Multichannel FSLMS algorithm based active headrest

Abstract

The multichannel Filtered-S LMS (FSLMS) algorithm has been used efficiently in nonlinear active noise control (ANC) because of its improved performance and low computational complexity. However, the performance of this algorithm has not yet been evaluated in a real-time ANC system. This correspondence shows the real-time performance of the trigonometric functional expansion based FSLMS algorithm compared to the Filtered-X LMS (FXLMS) algorithm. A multichannel active headrest with one reference, two control sources and two error microphones is used to test the algorithm. Three different primary nonlinear noise cases are studied and it is shown that in all three cases, the FSLMS algorithm is capable of attenuating the primary noise whereas the FXLMS algorithm completely fails to achieve any noise reduction. Insight into the FSLMS algorithm as a suitable noise controller for transformer noise is also presented.