Blind separation of digital sources

This paper presents novel techniques for blind separation of instantaneously mixed digital sources, which are suitable for the case with less sensors than sources. First, a solvability analysis is presented for a general case. A necessary and sufficient condition for recoverability of sources is derived. Then the binary source case is considered, a new deterministic blind separation algorithm is proposed to estimate the mixing matrix and separate all sources efficiently in the noise-free or low noise case. By estimating the probability density function of the mixtures, the deterministic algorithm is extended to deal with the case of high noise. Finally, simulation results are presented to illustrate the validity of the algorithms.