The following information was submitted:
Transactions: WSEAS TRANSACTIONS ON INFORMATION SCIENCE AND APPLICATIONS
Transactions ID Number: 53-372
Full Name: Jiansheng Fu
Position: Ph.D. Candidate
Age: ON
Sex: Male
Address: School of Electronic Engineering, University of Electronic Science and Technology of China, Chengdu, Sichuan
Country: CHINA
Tel: 13608237285
Tel prefix: 86
Fax: +86-28-83207082
E-mail address: fujiansheng2010@126.com
Other E-mails: gladwolf01@163.com
Title of the Paper: radar HRRP target recognition based on discriminant information analysis
Authors as they appear in the Paper: Jiansheng Fu, Kuo Liao, Wanlin Yang
Email addresses of all the authors: fujiansheng2010@126.com,54900766@qq.com,wlyang@uestc.edu.cn
Number of paper pages: 17
Abstract: In radar HRRP target recognition, the quality and quantity of Discriminant Information (DI), which one is more important? Accompanied with this issue, the paper proceeds to delve into DI analysis, and accordingly, three fundamental DI extraction models are proposed, i.e., PGA, PIB and AIB. Among these models, PIB and AIB both aim to obtain Between-class DI (B-DI) from individual standpoints while PGA obtains Among-class DI (A-DI) from a general viewpoint; PGA and PIB are both used for passive recognition while AIB for active recognition. In order to externalize these models, we conduct Generalized Discriminant Analysis (GDA) into them, and two GDA variations come forth, i.e., PIB-based GDA (PIB-GDA) and AIB-based GDA (AIB-GDA). Theoretical analyses and experimental results indicate as follows. Firstly, although PGA prevails in pattern recognition, but the implementation prospect is hardly optimistic on account of the weak anti-fading ability of A-DI. Compared with !
PGA, PIB and AIB are both more suitable to multi-class discrimination due to the relative stability of B-DI. Secondly, in general, PIB-GDA is inferior to AIB-GDA but superior to GDA to many challenges, such as computational efficiency, target quantity, aspect and sample variation, noise disturbance, etc.
Keywords: Discriminant information, Feature extraction, Target recognition, Generalized discriminant analysis.
EXTENSION of the file: .pdf
Special (Invited) Session:
Organizer of the Session:
How Did you learn about congress:
IP ADDRESS: 125.71.231.82
