Presentation 1997/6/26
A Criterion for Signal Detection Based on Statistics of Maximum Likelihood
Fumiki Kitagawa, Masakiyo Suzuki, Nobuo Nagai,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This paper presents a new criterion for detecting the number of superimposed signals that can be applied to the case of small data samples available, while conventional methods based on infomation criteria have defficulties. The propposed method is based on the statistical property of maximum likelihood for the observational model supposed a signal number. This approach can be used for super resolution techniques for bearing estimation or harmonic retrieval, such as MUSIC and Maximum Likelihood method that provide high resolution with small samples. At last, simulation results explain the property and validity of the proposed method.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) signal number detection / MDL / Maximum Likelihood method / bearing estimate / harmonic retrieval
Paper # CAS97-1
Date of Issue

Conference Information
Committee VLD
Conference Date 1997/6/26(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 VLSI Design Technologies (VLD)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) A Criterion for Signal Detection Based on Statistics of Maximum Likelihood
Sub Title (in English)
Keyword(1) signal number detection
Keyword(2) MDL
Keyword(3) Maximum Likelihood method
Keyword(4) bearing estimate
Keyword(5) harmonic retrieval
1st Author's Name Fumiki Kitagawa
1st Author's Affiliation Reserch Institute for Electronic Science, Hokkaido Univ.()
2nd Author's Name Masakiyo Suzuki
2nd Author's Affiliation Reserch Institute for Electronic Science, Hokkaido Univ.
3rd Author's Name Nobuo Nagai
3rd Author's Affiliation Reserch Institute for Electronic Science, Hokkaido Univ.
Date 1997/6/26
Paper # CAS97-1
Volume (vol) vol.97
Number (no) 138
Page pp.pp.-
#Pages 8
Date of Issue