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

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 sta...

Ausführliche Beschreibung

Gespeichert in:

Bibliographische Detailangaben
Veröffentlicht in:	IEEE transactions on signal processing 1998-05, Vol.46 (5), p.1439-1443
Hauptverfasser:	Willis, A.J., De Mello Koch, R., Spear, B., Klopper, A.
Format:	Artikel
Sprache:	eng
Schlagworte:	Acoustic noise Africa Applied sciences Bayesian methods Detection, estimation, filtering, equalization, prediction Direction of arrival estimation Exact sciences and technology Frequency estimation Gaussian noise Information, signal and communications theory Narrowband Phased arrays Sensor arrays Signal and communications theory Signal, noise Telecommunications and information theory Vectors
Online-Zugang:	Volltext bestellen
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

Beschreibung
Zusammenfassung:	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.
ISSN:	1053-587X 1941-0476
DOI:	10.1109/78.668807