Abstract
In this work, expressions are derived for the steady-state excess-mean-square error (EMSE) of the epsilon-normalized sign-error least mean square (NSLMS) adaptive algorithm for both cases of real-and complex-valued data. Moreover, a comparison between the computational load of the epsilon-NSLMS algorithm and the epsilon-normalized least mean square (NLMS) algorithm is also presented. Finally, simulation results to substantiate the theoretical findings are presented.