Online estimation of secondary path in active noise control systems using Generalized Levinson Durbin algorithm

Abstract

A novel online secondary path modeling scheme for active noise control (ANC) systems based on the Generalized Levinson Durbin (GLD) algorithm is proposed in this paper. A short duration zero mean white Gaussian noise is injected into the system using an active loudspeaker and the GLD algorithm is employed to recursively estimate the secondary path. The performance of the proposed scheme is compared with that obtained by filtered-x least mean square (FxLMS) algorithm based ANC system and Eriksson's method in terms of ensemble mean square error and convergence rate. The new scheme is shown to effectively model the secondary path even without prior knowledge of its order.