Deep Learning-Based Anti-Jamming Beamforming Designs Against Adversarial Jamming Attacks
IEEE Transactions on Wireless Communications · 2026 · 25 · 20900-20912
DOI: 10.1109/twc.2026.3713089 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Wireless Communications, 2026. Verfasst von Ohseung Kwon und 3 weiteren Autor:innen. Die Arbeit nennt 45 Literaturangaben. Gefördert von: National Research Foundation of Korea, Korean Government Ministry of Science and ICT, Korea Research Institute for Defense Technology Planning and Advancement.
Abstract (Originalsprache)
This paper proposes a deep learning approach for designing anti-jamming transmission techniques in multi-user multi-antenna systems in the presence of an intelligent jammer. A base station and users are interrupted by the jammer in both the uplink pilot and downlink data transmissions. To defend against these jamming attacks, we develop jamming-resistant beamforming techniques which are robust to arbitrary jamming strategies. This can be formalized as a worst-case optimization problem that identifies beamforming vectors against the jammer designed to minimize the performance of legitimate networks. To address this nonconvex task, we leverage deep neural networks (DNNs) for jamming and beamforming optimization procedures. The worst-case jammer DNN is determined to generate the jamming signals interfering with the uplink pilot and downlink data transmissions such that the sum rate performance of users is minimized. Then, we optimize a beamformer DNN at the BS to yield anti-jamming beamforming vectors against the jammer DNN. To further enhance the robustness to various jamming policies, we propose a joint training mechanism that optimizes the beamformer and jammer DNNs in an alternating manner. Numerical results demonstrate the superiority of the proposed DNN-based approach over existing anti-jamming techniques.
Förderung
- National Research Foundation of Korea
- Korean Government Ministry of Science and ICT
- Korea Research Institute for Defense Technology Planning and Advancement
- Korean Government [Defense Acquisition Program Administration (DAPA)] through the Space-Layer Intelligent Communication Network Laboratory in 2022
Zitieren
APA
Kwon, O., Lee, H., Debbah, M., & Lee, I. (2026). Deep Learning-Based Anti-Jamming Beamforming Designs Against Adversarial Jamming Attacks. IEEE Transactions on Wireless Communications, 25, 20900-20912. https://doi.org/10.1109/twc.2026.3713089
Vancouver
Kwon O, Lee H, Debbah M, Lee I. Deep Learning-Based Anti-Jamming Beamforming Designs Against Adversarial Jamming Attacks. IEEE Transactions on Wireless Communications. 2026;25:20900-20912. doi:10.1109/twc.2026.3713089
BibTeX
@article{Kwon2026,
author = {Kwon, Ohseung and Lee, Hoon and Debbah, Merouane and Lee, Inkyu},
title = {Deep Learning-Based Anti-Jamming Beamforming Designs Against Adversarial Jamming Attacks},
journal = {IEEE Transactions on Wireless Communications},
year = {2026},
volume = {25},
pages = {20900--20912},
doi = {10.1109/twc.2026.3713089}
}
EUVIMED