A FIR Filter-based Online Secondary Path Identification Algorithm for Active Noise Control

The existing conventional online secondary path modeling algorithms have the characteristics that the operation of the ANC controller and the modeling of the secondary path are mutually dependent and interactive. So the mutual undesirable disturbances can badly affect the performance of the system. In the present paper, a new FIR filter-based online secondary path identification algorithm is proposed. Compared with existing algorithms, the method doesn't need feeding extra noise to the secondary source, and is also different from the overall modeling method using the control output. Instead, the facts are utilized that the FIR filters have coefficient vectors equivalent to impulse responses corresponding to transfer functions, and when the coefficients of control filter are updated, the filter coefficient vectors are different at different iteration steps because of the estimation errors. Furthermore, in the method, the modeling of the secondary path is relatively independent of the active noise control system, and the reference signal is used as the input of the system identification. Therefore, the mutual bad disturbances between the operation of the ANC and the identification of the secondary path are eliminated completely, and the complexity of ANC system is greatly reduced. Finally, computer simulations show that the proposed algorithm is feasible and can track the change of the secondary path quickly.