Adaptive Step-size Normalised Least Mean Square Algorithm for Spline Adaptive Filtering

This paper proposes an adaptive step-size normalised least mean square algorithm based on spline adaptive filtering. An adaptive step-size method is used by exploiting an estimation of autocorrelation between previous and present error estimation in order to control step-size update. The proposed algorithm combines the adaptive step-size approach and normalised least mean square scheme for achieving fast convergence rate on the filter weight and control point adaptation. Simulation results demonstrate that the proposed algorithm exhibits more robust performance compared with the conventional spline adaptive filtering algorithms.