Individual control of narrowband noise with disturbance

Abstract

The disturbance picked up by error sensors may degrade the performance of narrowband active noise cancellation (NANC) systems, resulting in slow convergence of algorithm and low noise reduction. This paper proposes modified algorithms to enhance the transient and steady-state performance. A delayless filter bank is designed to separate the residual noise into parallel channels to update the corresponding adaptive filters using individual error signals. The improved performance is verified by computer simulations.