Paper Abstract and Keywords |
Presentation |
2018-05-14 13:15
Closely Spaced Multiple Target Tracking with Super-Resolution DoA Estimation using Multiple Hypothesis Masanari Nakamura, Tetsutaro Yamada, Yasushi Obata, Hiroshi Kameda (Mitsubishi Electric Corp.) SANE2018-6 |
Abstract |
(in Japanese) |
(See Japanese page) |
(in English) |
In this paper, we propose multiple target tracking algorithm for dense targets which move toward the radar. When FBSSP-MUSIC is utilized for the estimation of the direction of arrival, the measurements could be unresolved randomly, since echoes of the dense targets are strongly correlated. In target tracking, the unresolved measurements degrade multiple target tracking performance, due to multiple tracks updated using the same unresolved measurement. The proposed method employs EM algorithm to resolve the measurement only when the measurement is unresolved. In EM algorithm, we use the mean values of the targets’ predictive distribution as the initial value. Therefore, the proposed method can improve tracking performance since each target is updated by the resolved measurement. Through the experiment which simulates two closely spaced targets moving toward the radar, we confirmed the effectiveness of the proposed method. |
Keyword |
(in Japanese) |
(See Japanese page) |
(in English) |
Radar / Multiple Target Tracking / Super-resolution Direction-of-Arrival / Unresolved Measurement / Expectation Maximization algorithm / / / |
Reference Info. |
IEICE Tech. Rep., vol. 118, no. 28, SANE2018-6, pp. 29-34, May 2018. |
Paper # |
SANE2018-6 |
Date of Issue |
2018-05-07 (SANE) |
ISSN |
Online edition: ISSN 2432-6380 |
Copyright and reproduction |
All rights are reserved and no part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publisher. Notwithstanding, instructors are permitted to photocopy isolated articles for noncommercial classroom use without fee. (License No.: 10GA0019/12GB0052/13GB0056/17GB0034/18GB0034) |
Download PDF |
SANE2018-6 |
|