Variable Matrix-Type Step-Size Affine Projection Sign Algorithm for System Identification in the Presence of Impulsive Noise

This paper presents a novel variable matrix-type step-size affine projection sign algorithm (VMSS-APSA) characterized by robustness against impulsive noise. To mathematically derive step size, VMSS-APSA utilizes mean-square deviation (MSD) for the modified version of original APSA. Accurately establishing MSD APSA is impossible. Therefore, proposed derives upper bound using L1-norm measurement The optimal size calculated at each iteration minimizing MSD, thereby improving filter performance in terms convergence rate and steady-state estimation error. Because cost function was designed to maintain form similar APSA, they have symmetric characteristics. Simulation results demonstrate that improves system-identification scenario presence