High-order cumulant-based adaptive filter using particle swarm optimization

High-order cumulant-based (HOC) adaptive filter can limit Gauss noise or other noise with symmetric probability distribution function. Current HOC-based adaptive filter commonly adopt gradient search method, but gradient search process is hard to avoid local convergence and complexity. Particle swarm optimization (PSO) is simple and easy to implement, and with no gradient information and other advantages, which can be used to solve many complex problems. Using PSO algorithm to optimize the filter coefficients was proposed as a new method, considering HOC-based coefficients adjustment of adaptive filter as an optimization problem. The simulation results show that using PSO can get higher precision in HOC-based coefficients optimization of adaptive filter. In addition, PSO algorithm is relatively affected little by system jump, which has certain advantage in non-stationary process model.