutive blind source separation. A common way to separate convolutive mixtures is the transformation to the time-frequency domain and the separation of the single bins using an instantaneous ICA algorithm. This approach reduces the complexity but leads to the so-called permutation problem which has been widely studied. Another problem arises when the single bins are only poorly separable or even not separable at all. These bins can significantly reduce the overall performance. In this paper we propose a method for detecting such bins based on properties of the unmixing matrices.

AB - In this paper we present a method for estimating the bin-wise separation performance in convolutive blind source separation. A common way to separate convolutive mixtures is the transformation to the time-frequency domain and the separation of the single bins using an instantaneous ICA algorithm. This approach reduces the complexity but leads to the so-called permutation problem which has been widely studied. Another problem arises when the single bins are only poorly separable or even not separable at all. These bins can significantly reduce the overall performance. In this paper we propose a method for detecting such bins based on properties of the unmixing matrices.

UR - https://www.scopus.com/record/display.uri?eid=2-s2.0-70349690007&origin=inward&txGid=3103670e4ddc332cce9ea3402cfa99a8

U2 - 10.1109/ACSSC.2008.5074719

DO - 10.1109/ACSSC.2008.5074719

M3 - Conference contribution

SN - 978-1-4244-2940-0

T3 - Conference Record - Asilomar Conference on Signals, Systems and Computers

SP - 1718

EP - 1721

BT - 2008 42nd Asilomar Conference on Signals, Systems and Computers

PB - IEEE

T2 - 2008 42nd Asilomar Conference on Signals, Systems and Computers

Y2 - 26 October 2008 through 29 October 2008

ER -

On separation performance enhancement in convolutive blind source separation

Abstract

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this