SMEN: Scattering Mechanism Embedded Network for Polarimetric Radar Corner Reflector Array Identification
IEEE Transactions on Aerospace and Electronic Systems · 2026 · 62 · 16728-16744
DOI: 10.1109/taes.2026.3723647 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Aerospace and Electronic Systems, 2026. Verfasst von Hao-Liang Li und 2 weiteren Autor:innen. Die Arbeit wurde bisher 13-mal zitiert (Crossref) und nennt 61 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Science and Technology Innovation Program of Hunan Province, Innovation Project for Graduate Student of National University of Defense Technology.
Abstract (Originalsprache)
Polarimetric radar plays a pivotal role in radar automatic target recognition. As a dominant passive jamming source, corner reflector arrays can significantly reduce the target detection performance of polarimetric radar systems, rendering their accurate identification a critical research focus in recent years. However, it remains challenging to achieve fine interpretation of target scattering mechanisms using only single polarimetric domain information. Furthermore, most existing identification methods rely on network structures originally designed for optical images and fail to account for the inherent scattering mechanisms of radar targets, which limits their performance potential. To address these issues, this article proposes a scattering mechanism embedded network (SMEN) framework for polarimetric radar corner reflector array identification. The core idea is to integrate expert knowledge of target joint-domain scattering mechanism interpretation into the design and training of deep-learning models. The main contributions are threefold. First, a general copolarization correlation pattern (GCPCP) interpretation tool is developed to characterize and visualize the target scattering properties in the joint polarimetric rotation-frequency domain. Second, a feature extraction and weighting module and a scattering mechanism learning module (SMLM) are designed to learn joint-domain scattering mechanism knowledge using GCPCP; subsequently, the physics-aware features extracted by SMLM are embedded into the deep-learning network via a multilevel feature fusion strategy, thereby guiding and constraining the model's learning process. Third, extensive experimental validations on the constructed electromagnetic simulation dataset and measured dataset demonstrate that the proposed SMEN not only achieves superior corner reflector array identification performance but also significantly enhances the algorithm's generalization and robustness.
Förderung
- National Natural Science Foundation of China
- Science and Technology Innovation Program of Hunan Province
- Innovation Project for Graduate Student of National University of Defense Technology
Zitieren
APA
Li, H., Hu, Y., & Chen, S. (2026). SMEN: Scattering Mechanism Embedded Network for Polarimetric Radar Corner Reflector Array Identification. IEEE Transactions on Aerospace and Electronic Systems, 62, 16728-16744. https://doi.org/10.1109/taes.2026.3723647
Vancouver
Li H, Hu Y, Chen S. SMEN: Scattering Mechanism Embedded Network for Polarimetric Radar Corner Reflector Array Identification. IEEE Transactions on Aerospace and Electronic Systems. 2026;62:16728-16744. doi:10.1109/taes.2026.3723647
BibTeX
@article{Li2026,
author = {Li, Hao-Liang and Hu, Yan-Shan and Chen, Si-Wei},
title = {SMEN: Scattering Mechanism Embedded Network for Polarimetric Radar Corner Reflector Array Identification},
journal = {IEEE Transactions on Aerospace and Electronic Systems},
year = {2026},
volume = {62},
pages = {16728--16744},
doi = {10.1109/taes.2026.3723647}
}
EUVIMED