A Realization of Adaptive Noise Canceler by Multi-Input Multi-Output State Equation Model

概要

論文の詳細を見る
The purpose of this paper is to present a new multiinput multi-output (MIMO) adaptive system model and a general adaptive algorithm for acoustical active noise control (ANC). First we give an extension of an MIMO adaptive filter model described by state equation for ANC, and provide an adaptive algorithm for this model based on the steepest descent method. Then using this model we give a realization of noise canceler for a closed room by considering the propagation path and feedback path of acoustical control output, and show two methods for eliminating the influence of feedback path. Finally, we give some results of computer simulation, which show that the observed convergence and steady-state properties agree well with theoretical predictions. A comparison with a multi-channel tapped-line adaptive system leads to the conclusion that the proposed algorithm has a more benefit of improved convergence rate in the enclosure than classical multiple error filtered-x LMS algorithm.
東海大学の論文