Amplitude Banded Sato Algorithm for Blind Channel Equalization

In this paper we propose a novel non-linear blind adaptive algorithm called the Amplitude Banded Sato (ABSato) algorithm for equalization of communication channels. The ABSato algorithm is derived as a modified version of the Amplitude Banded Least Mean Square (ABLMS) algorithm addressed by Shimamura et al. recently. The capability of nonlinear classification the ABLMS algorithm inherently possesses is kept in the ABSato algorithm, resulting in an improvement of equalization performance. Mean square error as well as bit error rate performances are investigated on simple communication channel models. Observation of simulation results show that the ABSato algorithm provides better performance than the standard Sato algorithm on all the communication channel models.