Lorentz adaptive prior based FastICA BSS

摘要

It was well known that an appropriate selection of the nonlinear contrast function was the key for achieving successful blind source separation through FastICA algorithm. In this study, it was first suggested that the estimation of the nonlinear contrast function could be implemented by the way of adjusting the prior densities of the sources from the data; it was then discussed how to apply the Lorentz prior model into the FastICA to estimate the contrast function adaptively; it was finally testified in the experiment that the Lorentz adaptive prior based FastICA is more effective than the conventional one with fixed prior.