Contribution of Statistical Tests to Sparseness-Based Blind Source Separation - Archive ouverte HAL Access content directly
Journal Articles Eurasip Journal on Applied Signal Processing Year : 2012

Contribution of Statistical Tests to Sparseness-Based Blind Source Separation

(1, 2) , (3, 1) , (4, 1)
1
2
3
4

Abstract

We address the problem of blind source separation in the underdetermined mixture case. Two statistical tests are proposed to reduce the number of empirical parameters involved in standard sparseness-based underdetermined blind source separation (UBSS) methods. The first test performs multisource selection of the suitable time-frequency points for source recovery and is full automatic. The second one is dedicated to autosource selection for mixing matrix estimation and requires fixing two parameters only, regardless of the instrumented SNRs. We experimentally show that the use of these tests incurs no performance loss and even improves the performance of standard weak-sparseness UBSS approaches.
Fichier principal
Vignette du fichier
RUBSS_revue_JASPvAvril2012_whl.pdf (526.24 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-00739565 , version 1 (08-10-2012)

Identifiers

Cite

Si Mohamed Aziz Sbai, Abdeldjalil Aissa El Bey, Dominique Pastor. Contribution of Statistical Tests to Sparseness-Based Blind Source Separation. Eurasip Journal on Applied Signal Processing, 2012, 2012 (169), ⟨10.1186/1687-6180-2012-169⟩. ⟨hal-00739565⟩
169 View
260 Download

Altmetric

Share

Gmail Facebook Twitter LinkedIn More