Presentation 2007-12-07
An exact direct method for short-time multiple sinusoidal parameter estimation by means of finite Fourier transform
Shigeru ANDO,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In previous papers, we have proposed a novel method for estimating directly and exactly the frequency, amplitude, and phase of a real sinusoid from their mixture. This method is based on weighted finite integrals of the characteristic differential equation of a sinusoid, and arrives at algebraic equations of its frequency and integral boundary terms. The integral boundary terms can be treated as unknown variables or eliminated by three window functions. Thanks to the Fourier transform and the windowing operation, multiple sinusoids can be decomposed and estimated mostly independently and accurately. Contrarily in this paper, we extend the method rigorously into multiple sinusoidal conditions using higher order characteristic differential equations of mixture of two sinusoids. We apply this method to remove the interference between closely adjacent sinusoids in the original algorithm.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) sinusoidal parameter estimation / differential equation / weighted integral method
Paper # EA2007-92
Date of Issue

Conference Information
Committee EA
Conference Date 2007/11/30(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 Engineering Acoustics (EA)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) An exact direct method for short-time multiple sinusoidal parameter estimation by means of finite Fourier transform
Sub Title (in English)
Keyword(1) sinusoidal parameter estimation
Keyword(2) differential equation
Keyword(3) weighted integral method
1st Author's Name Shigeru ANDO
1st Author's Affiliation Department of Information Physics and Computing, University of Tokyo()
Date 2007-12-07
Paper # EA2007-92
Volume (vol) vol.107
Number (no) 370
Page pp.pp.-
#Pages 6
Date of Issue