Abstract
Single-channel feedforward control is a commonly used approach for active noise control (ANC). In this paper, a time-domain underdetermined multichannel inverse filtering technique is proposed for the feedforward active control (ANC) of duct noise. In the commonly used filtered-x least-mean-square (FXLMS) algorithm, the feedforward control problem is formulated as an overdetermined inverse filtering problem which generally results in non-zero residual noise. By introducing multiple secondary sources, the problem can be reformulated into an underdetermined system, which admits infinite number of exact solutions with zero residual noise. Linearly constrained minimum variance (LCMV) method is employed in the controller design. However, as a major shortcoming of the time-domain approach, the length of the filters tends to be too long to admit digital signal processor (DSP) implementation. To address the problem, the least absolute shrinkage and selection operator (LASSO) algorithm, is exploited to effectively reduce the controller orders. A two-channel system is implemented to suppress the fan noise of an air-cleaner. Simulation and experiment results have demonstrated that the proposed approach has achieved significantly higher noise reduction than the conventional FXLMS algorithm.