Cross Entropy Approximation of Structured Gaussian Covariance Matrices

@article{Liou2008CrossEA,
  title={Cross Entropy Approximation of Structured Gaussian Covariance Matrices},
  author={Cheng-Yuan Liou and Bruce R. Musicus},
  journal={IEEE Transactions on Signal Processing},
  year={2008},
  volume={56},
  pages={3362-3367}
}
  • C. Liou, B. Musicus
  • Published 1 July 2008
  • Computer Science
  • IEEE Transactions on Signal Processing
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, 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… 
View on IEEE
ntur.lib.ntu.edu.tw
9 Citations
Background Citations
2
Methods Citations
2

9 Citations

Citation Type

Citation Type

On maximum likelihood estimation in factor analysis - An algebraic derivation
Atmospheric effects estimation on SAR images using a pure statistical method
The values of the unwrapped phase produced by interferometric pairs can be parameterized for phase components, such as height, atmospheric path delay and deformation term, and estimated through…
Manifold Construction by Local Neighborhood Preservation
TLDR
This work presents a neighborhood preservation method to construct the latent manifold that preserves the relative Euclidean distances among neighboring data points and its performance in preserving the local relationships is promising when compared with the methods, LLE and Isomap.
A Novel Method for Manifold Construction
This paper presents a distance invariance method to construct the low dimension manifold that preserves the neighborhood topological relations among data patterns. This manifold can display close…
A Comprehensive Review of Factor Analysis
Factor analysis is one of the dimension reduction methods. It is widely used in any areas even though its history is very long. Researchers and engineers are applying factor analysis much as thy…
ABehavioralModelofDrivers’MeanSpeed InfluencedbyWeather Conditions, Road Geometry, and Driver Characteristics Using a Driving Simulator Study
TLDR
It can be concluded that the existence of a curve on the road encourages drivers to slow down, even young drivers, and the parts of the simulated road with a horizontal curve act as a speed reduction tool for drivers.
A Behavioral Model of Drivers’ Mean Speed Influenced by Weather Conditions, Road Geometry, and Driver Characteristics Using a Driving Simulator Study
TLDR
It can be concluded that the existence of a curve on the road encourages drivers to slow down, even young drivers, and it is noteworthy that the parts of the simulated road with a horizontal curve act as a speed reduction tool for drivers.
Information in hierarchical modular system
TLDR
This paper shows how to generate the hierarchical modular system by using the principle of extreme physical information and discusses the equal temperament music scale that has many shared properties with the HMS.
Comprehensive assessment of reliability and validity for the clinical cases in simulated community pharmacy
TLDR
The internal consistency, sample adequacy, factor loading, test for an exact fit, and fit measurements should be ensured for the clinical cases included in OSCE to ensure the accurate assessment of the clinical competence of the student in a simulated community pharmacy.

References

SHOWING 1-10 OF 23 REFERENCES
A separable cross-entropy approach to power spectral estimation
TLDR
An approach to power spectrum estimation that is based on a separable cross-entropy modeling procedure that is a generalized version of Capon's maximum-likelihood method and a spectral estimation algorithm that is similar to classical windowing methods is presented.
Effects of estimated noise covariance matrix in optimal signal detection
TLDR
An exact confidence bound is obtained for the realized signal-to-noise ratio which is useful in optimum selection of features in the case of detecting a single signal specified by a real or a complex vector.
Iterative algorithms for optimal signal reconstruction and parameter identification given noisy and incomplete data
We present a new approach to the problem of estimating multiple signal and parameter unknowns given noisy and incomplete data. Using cross-entropy, we fit a separable density to the given model…
Estimation of structured covariances with application to array beamforming
The estimation of covariance matrices which are structured, for example, of Toeplitz type, from measurement data is considered. The problem is considered in the context of array beamforming, and…
Estimation of structured covariance matrices
TLDR
An iterative algorithm that solves the necessary gradient equations for moderate-sized problems with reasonable computational ease is presented and theoretical convergence properties of the basic algorithm are investigated and robust modifications discussed.
On the uniqueness of the maximum-likeliwood estimate of structured covariance matrices
A generalized Burg technique has been developed recently by Burg, Luenberger, and Wenger for maximum likelihood estimation of structured covariance matrices. Uniqueness of the estimate in the…
Expectations of useful complex Wishart forms
TLDR
This paper derives concise formulas for E{WAW},E{W−1AW−1}, and other useful expressions, whereW follows the complex Wishart distribution andA is a deterministic matrix, which need not be symmetric.
A Bayesian Approach to Adaptive Detection in Nonhomogeneous Environments
TLDR
Two Bayesian versions of the adaptive matched filter are presented, where the conventional maximum likelihood estimate of the primary data covariance matrix is replaced either by its minimum mean-square error estimate or by its maximum a posteriori estimate.
Multiple emitter location and signal Parameter estimation
TLDR
The multiple signal classification (MUSIC) algorithm is described, which provides asymptotically unbiased estimates of number of incident wavefronts present and directions of arrival (DOA) (or emitter locations) and strengths and cross correlations among the incident waveforms.
Detection of signals by information theoretic criteria
  • M. Wax, T. Kailath
  • Computer Science
    IEEE Trans. Acoust. Speech Signal Process.
  • 1985
TLDR
Simulation results that illustrate the performance of the new method for the detection of the number of signals received by a sensor array are presented.
...
...