Equivariant adaptive source separation
IEEE Transactions on Signal Processing
Flexible Independent Component Analysis
Journal of VLSI Signal Processing Systems
Hi-index | 0.00 |
We present a new learning algorithm for the blind separation ofindependent source signals having non-zero skewness (the 3rd-order cumulant)(the source signals have non-symmetric probability distribution.), fromtheir linear mixtures. It is shown that for a class of source signals whoseprobability distribution functions is not symmetric, a simple adaptivelearning algorithm using quadratic function (f(x)=x^2) is veryefficient for blind source separation task. It is proved that all stableequilibria of the proposed learning algorithm are desirable solutions.Extensive computer simulation experiments confirmed the validity of theproposed algorithm.