A Multi-population Particle Swarm Optimizer and its Application to Blind Multichannel Estimation

Abstract

In this paper, a multi-population particle swarm optimizer based on Lotka-Volterra competition equation is first proposed. The cooperative coevolution in the field of is involved into original particle swarm optimizer, and populations size is adjusted adaptively based on multi- population Lotka-Volterra competition equation. Then, the algorithm is applied to blind multichannel estimation by optimizing an error function for the outputs of a multichannel system. The experiment results demonstrate that the proposed algorithm is superior to original particle swarm optimization algorithm, and is effective to blind multichannel estimation.