Theoretical analysis of an improved covariance matrix estimator in non-Gaussian noise [radar detection applications].
Frédéric PascalPhilippe ForsterJean Philippe OvarlezPascal LarzabalPublished in: ICASSP (4) (2005)
Keyphrases
- covariance matrix
- theoretical analysis
- estimation error
- covariance matrices
- millimeter wave
- principal component analysis
- sample size
- mahalanobis distance
- least squares
- geometrical interpretation
- eigendecomposition
- eigenvalues and eigenvectors
- correlation matrix
- positive definite
- pseudo inverse
- signal processing
- genetic algorithm
- symmetric matrix
- maximum a posteriori
- signal to noise ratio
- non stationary
- gaussian mixture
- evolutionary algorithm
- machine learning