Presentation 2013-03-14
Optimal Nonlinear Filter for Weak Signal Extraction under Colored Non-Gaussian Noise
Akihisa ICHIKI, Yukihiro TADOKORO,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) We propose the nonlinear filtering method to extract a weak signal in strong noise environment. For efficient filtering, we require the filter characteristics that the signal-to-noise ratio (SNR)of output from the filter becomes large. In this paper, we derive the optimal filter characteristics that exhibits maximum SNR. The obtained filter characteristics is determined by the probability density of the input noise of an arbitrary type, including col- ored and non-Gaussian noise. According to the result, it is concluded that a linear filter is optimal under Gaussian noise. On the other hand, nonlinear filtering is efficient under non-Gaussian noise. Moreover, the obtained filter characteristics achieves the Cramer-Rao lower bound, which is the lower bound of estimation error in unbiased estimator.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Nonlinear filtering / Non-Gaussian noise / Colored noise / Signal-to-noise ratio / Linear estimator
Paper # NLP2012-153
Date of Issue

Conference Information
Committee NLP
Conference Date 2013/3/7(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 Nonlinear Problems (NLP)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Optimal Nonlinear Filter for Weak Signal Extraction under Colored Non-Gaussian Noise
Sub Title (in English)
Keyword(1) Nonlinear filtering
Keyword(2) Non-Gaussian noise
Keyword(3) Colored noise
Keyword(4) Signal-to-noise ratio
Keyword(5) Linear estimator
1st Author's Name Akihisa ICHIKI
1st Author's Affiliation TOYOTA Central R&D Labs., Inc.()
2nd Author's Name Yukihiro TADOKORO
2nd Author's Affiliation TOYOTA Central R&D Labs., Inc.
Date 2013-03-14
Paper # NLP2012-153
Volume (vol) vol.112
Number (no) 487
Page pp.pp.-
#Pages 6
Date of Issue