Separation of non stationary sources; achievable performance
J.-F. Cardoso · 2002
We consider the blind separation of an instantaneous mixture of non-stationary source signals, possibly normally distributed. The asymptotic Cramer-Rao bound is exhibited in the case of known source distributions: it reveals how non-stationarity and non-Gaussianity jointly governs the achievable performance via an index of non-stationarity and an index of non-Gaussianity.