Abstract

After analysis of the cyclostationary signal characteristics and research of the traditional blind source separate algorithm, a new variable step size algorithm for blind source separation (BSS) is proposed in this paper. This algorithm takes the mutual information as the objective function, and uses the adaptive variable step size kurtosis, which effectively solves the contradiction between convergence speed and steady-state error. Simulation results show that the new algorithm can achieve better separation result for the cyclostationary source signals, both the convergence speed is faster and its smooth performance is better than traditional gradient algorithm.

Paper Details