An improved variable step size LMS algorithm

Abstract

The traditional variable step size least mean square (LMS) algorithm has many weakness, such as poor convergence speed and susceptible to noise interference. In order to improve the performance of the algorithm, an improved variable step size LMS algorithm is presented, and it is applied to the noise cancelling system of communication. By using the hyperbolic tangent function, the relationship between step size and error signal is established. In this algorithm, the step size factor is adjusted by the absolute value of the product of the current and former errors. The algorithm also introduces the disturbance of the absolute estimation error to update the tapping vector of the self-adaptive filter. The simulation results show that the proposed algorithm has a faster convergence speed than traditional LMS algorithm and CTanh-LMS algorithm. And it obtains a good results in the noise cancelling system of communication.