DocumentCode :
3069176
Title :
Efficient maximum entropy spectral estimation using non-multiplication methods
Author :
Koh, Taiho ; Powers, Edward J.
Author_Institution :
The University of Texas, Austin, Texas
Volume :
9
fYear :
1984
fDate :
30742
Firstpage :
535
Lastpage :
538
Abstract :
New efficient autocorrelation and lattice methods of MEM spectral estimation are presented. In these methods, the "covariance property" of Gaussian distributions is exploited such that autocorrelation samples or lattice parameters can be computed with addition operations only. Also the performance of the proposed lattice method is studied through simulations and its application to the problem of spectral analysis of the vibrational modes of damaged and undamaged aircraft wings.
Keywords :
Analytical models; Autocorrelation; Computational modeling; Distributed computing; Entropy; Equations; Gaussian distribution; Lattices; Random variables; Spectral analysis;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Acoustics, Speech, and Signal Processing, IEEE International Conference on ICASSP '84.
Type :
conf
DOI :
10.1109/ICASSP.1984.1172311
Filename :
1172311
Link To Document :
بازگشت