Blind Separation of Instantaneous Mixtures of Dependent Sources

Abstract : This paper deals with the problem of Blind Source Separation. Contrary to the vast majority of works, we do not assume the statistical independence between sources and explicitly consider that they are dependent. We introduce three particular types of dependent sources and show that their cumulants have interesting properties. Based on these properties, we investigate the behavior of classical Blind Source Separation algorithms when applied to these sources: depending on the source vector, the separation may be successful or some additionnal indeterminacies can be identified.
Liste complète des métadonnées

Cited literature [9 references]  Display  Hide  Download
Contributor : Pierre Comon <>
Submitted on : Thursday, October 9, 2008 - 10:43:58 AM
Last modification on : Thursday, February 7, 2019 - 4:44:49 PM
Document(s) archivé(s) le : Friday, June 4, 2010 - 12:23:46 PM


Files produced by the author(s)


  • HAL Id : hal-00327652, version 1


Marc Castella, Pierre Comon. Blind Separation of Instantaneous Mixtures of Dependent Sources. 7th International Conference on Independent Component Analysis and Signal Separation, Sep 2007, London, United Kingdom. pp.9-16. ⟨hal-00327652⟩



Record views


Files downloads