A neural architecture for blind source separation
Tapia, Ernesto ;  Rojas, Raúl ;  Universität <Berlin, Freie Universität> / Fachbereich Mathematik und Informatik

Main titleA neural architecture for blind source separation
AuthorTapia, Ernesto
AuthorRojas, Raúl
InstitutionUniversität <Berlin, Freie Universität> / Fachbereich Mathematik und Informatik
No. of Pages8 S.
Series Freie Universität Berlin, Fachbereich Mathematik und Informatik : Ser. B, Informatik ; [20]06,04
Classification (DDC)003 Systems
AbstractA neural architecture based on linear predictability is used to separate linear mixtures of signals. The architecture is divided in two parameterers groups, one modeling the linear mixture of signals and the other computing the linear predictions of the reconstructed signals. The network weights correspond to the mixing matrices and coefficients of the linear predictions, while the values computed by the network units correspond to the predicted and reconstructed signal values. A quadratic error is iteratively minimized to approximate the mixing matrix and to maximize the linear predictability. Experiments with toy and acoustic signals show the feasibility of the architecture.
If your browser can't open the file, please download the file first and then open it
FU DepartmentDepartment of Mathematics and Computer Science
Other affiliation(s)Institut für Informatik
Year of publication2006
Type of documentMaps
Terms of use/Rights Nutzungsbedingungen
Created at2009-10-13 : 01:12:03
Last changed2015-01-22 : 04:42:24
Static URLhttp://edocs.fu-berlin.de/docs/receive/FUDOCS_document_000000003868