Efficient variant of algorithm FastICA for independent component analysis attaining the Cramer-Rao lower bound.
Petr Tichavsky, Zbynek Koldovski and Erkki Oja
IEEE Transactions on Neural Networks Volume 17, Number 5, pp. 1265-1277, 2006.

Abstract

An efficient variant of the algorithm FastICA for independent component analysis is derived, that attains the theoretically minimal Cramer-Rao lower bound. Thus the algorithm is asymptotically as efficient as possible.

PDF - Requires Adobe Acrobat Reader or other PDF viewer.

[Edit]