Efficient Protocol to Discriminate Multiple Ballistic Targets
In this paper, we propose a novel protocol for discriminating multiple ballistic targets during the midcourse and terminal phases of their trajectories. The midcourse-phase protocol includes the following steps: 1) estimating the number of targets, 2) separating the trajectories, and 3) performing discrimination using either a convolutional neural network (CNN) or a nearest-neighbor classifier. The terminal-phase protocol performs discrimination using trajectory information or a CNN. Simulations that consider radar signals calculated using physical optics, indicate that the proposed protocol effectively discriminates between warheads and decoys in scenarios that involve multiple targets.
Authors
- Sewon Yoon (ORCID: https://orcid.org/0000-0003-0514-708X)
- Inoh Choi (ORCID: https://orcid.org/0000-0002-0410-1066)
- Sang‐Hong Park (ORCID: https://orcid.org/0000-0001-8512-1431)
- Joo‐Ho Jung (ORCID: https://orcid.org/0000-0001-7087-8079)
- Min Kim (ORCID: https://orcid.org/0000-0002-9528-4282)
Institutions
- Korea Institute of Ocean Science and Technology (KR)
- Korea Maritime and Ocean University (KR)
- Konkuk University (KR)
- Korea Maritime Institute (KR)
- Korea Institute of Maritime and Fisheries Technology (KR)
- Pukyong National University (KR)
Publication Details
- Journal
- IEEE Access
- Published
- 2025-01-01
- DOI
- https://doi.org/10.1109/access.2025.3584058
- Primary Topic
- Biometric Identification and Security
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- National Research Foundation of Korea
- Ministry of Science and ICT, South Korea