A robust algorithm for convolutive blind source separation in presence of noise

M. El Rhabi (1) , H. Fenniri (2) , A. Keziou (2) , E. Moreau (3)

1 IMI - Département Ingénierie Mathématique et Informatique, École des Ponts ParisTech
2 CRESTIC - Centre de Recherche en Sciences et Technologies de l'Information et de la Communication - EA 3804
3 LIS - Laboratoire d'Informatique et Systèmes

We consider the blind source separation (BSS) problem in the noisy context. We propose a new methodology in order to enhance separation performances in terms of efficiency and robustness. Our approach consists in denoising the observed signals through the minimization of their total variation, and then minimizing divergence separation criteria combined with the total variation of the estimated source signals. We show by the way that the method leads to some projection problems that are solved by means of projected gradient algorithms. The efficiency and robustness of the proposed algorithm using Hellinger divergence are illustrated and compared with the classical mutual information approach through numerical simulations.

Mots clés: Divergences; Blind source sparation; Divergences * Corresponding author; Noisy convolutive mixtures; Blind source separation; Stochastic Optimization; Total variation

Domaines: Mathématiques [math]; Mathématiques [math]; Statistiques [math.ST]; Informatique [cs]; Informatique [cs]; Théorie de l'information [cs.IT]; Informatique [cs]; Traitement du signal et de l'image [eess.SP]

M. El Rhabi, H. Fenniri, A. Keziou, E. Moreau. A robust algorithm for convolutive blind source separation in presence of noise. Signal Processing, 2013, 93 (4), pp.818 - 827. ⟨10.1016/j.sigpro.2012.09.026⟩. ⟨hal-01811756⟩