A partitioned frequency domain algorithm for convolutive blind source separation

Author

Scarpiniti, Michele ; Picaro, Andrea ; Parisi, Raffaele ; Uncini, Aurelio

Author_Institution

Infocom Dept., Sapienza Univ. of Rome, Rome, Italy

fYear

2009

fDate

1-4 Sept. 2009

Firstpage

1

Lastpage

6

Abstract

The aim of this paper is to introduce a blind source separation algorithm in reverberant environment, usually characterized by long impulsive responses. In order to reduce the computational complexity of this kind of algorithms a partitioned frequency domain approach is proposed by partitioning the demixing filter in an optimal number of sub-filters. Several experimental results are shown to demonstrate the effectiveness of the proposed method.

Keywords

blind source separation; computational complexity; convolution; filtering theory; frequency-domain analysis; transient response; computational complexity; convolutive blind source separation algorithm; demixing filter; impulsive response; partitioned frequency domain algorithm; Blind source separation; Computational complexity; Convolution; Finite impulse response filter; Frequency domain analysis; Frequency estimation; Partitioning algorithms; Reverberation; Sensor arrays; Source separation;

fLanguage

English

Publisher

ieee

Conference_Titel

Machine Learning for Signal Processing, 2009. MLSP 2009. IEEE International Workshop on

Conference_Location

Grenoble

Print_ISBN

978-1-4244-4947-7

Electronic_ISBN

978-1-4244-4948-4

Type

conf

DOI

10.1109/MLSP.2009.5306200

Filename

5306200