Evolutionary variable step-size algorithm for adaptive filtering

A new variable step size scheme namely evolutionary variable step-size algorithm (EVS) for adaptive filtering is proposed. The algorithm is basically a kind of evolutionary method based on evolving the step size of the least-mean-square (LMS) algorithm. The step size candidates are generated by random or deterministic perturbation and then evaluated by calculating a square error measure based on a priori and a posteriori errors. The fittest candidate is selected for subsequent adaptation. The composition of the square error measure is regulated according to the mode of adaptation in order to provide fast converging and tracking capability. The convergence performance is significantly improved and is less sensitive to the eigenvalue spread.