Convolutive blind source separation for more than two sources in the frequency domain

Hiroshi Sawada, Ryo Mukai, Shoko Araki, Shoji Makino

Research output: Contribution to journalArticlepeer-review

4 Citations (Scopus)

Abstract

The use of blind source separation (BSS) technique for the recovery of more than two sources inthe frequency domain was iprensented. It was found that frequency-domain BSS method was practically applicable for more than two sources by overcoming problem of permutation and circularity. The minimization error could be done by adjusting the scaling ambiguity of the independent component analysis (ICA) solution before windowing. The result shows that the effectiveness and efficiency of the BSS method and the separation of six sources with a planar array of eight sensors.

Original languageEnglish
Pages (from-to)296-298
Number of pages3
JournalAcoustical Science and Technology
Volume25
Issue number4
DOIs
Publication statusPublished - 2004 Jul
Externally publishedYes

Keywords

  • Blind source separation
  • Convolutive mixtures
  • Independent component analysis

ASJC Scopus subject areas

  • Acoustics and Ultrasonics

Fingerprint Dive into the research topics of 'Convolutive blind source separation for more than two sources in the frequency domain'. Together they form a unique fingerprint.

Cite this