Quadratic MIMO contrast functions for blind source separation in a convolutive context

Authors:
Saloua Rhioui;Marc Castella;Eric Moreau
Affiliations:
STD, ISITV, La Valette du Var, France;Département CITI, UMR-CNRS 5157, GET/INT, Évry, France;STD, ISITV, La Valette du Var, France
Venue:
ICA'06 Proceedings of the 6th international conference on Independent Component Analysis and Blind Signal Separation
Year:
2006

Citing 3
Cited 3

Separation of a class of convolutive mixtures: a contrast function approach

Signal Processing
Identification and deconvolution of multichannel linearnon-Gaussian processes using higher order statistics and inverse filtercriteria

IEEE Transactions on Signal Processing
Blind equalization of MIMO-FIR channels driven by white but higher order colored source signals

IEEE Transactions on Information Theory

A super-exponential deflation method incorporated with higher-order correlations for blind deconvolution of MIMO linear systems

Signal Processing - Special section: Distributed source coding
Blind Deconvolution of MIMO-IIR Systems: A Two-Stage EVA

Neural Information Processing
An eigenvector algorithm with reference signals using a deflation approach for blind deconvolution

ICA'07 Proceedings of the 7th international conference on Independent component analysis and signal separation

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper considers the problem of blind separation of sources mixed by a MIMO convolutive system. For both i.i.d. and non i.i.d. sources, quadratic separation criteria previously designed for the extraction of a single source are extended to parallel extraction in the MIMO case. These criteria are based on the use of so-called reference signals and a condition is given under which we obtain MIMO contrast functions. Simulations demonstrate that a particular choice of a set of reference signals ensures the contrast property. The performance offered by these criteria is investigated through simulations: it is shown that the proposed contrast functions avoid accumulation errors, contrary to deflation methods.