Auxiliary Noise Power Scheduling Algorithm for Active Noise Control with On-line Secondary Path Modelling and Sudden Changes

Abstract

In active noise control (ANC), online secondary path modeling can be achieved by adding a small auxiliary noise signal. If the secondary path changes are slow, then this signal can be low, but keeping the stability of the ANC system with sudden secondary path changes requires higher values for this signal, so auxiliary noise power scheduling is required. The proposed algorithm deals well with sudden (and strong) changes, due to the fast convergence of the secondary path model. It is compared with other similar algorithms in the literature. The ability to deal with different physical conditions without changing the algorithms parameters is also compared. The proposed algorithm increases the auxiliary noise to levels close to the acoustic noise when no cancelation is done and reduces it to lower levels when noise cancelation is being performed. In addition variants of the LMS that improve performance are proposed.