Presentation 2011-06-30
Statistical Mechanical Analysis of the Filtered-X LMS Algorithm (II)
Seiji MIYOSHI, Yoshinobu KAJIKAWA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) We analyze the dynamical behaviors (learning curves) of active noise control with Filtered-X LMS algorithm using statistical mechanical method. Direction cosines among coefficient vectors of an adaptive filter, its shifted filters, and an unknown system are treated as the macroscopic variables. We obtain the simultaneous differential equations that describe the dynamical behaviors of the macroscopic variables when the tapped-delay line is long. We analytically solve the equations. Neither the independence assumption nor the small step sizes condition are used. The obtained theory quantitatively agrees with computer simulations, regardless of whether there is noises on error microphone.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Filtered-X LMS algorithm / adaptive filter / active noise control / statistical mechanics of information / thermodynamic limit
Paper # CAS2011-4,VLD2011-11,SIP2011-33,MSS2011-4
Date of Issue

Conference Information
Committee MSS
Conference Date 2011/6/23(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Mathematical Systems Science and its applications(MSS)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Statistical Mechanical Analysis of the Filtered-X LMS Algorithm (II)
Sub Title (in English)
Keyword(1) Filtered-X LMS algorithm
Keyword(2) adaptive filter
Keyword(3) active noise control
Keyword(4) statistical mechanics of information
Keyword(5) thermodynamic limit
1st Author's Name Seiji MIYOSHI
1st Author's Affiliation Faculty of Engineering Science, Kansai University()
2nd Author's Name Yoshinobu KAJIKAWA
2nd Author's Affiliation Faculty of Engineering Science, Kansai University
Date 2011-06-30
Paper # CAS2011-4,VLD2011-11,SIP2011-33,MSS2011-4
Volume (vol) vol.111
Number (no) 105
Page pp.pp.-
#Pages 6
Date of Issue