A method to improve the performance of the LMS adaptive filter in processing signals of non-zero mean with sharp changes

摘要

A method is proposed to improve the performance of the conventional least-mean-square (LMS) adaptive filter in processing signals of nonzero mean with abrupt changes. It is theoretically determined that, with the aid of a local-mean estimator (LME), the working condition and thus the performance of the adaptive filter can be improved. The implementation of the LME is discussed and a varying window average filter is introduced. The improved performance of the proposed method compared to the direct use of the conventional LMS has been demonstrated by computer simulations.