Optimization of a Geman-McClure like criterion for sparse signal deconvolution - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2015

Optimization of a Geman-McClure like criterion for sparse signal deconvolution

Résumé

This paper deals with the problem of recovering a sparse unknown signal from a set of observations. The latter are obtained by convolution of the original signal and corruption with additive noise. We tackle the problem by minimizing a least-squares fit criterion penalized by a Geman-McClure like potential. The resulting criterion is a rational function, which makes it possible to formulate its minimization as a generalized problem of moments for which a hierarchy of semidefinite programming relaxations can be proposed. These convex relaxations yield a monotone sequence of values which converges to the global optimum. To overcome the computational limitations due to the large number of involved variables, a stochastic block-coordinate descent method is proposed. The algorithm has been implemented and shows promising results
Fichier principal
Vignette du fichier
camsap15.pdf (107.53 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-01263352 , version 1 (27-01-2016)
hal-01263352 , version 2 (15-02-2016)

Identifiants

  • HAL Id : hal-01263352 , version 1

Citer

Marc Castella, Jean-Christophe Pesquet. Optimization of a Geman-McClure like criterion for sparse signal deconvolution. CAMSAP 2015 : 6th International Workshop on Computational Advances in Multi-Sensor Adaptive Processing, Dec 2015, Cancun, Mexico. pp.317 - 320. ⟨hal-01263352v1⟩
181 Consultations
600 Téléchargements

Partager

Gmail Mastodon Facebook X LinkedIn More