Prewhitening of mixing matrices: impact on the contraction of their probability mass priors
Victor A. N. Barroso, P. Lopes, T. Patrao · 2003
The authors consider a linear mixture of n sources, x(t) = As(t) + w(t), where x(t) denotes the observations, A /spl isin/ /spl Ropf//sup n/spl times/n/ is the mixing matrix, s(t) contains the n source signals, and w(t) models observation noise. The sources are assumed to be wide-sense stationary, zero-mean, R/sub s/ = E{s(t)s(t)/sup T/} = I/sub n/, and A is invertible.