Blind Source Separation of Post-Nonlinear Mixtures Using Evolutionary Computation and a Local Search Procedure

In this work, we propose a new method for source separation of post-nonlinear mixtures that brings together evolutionary-based global search, entropy estimation via order-statistics and a local search step based on the FastICA algorithm. The rationale of the proposal is to attempt to obtain eficient and precise solutions using with parsimony the available computational resources. The new proposal was tested in different scenarios, and, in all cases, we attempted to establish grounds for comparison with an alternative approach whose optimization step does not include the local (memetic) search stage. Simulation results indicate that a good tradeoff between performance and computational cost was indeed reached.