A frequency band constrained filtered-x least mean square algorithm for feedback active control systems.

Abstract

The generalized leaky filtered-x least mean square (GLFxLMS) algorithm can reduce the noise amplification caused by the waterbed effect in feedback active control systems effectively; however, it suffers from a high computation complexity. Hence, a frequency band constrained filtered-x least mean square algorithm is proposed to reduce the computation complexity of the GLFxLMS algorithm by replacing the penalty term containing a symmetric Toeplitz matrix in the cost function with the mean square of a penalty signal. The simulation results based on the measured transfer functions of an active headrest system show that the proposed algorithm has the same performance as the GLFxLMS algorithm, but with much lower computation complexity.