Blind Source Separation Based on Adaptive Artificial Bee Colony Optimization and Kurtosis

Abstract

A novel technique is proposed for selecting iterative updates and step sizes based on adaptive function values to compensate for the slow convergence rate of artificial bee colony optimization (ABCO). On this basis, a blind source separation (BSS) algorithm is proposed based on adaptive ABCO and kurtosis, which does not impose any hypothetical requirements on the source signal. By using kurtosis as the objective function, the algorithm can separate signals that follow any distribution. BSS results from various test distributions demonstrate the superior performance of the proposed algorithm compared to conventional methods.