Abstract
The key issue of three-dimensional active noise control (ANC) problems is that global control without spatial spillover is generally difficult, given limited number of discrete sensors. In this paper, a feedforward multi-channel ANC approach is proposed to circumvent this difficulty. In light of the multi-channel model-matching principle, an underdetermined multi-channel inverse filtering (UMIF) system is formulated. With this UMIF system as a design constraint, a cost function is introduced to minimize the noise energy at the control points. The linearly constrained minimum variance (LCMV) beamformer proves effective in broadening the control area. Deployment of control points and the regularization parameter of the LCMV approach are also optimized. To implement the proposed ANC system in a non-freefield environment, sensor interpolation is employed to find the frequency responses between the control points and the loudspeakers in light of the plane wave decomposition. Only a small number of room response measurements are required. The proposed LCMV algorithm has been implemented on a six-element linear loudspeaker array. Simulation results have demonstrated that the propose approach has yielded significant noise reduction performance in a large control area.