In this work we introduce a new ICA algorithm, designed to take available prior information on the sources into account. This prior information is included in the form of a 'weak' constraint and is exploited simultaneously with independence in order to separate the sources. This new algorithm outperforms classical ICA algorithms in the case of low SNR. Furthermore, inclusion of additional information about the sources may help enforcing the ordering of the extracted components according to their significance.
Optimizing ICA using generic knowledge of the sources / VALENTE, Giancarlo; De Martino, F.; Balsi, M.; Formisano, E.. - STAMPA. - II:(2005), pp. 27-34. (Intervento presentato al convegno 1st International Workshop on Biosignal Processing and Classification (BPC 2005) tenutosi a Lausanne; Switzerland nel 14-17/9/2005) [10.1109/RME.2005.1542974].
Optimizing ICA using generic knowledge of the sources
VALENTE, Giancarlo;F. De Martino;M. Balsi;
2005
Abstract
In this work we introduce a new ICA algorithm, designed to take available prior information on the sources into account. This prior information is included in the form of a 'weak' constraint and is exploited simultaneously with independence in order to separate the sources. This new algorithm outperforms classical ICA algorithms in the case of low SNR. Furthermore, inclusion of additional information about the sources may help enforcing the ordering of the extracted components according to their significance.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
