Maximum likelihood blind source separation in Gaussian noise
J. Miquez, Luis Castedo · 2003
This paper presents a new maximum likelihood (ML) approach to the separation of convolutive mixtures of unobserved sources in the presence of additive temporally white Gaussian noise (ATWGN). The proposed method proceeds in two steps. First, the ML estimate of the mixing system is computed and, afterwards, this estimate is employed to obtain the ML estimates of the sources. The proposed algorithms rely on the knowledge of the sources probability density function (p.d.f.) and the noise second order statistics. These are fairly realistic assumptions in applications such as digital communications.