Adjoint nonlinear active noise control algorithm for virtual microphone

Abstract

Controlling nonlinear noise processes at a virtual location involves higher computational complexity compared to traditional active noise control (ANC) at a physical sensor. In an attempt to reduce the computational burden, a filtered-error LMS based ANC algorithm is proposed in this paper for controlling a nonlinear noise process at a virtual location. The filtered-error based algorithm using the adjoint of the secondary path is used to develop both linear and nonlinear ANC controllers with the later based on the functional link artificial neural network. A computer simulation and real-time experimental study with detailed computational complexity analysis are presented to support the proposed algorithm. © 2011 Elsevier Ltd. All rights reserved.