Title :
A method of dynamic DOA estimation with an unknown number of sources
Author :
Rongjie Wang ; Yiju Zhan
Author_Institution :
Marine Eng. Inst., Jimei Univ., Xiamen, China
Abstract :
Aiming to source number determination and DOA estimation under the case of time-varying source number, we propose a method of DOA estimation with an unknown number of sources. Firstly, we introduced an algorithm based on cross-validation technique to determine the number of sources. Then dynamic DOAs of source are estimated using an algorithm based on blind source separation under the case that number of sources are unknown in advanced and it is time-varying. The effectiveness of the proposed method was validated by simulation of time-invariant and time-varying numbers of source, and the simulation results show the proposed method has superior evaluation performance to other conventional methods.
Keywords :
blind source separation; direction-of-arrival estimation; principal component analysis; blind source separation; cross-validation technique; dynamic DOA estimation; source number determination; time-varying source number; Arrays; Blind source separation; Direction-of-arrival estimation; Heuristic algorithms; Maximum likelihood estimation; Sensors; Blind source separation; Cross validation; Direction of arrival; Number of sources; Principal component analysis;
Conference_Titel :
Mechatronics and Automation (ICMA), 2015 IEEE International Conference on
Conference_Location :
Beijing
Print_ISBN :
978-1-4799-7097-1
DOI :
10.1109/ICMA.2015.7237465