Blind Source Separation Algorithm Based on Improved Particle Swarm Optimization Under Noisy Condition

In view of the shortcomings of particle swarm optimization algorithm, it is easy to fall into early maturation and low separation accuracy. On the basis of particle swarm optimization algorithm, an improved particle swarm optimization algorithm based on adaptive adjustment inertia weight is proposed. Using the correlation coefficient matrix as the evaluation criterion of separation performance, the improved particle swarm optimization algorithm, the basic particle swarm optimization and the fast fixed point algorithm(FastICA) are compared under the noisy condition. The simulation results show that the proposed algorithm is effective to improve separation performance and has resistance to noise to some extent.