Improvement of the computational load of nonlinear ANC systems using Volterra filters

This paper proposes an approach to efficiently mitigate the perceptible distortion in the secondary sound field, which contributes to the degradation of the active noise control accuracy used for spatial sound field control system. Our proposed system uses a Volterra inverse filter to reduce these distortions. In this filtering for active noise control, the primary challenge lies in the high computational load associated with the processing related to the Volterra kernel. We propose a new method to enhance the sparsity of the Volterra kernel, which contributes to the reduction of computational load, by focusing on the physical mechanisms behind the generation of these distortions Simulation results show that the proposed method suppresses these distortions and improves noise reduction by approximately 10 dB in specific frequency bands.