An active noise control system using online secondary path modeling with reduced auxiliary noise

Abstract

In an active noise control (ANC) system using the filtered-x least mean square (FxLMS) algorithm, an online secondary path modeling method that uses an injected auxiliary noise is often applied. Such a method allows quick and full-band signal-independent modeling. In addition, it is suitable for multisecondary path modeling. Normally, the larger the auxiliary noise, the faster an accurate model can be obtained. However, it increases the residual noise of the ANC system. To mitigate this problem, in this letter, a new online secondary path modeling method is proposed. Rather than fixed, the power of auxiliary noise is varied according to the working status of the ANC system. More specifically, the auxiliary noise is large before the ANC system converges, and becomes small when the system converges. Computer simulations show its effectiveness and robustness.