Projekte pro Jahr
Abstract
This paper studies the problem of blind separation of convolutively mixed source signals on the basis of the joint diagonalization (JD) of power spectral density matrices (PSDMs) observed at the output of the separation system. Firstly, a general framework of JDbased blind source separation (BSS) is reviewed and summarized. Special emphasis is put on the separability conditions of sources and mixing system. Secondly, the JDbased BSS is generalized to the separation of convolutive mixtures. The definition of a time and frequency dependent characteristic matrix of sources allows us to state the conditions under which the separation of convolutive mixtures is possible. Lastly, a frequencydomain approach is proposed for convolutive mixture separation. The proposed approach exploits objective functions based on a set of PSDMs. These objective functions are defined in the frequency domain, but are jointly optimized with respect to the timedomain coefficients of the unmixing system. The local permutation ambiguity problems, which are inherent to most frequencydomain approaches, are effectively avoided with the proposed algorithm. Simulation results show that the proposed algorithm is valid for the separation of both simulated and realword recorded convolutive mixtures.
Originalsprache  Englisch 

Zeitschrift  Signal Processing 
Jahrgang  88 
Ausgabenummer  8 
Seiten (von  bis)  19902007 
Seitenumfang  18 
ISSN  01651684 
DOIs  
Publikationsstatus  Veröffentlicht  01.08.2008 
Fingerprint
Untersuchen Sie die Forschungsthemen von „Blind source separation for convolutive mixtures based on the joint diagonalization of power spectral density matrices“. Zusammen bilden sie einen einzigartigen Fingerprint.Projekte
 1 Abgeschlossen

Blinde Separierung akustischer Quellensignale in reflexionsbehafteter Umgebung
Mertins, A. & Kollmeier, B.
01.01.04 → 31.12.08
Projekt: DFGProjekte › DFG Einzelförderungen