Independent component analysis based on adaptive artificial bee colony

Independent component analysis has been more attractive in the signal processing field. An independent component analysis method based on adaptive artificial bee colony algorithm is proposed in this paper, aiming at the problems of slow convergence and low computational precision in existing independent component analysis methods. The algorithm uses the Givens rotation to reduce the amount of variables to be solved. An adaptive global guidance item is introduced in searching strategy to dynamically adjust optimal guiding role. Simulation results show that the adaptive algorithm can separate the linear combinations of sub-Gaussian and super-Gaussian sources successfully and improve the accuracy of separation.