Title :
ICA of correlated sources mismodeled as uncorrelated: performance analysis
Author :
Lahat, Dana ; Messer, Hagit ; Cardoso, Jean-François
Author_Institution :
Sch. of Electr. Eng., Tel-Aviv Univ., Tel-Aviv, Israel
Abstract :
In this contribution we present closed-form performance analysis for the problem of blind separation of correlated sources when the separation algorithm is ignorant of the source model, and assumes only scalar sources. Our data model is of one scalar source and one multidimensional source. We obtain a closed-form analytical expression for applying joint diagonalization on correlated sources. The obtained expression is different from the optimal expression (in the minimal mean square error (MMSE) sense), obtained by joint block diagonalization. Numerical evaluations quantify the loss of performance due to mismodeling.
Keywords :
blind source separation; independent component analysis; ICA; blind separation algorithm; block diagonalization; closed-form analytical expression; closed-form performance analysis; correlated sources mismodeled; data model; independent component analysis; multidimensional source; optimal expression; source model; Algorithm design and analysis; Blind source separation; Covariance matrix; Data models; Independent component analysis; Mean square error methods; Multidimensional systems; Performance analysis; Performance loss; Telecommunications; Blind Source Separation; Independent component analysis; joint diagonalization; mismodeling;
Conference_Titel :
Statistical Signal Processing, 2009. SSP '09. IEEE/SP 15th Workshop on
Conference_Location :
Cardiff
Print_ISBN :
978-1-4244-2709-3
Electronic_ISBN :
978-1-4244-2711-6
DOI :
10.1109/SSP.2009.5278531
