A Frequency-weighted Leaky FxLMS Algorithm with Application to Feedback Active Noise Control Systems

The trouble caused by instability of the controller and output saturation can destroy noise reduction performance of feedback active noise control (ANC) systems. Based on the traditional leaky filtered-x least mean square (FxLMS) algorithm, a frequency-weighted leaky FxLMS algorithm is proposed in this paper, where the weight factors of the proposed algorithm are characterized in frequency-domain and can be calculated directly by solving a constrained optimization problem. Moreover, it possesses the ability to depress the control filter only at a few frequencies, which are associated with instability of the controller or output saturation, to cut down loss in noise reduction performance. Simulations and experiments are conducted in a duct ANC system to illustrate the feasibility of our algorithm.