Multichannel narrowband active noise control system with a frequency estimator based on DFT coefficients

Abstract

The narrowband active noise control system (NANC) is an active control system that cancels periodic noise. The ability of the system to suppress noise is affected by the accuracy of the frequency estimator. The frequency estimator in the existing narrowband system has difficulty balancing the estimation accuracy and computational complexity. In this paper, a filtered-x discrete Fourier transform coefficients estimator least mean square (Fx-DFTCE-LMS) algorithm is proposed to enhance the noise reduction capability of multichannel NANC system. This system can handle any degree of frequency mismatch without increasing the complexity. Compared with most existing estimation methods, the proposed method does not need set initial frequency or estimator step size. Numerous simulations demonstrate that the proposed NANC system can track multiple frequencies more quickly and accurately, and effectively reduce noise.