Comparison of two proposed methods in adaptive noise canceling

In this paper, an adjusted step size LMS (least mean squares) algorithm is proposed for possible improvements in the performance of adaptive FIR filters in nonstationary environments. Nonstationary signals means that the statistical properties of the noise changes in time such as the high frequency channel time variations. We propose two methods to improve the adaptive filtering characteristics. One of the proposed methods is based on the signal to noise ratio value for adjusting the adaptive step size parameter. The other method is based on a self-correcting approach for a fast processing time. Both methods are compared to each other and to the classic approach used in the adaptive filtering area.