A Direct Localization Method of Multiple Distributed Sources Based on the Idea of Multiple Signal Classification

Yanqing REN  Zhiyu LU  Daming WANG  Jian LIU  

IEICE TRANSACTIONS on Communications   Vol.E101-B   No.5   pp.1246-1256
Publication Date: 2018/05/01
Publicized: 2017/11/16
Online ISSN: 1745-1345
DOI: 10.1587/transcom.2017EBP3199
Type of Manuscript: PAPER
Category: Antennas and Propagation
emitter localization,  distributed sources,  multiple signal classification,  Cramer-Rao bound,  array signal processing,  

Full Text: PDF>>
Buy this Article

The Localization of distributed sources has attracted significant interest recently. There mainly are two types of localization methods which are able to estimate distributed source positions: two-step methods and direct localization methods. Unfortunately, both fail to exploit the location information and so suffer a loss in localization accuracy. By utilizing the information not used in the above, a direct localization method of multiple distributed sources is proposed in this paper that offers improved location accuracy. We construct a direct localization model of multiple distributed sources and develop a direct localization estimator with the theory of multiple signal classification. The distributed source positions are estimated via a three-dimensional grid search. We also provide Cramer-Rao Bound, computational complexity analysis and Monte Carlo simulations. The simulations demonstrate that the proposed method outperforms the localization methods above in terms of accuracy and resolution.