Adaptive Volterra Filters for Active Control of Nonlinear Noise Processes

Abstract

Active noise controls (ANCs) are a challenging application, where there are many considerations such as a secondary path modeling and compensation, fast-changing environment and noise characteristics, interactions between acoustic and electrical domains, placement of microphones and loudspeakers as well as undesired acoustic feedback from the canceling loudspeaker to the reference microphone. ANC algorithms need to identify the secondary path transfer functions between the loudspeaker-to-microphone error sensors and then use the information to guide the direction of control filter coefficient updating. The conventional ANC systems are inefficient in the nonlinear or deterministic noise process, where noise comes from a dynamic system rather than a stochastic, white, or tonal noise process. This paper presents a Volterra filtering for feed-forward ANC which is suitable for the nonlinear controller.