Blind extraction of temporally correlated but statistically dependent acoustic signals

Andrzej Cichocki, Tomasz Rutkowski, Allan Kardec Barros, Sang Hoon Oh

Research output: Contribution to conferencePaperpeer-review

20 Citations (Scopus)

Abstract

In this paper we propose a batch learning algorithm for sequential blind extraction of arbitrary distributed but generally not i.i.d. (independent identically distributed) temporally correlated sources, possibly dependent speech signals from linear mixture of them. The proposed algorithm is computationally very simple and efficient, it is based only on the second order statistics and in contrast to the most known algorithms developed for the sequential blind extraction and independent component analysis, do not assume statistical independence of source signals neither non-zero kurtosis for the sources, thus statistical dependent signals including sources with extremely low or even zero kurtosis (colored Gaussian with different spectra) can be also successfully extracted. Extensive computer simulation confirm the validity and high performance of the proposed algorithm.

Original languageEnglish
Pages455-464
Number of pages10
Publication statusPublished - 2000
Externally publishedYes
Event10th IEEE Workshop on Neural Netwoks for Signal Processing (NNSP2000) - Sydney, Australia
Duration: 11 Dec 200013 Dec 2000

Conference

Conference10th IEEE Workshop on Neural Netwoks for Signal Processing (NNSP2000)
CitySydney, Australia
Period11/12/0013/12/00

Fingerprint

Dive into the research topics of 'Blind extraction of temporally correlated but statistically dependent acoustic signals'. Together they form a unique fingerprint.

Cite this