Web projektu DAR

Výzkumné centrum
Data - Algoritmy - Rozhodování

Založeno v roce 2005 s podporou MŠMT ČR (projekt 1M0572)

Publikace

Performance analysis of the FastICA algorithm and Cramér-Rao bounds for linear Independent component analysis

Typ:

Článek v odborném periodiku

Autoři publikace:

Tichavský P., Koldovský Z., Oja E.

Název periodika:

IEEE Transactions on Signal Processing

Ročník:

Rok:

2006

Číslo:

Strany:

1189-1203

ISSN:

1053-587X

Kontaktní osoba:

Ing. Petr Tichavský, CSc. (ÚTIA - Oddělení stochastické informatiky)

Klíčová slova:

blind source separation, independent component analysis ICA

příloha1:

13fastica [file]

Anotace:

This paper derives analytic closed form expressions that characterize the performance of the algorithm FastICA, which is one of the most successful algorithms for independent component analysis. Second, the Cramer-Rao lower bound for linear ICA is derived as an algorithm independent limit of the achievable separation quality. The FastICA algorithm is shown to approach this limit in certain scenarios. Extensive computer simulations supporting the theoretical findings are included.