Séparation de sources retardées et paramétriques

Abstract : This paper addresses delayed (or anechoic) source separation in the case of parameterized deterministic sources, possibly highly correlated. An alternating least squares scheme is proposed to estimate the source parameters, the mixing coefficients and the delays. The delays are estimated with an effective sparse deconvolution algorithm designed for delayed source separation and allowing continuous delay estimation. Numerical simulations demonstrate the effectiveness of the proposed algorithm compared to state-of-the-art methods for highly correlated Gaussian sources.
Document type :
Conference papers
Complete list of metadatas

Cited literature [20 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-01872698
Contributor : Vincent Mazet <>
Submitted on : Friday, October 26, 2018 - 5:34:34 PM
Last modification on : Thursday, November 8, 2018 - 1:12:52 AM
Long-term archiving on : Sunday, January 27, 2019 - 6:14:47 PM

File

Gretsi_francais.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01872698, version 1

Citation

Hassan Mortada, Mazet Vincent, Charles Soussen, Christophe Collet. Séparation de sources retardées et paramétriques. 26eme Colloque GRETSI Traitement du Signal & des Images (GRETSI 2017), Sep 2017, Juan-les-Pins, France. ⟨hal-01872698⟩

Share

Metrics

Record views

43

Files downloads

21