Blind Source Separation based on Refracted and Elite Opposition-based Learning Firefly Algorithm

To overcome the shortcomings of many parameters, slow convergence speed, and poor separation accuracy in applying traditional swarm intelligence algorithm in blind source separation, a Firefly Algorithm integrating Refracted and Elite Opposition-based learning is proposed. The proposed algorithm enhances the global search ability of the original algorithm, improves the population diversity, and accelerates the convergence speed. The simulation results show that in the Blind Source Separation, the similarity coefficients of the separated signals obtained by the proposed algorithm are all above 0.99.