DocumentCode
812293
Title
Maximum flatness spectral modeling
Author
Nadeu, Climent
Author_Institution
ETSE Telecommunicacio, Univ. Politecnica de Catalunya, Barcelona, Spain
Volume
38
Issue
11
fYear
1990
fDate
11/1/1990 12:00:00 AM
Firstpage
2006
Lastpage
2008
Abstract
The maximum entropy method obtains the flattest spectrum consistent with the given autocorrelation values, according to a specific flatness measure. The suitability of the maximum flatness criterion for spectral estimation is discussed. It is concluded that this criterion offers useful insight into the spectral models arising from the optimization approach
Keywords
correlation methods; optimisation; spectral analysis; autocorrelation values; maximum entropy method; maximum flatness criterion; optimization approach; spectral estimation; Acoustic measurements; Acoustic signal processing; Autocorrelation; Constraint optimization; Entropy; Iterative algorithms; Nonlinear equations; Shape; Signal processing algorithms; Speech;
fLanguage
English
Journal_Title
Acoustics, Speech and Signal Processing, IEEE Transactions on
Publisher
ieee
ISSN
0096-3518
Type
jour
DOI
10.1109/29.103103
Filename
103103
Link To Document