Blind separation of any source distributions via high-order statistics

An effective technique to perform instantaneous blind source separation by considering a reformulation of the referenced contrast (RC) [A. Adib, E. Moreau, D. Aboutajdine, Blind sources separation by simultaneous generalized referenced contrasts diagonalization, in: ICA, Nara, Japan, 2003, pp. 657–661] is presented. This novel method applies to separate all sources having any distributions including Gaussian ones. Our proposition allows one to assess the capabilities of the RC to break the obviousness that high-order statistics-based methods are restricted to the mixtures allowing at most one Gaussian source. Our design constrains the reference signals choice to avoid the cancellation of the cross-cumulants between this latter and the separator outputs. Simulation studies are presented to support the potential of the approach in terms of source separation.