Článek v odborném periodiku

Tichavský P., Koldovský Z., Yeredor A., Gómez-Herrero G., Do

IEEE Transactions on Neural Networks

2008

3 (2008)

421-430

1045-9227

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

blind source separation, independent component analysis

Blind inversion of a linear and instantaneous mixture of source signals is a problem often encountered in many signal processing applications. Efficient FastICA (EFICA) offers an asymptotically optimal solution to this problem when all of the sources obey a generalized Gaussian distribution, at most one of them is Gaussian, and each is independent and identically distributed in time. Likewise, Weights-Adjusted Second Order Blind Identification (WASOBI) is asymptotically optimal when all the sources are Gaussian and can be modeled as Autoregressive (AR) processes with distinct spectra. Nevertheless, real-life mixtures are likely to contain both Gaussian AR and non-Gaussian iid sources, rendering WASOBI and EFICA severely sub-optimal. In this paper we propose a novel scheme for combining the strengths of EFICA and WASOBI in order to deal with such hybrid mixtures. Simulations show that our approach outperforms competing algorithms designed for separating similar mixtures.