Cross entropy approximation of structured Gaussian covariance matrices
Resource
IEEE Transactions on Signal Processing 56 (7), Part 2: 3362-3367
Journal
IEEE Transactions on Signal Processing
Journal Volume
56
Journal Issue
7 II
Pages
3362-3367
Date Issued
2008
Author(s)
Musicus, Bruce R.
Abstract
We apply two variations of the principle of minimum cross entropy (the Kullback information measure) to fit parameterized probability density models to observed data densities. For an array beamforming problem with P incident narrowband point sources, N > P sensors, and colored noise, both approaches yield eigenvector fitting methods similar to that of the MUSIC algorithm and of the oblique transformation in factor analysis. Furthermore, the corresponding cross entropies (CE) are related to the MDL model order selection criterion. © 2008 IEEE.
Subjects
Array beamforming; Eigenvector methods; Factor analysis; Generalized principle component analysis; Kullback information measure; Minimum cross entropy (CE); Oblique transformation; Stochastic estimation; Structured covariance
Other Subjects
Covariance matrix; Entropy; Mathematical transformations; Wavelet analysis; array beamforming; Colored noises; Cross entropy; eigenvector fitting; Factor analysis (FA); gaussian; Kullback information; model order selection; MUSIC algorithms; Narrow bands; Observed data; Parameterized; point sources; Principle of minimum cross entropy; probability densities; Variations of; Risk assessment
Type
journal article
File(s)![Thumbnail Image]()
Loading...
Name
23.pdf
Size
243.58 KB
Format
Adobe PDF
Checksum
(MD5):2f84c882162dc9ab4d38fdc18100893d