Feedback supervising feedforward Minimum-Variance Control of acoustic noise

Minimum-Variance Control is one of the dominating strategies for actively suppressing acoustic noise. Feedforward structure is generally preferred if a reference signal is available. Otherwise, feedback structure is the interim solution. The purpose of this paper is to design an optimal combined control-weighted minimum variance noise control system. In the proposed approach feedforward is the primary control strategy. Feedback control is added to the already operating feedforward. It aims at controlling the residual noise of feedforward control. However, the main purpose of closing the external loop is to benefit from properties of feedback control. Appropriate controllers can be designed in the time domain, frequency domain or transform domain. The latter approach, based on polynomial operations, including Diophantine equations has been chosen. Correctly applied control weighting for both techniques guarantees stable operation for a non-minimum phase plant. Performance of the combined system is compared to performance of the individually operating feedforward system. Different time dependences are considered. Theoretical analysis is supported by simulation experiments based on data obtained from a real-world active headrest system.