NEW DIRECT ADAPTIVE ALGORITHM FOR MULTICHANNEL ACTIVE NOISECONTROL AND SOUND REPRODUCTION (WedAmOR5)

Author(s) : Masaki Kamata (Keio University, Japan) Yuhsuke Ohta (Keio University, Japan) Akira Sano (Keio University, Japan)

Abstract : A new direct adaptive algorithm is proposed for multichannel active noise control (ANC) and sound reproduction (SR) when primary and secondary path dynamics are all uncertain and changeable. To attenuate the canceling errors or reproduction errors, two kinds of virtual error vectors are introduced and are forced into zero by adjusting three adaptive FIR matrix filters in an on-line manner. It is shown that the convergence of the canceling or reproduction errors to zero can be attained at the objective points. The proposed algorithm can tune an inverse controller matrix directly without need of explicit identification of the secondary path channels, and requires neither any dither signals nor the PE property of the source signals for the identifiability.

Menu