A maximum a posteriori algorithm for reconstruction of targets in incompletely defined correlated noise

Author

Willis, A.J. ; Koch, R. De Mello ; Spear, B. ; Klopper, A.

Author_Institution

Dept. of Electr. Eng., Univ. of the Witwatersrand, Johannesburg, South Africa

Volume

46

Issue

5

fYear

1998

fDate

5/1/1998 12:00:00 AM

Firstpage

1439

Lastpage

1443

Abstract

The maximum a posteriori (MAP) line spectral estimator used to characterize sinusoids in data corrupted by Gaussian noise of unknown correlation is generalized to the case where an experimental estimate of noise covariance is available. The estimator is robust to noise with mean square error and standard deviation falling below that of the classical MAP for increasing number of samples, while approaching classical MAP for the case of no prior knowledge

Keywords

Bayes methods; Gaussian noise; array signal processing; correlation methods; covariance matrices; direction-of-arrival estimation; maximum likelihood estimation; signal reconstruction; Bayesian approach; DOA; Gaussian noise corrupted data; MAP line spectral estimator; experimental estimate; incompletely defined correlated noise; maximum a posteriori algorithm; mean square error; noise covariance; noise covariance matrix; robust estimator; samples; sinusoids; standard deviation; target reconstruction; Acoustic noise; Africa; Bayesian methods; Direction of arrival estimation; Frequency estimation; Gaussian noise; Narrowband; Phased arrays; Sensor arrays; Vectors;

fLanguage

English

Journal_Title

Signal Processing, IEEE Transactions on

Publisher

ieee

ISSN

1053-587X

Type

jour

DOI

10.1109/78.668807

Filename

668807