DocumentCode
2556876
Title
A kind of method for direction of arrival estimation based on blind source separation demixing matrix
Author
Kang Chun-yu ; Fan Wen-Tao ; Zhang Xin-hua ; Li Jun
Author_Institution
Res. Center of Signal & Inf., DaLian Navy Acad., Dalian, China
fYear
2012
fDate
29-31 May 2012
Firstpage
134
Lastpage
137
Abstract
Direction of arrival(DOA) estimation is the base of the underwater target orientation, tracking. Based on the characteristic of the array manifold can be estimated by complex blind source separation using the singular value decomposition, a new kind of model and method for DOA estimation is proposed using the directivity patterns of the blind source separation demixing matrix. Its efficiency and stability was tested by simulation data and recorded data in real sea. Results show that it can complete the real-time estimation of the target direction. And it is superior to the minimum variance distortionless response(MVDR) method and the method proposed in relevant literatures, it can obviously improves the detection capability of the sonar system for the faint target signal obviously.
Keywords
blind source separation; direction-of-arrival estimation; singular value decomposition; sonar detection; sonar signal processing; sonar tracking; target tracking; underwater sound; DOA estimation; MVDR method; array manifold; blind source separation; demixing matrix; direction of arrival estimation; directivity pattern; faint target signal detection; minimum variance distortionless response; real sea; singular value decomposition; sonar system detection; stability; target direction estimation; underwater target orientation; underwater target tracking; Arrays; Blind source separation; Direction of arrival estimation; Estimation; Manifolds; Marine vehicles; Sensors; Direction of arrival estimation; blind source separation; demixing matrix; minimum variance distortionless response;
fLanguage
English
Publisher
ieee
Conference_Titel
Natural Computation (ICNC), 2012 Eighth International Conference on
Conference_Location
Chongqing
ISSN
2157-9555
Print_ISBN
978-1-4577-2130-4
Type
conf
DOI
10.1109/ICNC.2012.6234543
Filename
6234543
Link To Document