Title :
Bearing estimation in the unknown noise fields
Author :
Frikel, M. ; Bourennane, S.
Author_Institution :
CNRS, Corte, France
Abstract :
In most algorithms of estimation of the direction of arrival (DOA) of the sources, the additive noise term is assumed to be spatially white or known to within a multiplicative scalar. When the noise is nonwhite but has a known spectral, we can still handle the problem through pre-whitening. However there are no techniques presently available to deal with completely unknown noise fields. In this paper, a proposed algorithm is used at each frequency to estimate the noise spectral matrix, which is used for whitening the received spectral matrix. A fast focusing operator is used to construct coherent source subspace. To illustrate the performances of the proposed algorithm, the problem of the localization of the sources is studied, without and with pre-whitening pre-processing. The simulation results show the improvement of the DOA when the perfect modelisation of the noise spectral matrix is used
Keywords :
direction-of-arrival estimation; matrix algebra; noise; spectral analysis; DOA estimation; additive noise term; bearing estimation; coherent source subspace; direction of arrival estimation; noise spectral matrix estimation; unknown noise fields; Additive noise; Direction of arrival estimation; Frequency estimation; Linear antenna arrays; Noise measurement; Noise reduction; Parameter estimation; Principal component analysis; Sensor arrays; White noise;
Conference_Titel :
Decision and Control, 1996., Proceedings of the 35th IEEE Conference on
Conference_Location :
Kobe
Print_ISBN :
0-7803-3590-2
DOI :
10.1109/CDC.1996.574646