Kalman filter based modeling and constrained H∞ optimal control for active noise cancellation

In this paper, a system identification and a robust control strategy are considered and investigated both theoretically and experimentally for active noise cancellation (ANC) in an acoustic duct. The system identification is based on recursive Kalman filter (RKF) estimation technique, and the control is according to a new constrained H_infin optimal control strategy. A continuous-time model of the acoustic duct is found first, in state space form, using RKF technique. The new H_infin optimal control with closed-loop pole assignment is then applied, which improves the performance of the robust controller and ensures the stability of the closed-loop system under the modeling errors and measurement noise. As a consequence, this methodology results in a noise reduction in a wide bandwidth that is desirable in the area of ANC. Actual hardware experiments using a DSP have been carried out in order to verify the applicability and the performance of the proposed method.