Real-time DSP Implementation of Active Noise Control Using Robust Adaptive Filter

Abstract

In this paper we investigate an approach to reduce undesired level of acoustic noise by taking advantage of the additivity property of sound propagation. By introducing another controlled sound source, an active noise control (ANC) system generates a sound wave with its phase just opposite of the original noise to be canceled in a specific area (quiet zone). In this research adaptive H∞ method is employed in order effectively cancel out the primary noise at the error sensor which estimates the primary noise. The adaptive algorithm is used because the propagation paths for the primary noise are often very complex and timevarying, while the H∞ theory is proposed to cope with uncertain dynamics of the path. The proposed active noise control is experimentally demonstrated using Digital Signal Processor (DSP) to cancel acoustic noise in a specific area within an enclosure as well as in an open space.