Blind Source Separation of Overdetermined Linear-Quadratic Mixtures

Leonardo Tomazeli Duarte (1) , Ricardo Suyama (2) , Romis Attux (1) , Yannick Deville (3) , Joao M. T. Romano (1) , Christian Jutten (4)

1 UNICAMP - Universidade Estadual de Campinas = University of Campinas
2 UFABC - Universidade Federal do ABC = Federal University of ABC = Université Fédérale de l'ABC [Brazil]
3 LATT - Laboratoire d'Astrophysique de l'Observatoire Midi-Pyrénées
4 GIPSA-VIBS - GIPSA - Vision and Brain Signal Processing

This work deals with the problem of source separation in overdetermined linear-quadratic (LQ) models. Although the mixing model in this situation can be inverted by linear structures, we show that some simple independent component analysis (ICA) strategies that are often employed in the linear case cannot be used with the studied model. Motivated by this fact, we consider the more complex yet more robust ICA framework based on the minimization of the mutual information. Special attention is given to the development of a solution that be as robust as possible to suboptimal convergences. This is achieved by defining a method composed of a global optimization step followed by a local search procedure. Simulations confirm the effectiveness of the proposal.

Mots clés: nonlinear; linear-quadratic; ICA; mutual information

Domaines: Traitement du signal et de l'image [eess.SP]; Traitement du signal et de l'image [eess.SP]

Leonardo Tomazeli Duarte, Ricardo Suyama, Romis Attux, Yannick Deville, Joao M. T. Romano, et al.. Blind Source Separation of Overdetermined Linear-Quadratic Mixtures. LVA/ICA 2010 - 9th International Conference on Latent Variable Analysis and Signal Separation, Sep 2010, Saint Malo, France. pp.263-270. ⟨hal-00525945⟩