Variable step-size affine projection sign algorithm using selective input vectors

Affine projection sign algorithm (APSA) is a useful adaptive filter for a highly correlated input signal in the presence of impulsive noise. In this study, a novel variable step-size APSA is proposed using selective input vectors to achieve both fast convergence rate and low steady-state mean-square deviation (MSD) with low computational cost. The selective input vectors and step size are chosen so as to maximize the theoretical MSD difference derived using Price׳s theorem. The simulation results show that the proposed algorithm has the fastest convergence rate and lowest steady-state MSD when compared with recent variable step-size APSAs. Moreover, it effectively reduces computational cost.