Maximum Likelihood DOA Estimation of Multiple Wideband Sources in the Presence of Nonuniform Sensor Noise

We investigate the maximum likelihood (ML) direction-of-arrival (DOA) estimation of multiple wideband sources in the presence of unknown nonuniform sensor noise. New closed-form expression for the direction estimation Cramér-Rao-Bound (CRB) has been derived. The performance of the conventional wi...

Full description

Bibliographic Details
Main Authors: K. Yao, R. E. Hudson, F. Lorenzelli, C. E. Chen
Format: Article
Language:English
Published: SpringerOpen 2007-12-01
Series:EURASIP Journal on Advances in Signal Processing
Online Access:http://dx.doi.org/10.1155/2008/835079
id doaj-1a6830a6d50b45c59d2e7d99e4d7b76a
record_format Article
spelling doaj-1a6830a6d50b45c59d2e7d99e4d7b76a2020-11-25T00:31:56ZengSpringerOpenEURASIP Journal on Advances in Signal Processing1687-61722007-12-01200810.1155/2008/835079Maximum Likelihood DOA Estimation of Multiple Wideband Sources in the Presence of Nonuniform Sensor NoiseK. YaoR. E. HudsonF. LorenzelliC. E. ChenWe investigate the maximum likelihood (ML) direction-of-arrival (DOA) estimation of multiple wideband sources in the presence of unknown nonuniform sensor noise. New closed-form expression for the direction estimation Cramér-Rao-Bound (CRB) has been derived. The performance of the conventional wideband uniform ML estimator under nonuniform noise has been studied. In order to mitigate the performance degradation caused by the nonuniformity of the noise, a new deterministic wideband nonuniform ML DOA estimator is derived and two associated processing algorithms are proposed. The first algorithm is based on an iterative procedure which stepwise concentrates the log-likelihood function with respect to the DOAs and the noise nuisance parameters, while the second is a noniterative algorithm that maximizes the derived approximately concentrated log-likelihood function. The performance of the proposed algorithms is tested through extensive computer simulations. Simulation results show the stepwise-concentrated ML algorithm (SC-ML) requires only a few iterations to converge and both the SC-ML and the approximately-concentrated ML algorithm (AC-ML) attain a solution close to the derived CRB at high signal-to-noise ratio.http://dx.doi.org/10.1155/2008/835079
collection DOAJ
language English
format Article
sources DOAJ
author K. Yao
R. E. Hudson
F. Lorenzelli
C. E. Chen
spellingShingle K. Yao
R. E. Hudson
F. Lorenzelli
C. E. Chen
Maximum Likelihood DOA Estimation of Multiple Wideband Sources in the Presence of Nonuniform Sensor Noise
EURASIP Journal on Advances in Signal Processing
author_facet K. Yao
R. E. Hudson
F. Lorenzelli
C. E. Chen
author_sort K. Yao
title Maximum Likelihood DOA Estimation of Multiple Wideband Sources in the Presence of Nonuniform Sensor Noise
title_short Maximum Likelihood DOA Estimation of Multiple Wideband Sources in the Presence of Nonuniform Sensor Noise
title_full Maximum Likelihood DOA Estimation of Multiple Wideband Sources in the Presence of Nonuniform Sensor Noise
title_fullStr Maximum Likelihood DOA Estimation of Multiple Wideband Sources in the Presence of Nonuniform Sensor Noise
title_full_unstemmed Maximum Likelihood DOA Estimation of Multiple Wideband Sources in the Presence of Nonuniform Sensor Noise
title_sort maximum likelihood doa estimation of multiple wideband sources in the presence of nonuniform sensor noise
publisher SpringerOpen
series EURASIP Journal on Advances in Signal Processing
issn 1687-6172
publishDate 2007-12-01
description We investigate the maximum likelihood (ML) direction-of-arrival (DOA) estimation of multiple wideband sources in the presence of unknown nonuniform sensor noise. New closed-form expression for the direction estimation Cramér-Rao-Bound (CRB) has been derived. The performance of the conventional wideband uniform ML estimator under nonuniform noise has been studied. In order to mitigate the performance degradation caused by the nonuniformity of the noise, a new deterministic wideband nonuniform ML DOA estimator is derived and two associated processing algorithms are proposed. The first algorithm is based on an iterative procedure which stepwise concentrates the log-likelihood function with respect to the DOAs and the noise nuisance parameters, while the second is a noniterative algorithm that maximizes the derived approximately concentrated log-likelihood function. The performance of the proposed algorithms is tested through extensive computer simulations. Simulation results show the stepwise-concentrated ML algorithm (SC-ML) requires only a few iterations to converge and both the SC-ML and the approximately-concentrated ML algorithm (AC-ML) attain a solution close to the derived CRB at high signal-to-noise ratio.
url http://dx.doi.org/10.1155/2008/835079
work_keys_str_mv AT kyao maximumlikelihooddoaestimationofmultiplewidebandsourcesinthepresenceofnonuniformsensornoise
AT rehudson maximumlikelihooddoaestimationofmultiplewidebandsourcesinthepresenceofnonuniformsensornoise
AT florenzelli maximumlikelihooddoaestimationofmultiplewidebandsourcesinthepresenceofnonuniformsensornoise
AT cechen maximumlikelihooddoaestimationofmultiplewidebandsourcesinthepresenceofnonuniformsensornoise
_version_ 1725321709109641216