Blind identification and separation of convolutively mixed independent sources

A trispectra method for solving the m-input n-output (n/spl ges/m) wideband blind identification and signal separation problem with unknown number of sources m is presented. The method is universal in the sense that it does not impose any restriction on the probability distribution of the input signals provided that they are non-Gaussian. A criterion, which states a sufficient condition for identification and separation, has been proved. An algorithm is also developed based on the criterion, whose efficiency is verified by the simulations.