Bayesian source separation of linear-quadratic and linear mixtures through a MCMC method - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2009

Bayesian source separation of linear-quadratic and linear mixtures through a MCMC method

Résumé

In this work, we deal with source separation of linear - quadratic (LQ) and linear mixtures. By relying on a Bayesian approach, the developed method allows one to take into account prior informations such as the non-negativity and the temporal structure of the sources. Concerning the inference scheme, the implementation of a Gibbs' sampler equipped with latent variables simplifies the sampling steps. The obtained results confirm the effectiveness of the new proposal and indicate that it may be particularly useful in situations where classical ICA-based solutions fail to separate the sources.
Fichier principal
Vignette du fichier
final20.pdf (235.04 Ko) Télécharger le fichier
Origine : Fichiers éditeurs autorisés sur une archive ouverte
Loading...

Dates et versions

hal-00424171 , version 1 (14-10-2009)

Identifiants

  • HAL Id : hal-00424171 , version 1

Citer

Leonardo Tomazeli Duarte, Christian Jutten, Saïd Moussaoui. Bayesian source separation of linear-quadratic and linear mixtures through a MCMC method. MLSP 2009 - IEEE 19th International Workshop on Machine Learning for Signal Processing, Sep 2009, Grenoble, France. 6 p. ⟨hal-00424171⟩
315 Consultations
154 Téléchargements

Partager

Gmail Facebook X LinkedIn More