Steady-state and tracking analysis of the SSLMS algorithm

This paper presents expressions for the steady-state mean-square error (MSE), the optimum stepsize, and the corresponding minimum value of the tracking MSE of the sign-sign least mean square (SSLMS) algorithm for the case of real-valued data. Then, simulation results are presented to support our analytical findings and found to corroborate them very well. Also, the performance of the SSLMS algorithm is compared to that of the LMS algorithm in the case where the noise statistics are uniformly distributed