EUVIMEDEUVIMED entdecken
Crossref

Multiple Access-Enabled Radar Antijamming Against Multimode Jammers via Partially Observable Reinforcement Learning

Yayun Gao, Ye Yuan, Yongfei Mao, Wei Yi, Lingjiang Kong

IEEE Transactions on Aerospace and Electronic Systems · 2026 · 62 · 19551-19567

DOI: 10.1109/taes.2026.3743295 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Aerospace and Electronic Systems, 2026. Verfasst von Yayun Gao und 4 weiteren Autor:innen. Die Arbeit nennt 48 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Natural Science Foundation of Sichuan Province, Fundamental Research Funds for the Central Universities.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Modern jamming techniques have evolved to possess increasing degrees of freedom, resulting in mismatch between the radar's decision space and the diverse jamming modes. Traditional active radar antijamming strategies typically rely on a 1-D decision process, assuming that each radar observation corresponds to a single jamming mode. However, in practice, different jamming modes can produce similar radar observations regarding whether the radar is being jammed or not, which introduces substantial challenges in identifying and countering dynamically switching jamming modes, especially under partial observability and unknown jamming environments. To address these challenges, we propose a multidimensional radar decision-making framework to enhance radar's ability to sense and evade jamming by integrating multiple-access (MA) techniques. This framework extends the radar's decision process from a single dimension to multiple dimensions, enabling the mapping of radar observations to multiple jamming modes and corresponding radar responses. The interaction between the radar and jammer is formulated as a partially observable Markov decision process (POMDP), modeling both the MA-based multidimensional decision process and the partial observation in antijamming. To solve this POMDP in unknown and dynamic jamming environments, we introduce an MA-based partially observable reinforcement learning (MA-PORL) strategy. Simulation results demonstrate that the expanded decision space provided by MA-PORL effectively reduces successful jamming events and frequency hopping, while simultaneously improving the signal-to-noise ratio and detection probability.

Förderung

  • National Natural Science Foundation of China
  • Natural Science Foundation of Sichuan Province
  • Fundamental Research Funds for the Central Universities

Zitieren

APA

Gao, Y., Yuan, Y., Mao, Y., Yi, W., & Kong, L. (2026). Multiple Access-Enabled Radar Antijamming Against Multimode Jammers via Partially Observable Reinforcement Learning. IEEE Transactions on Aerospace and Electronic Systems, 62, 19551-19567. https://doi.org/10.1109/taes.2026.3743295

Vancouver

Gao Y, Yuan Y, Mao Y, Yi W, Kong L. Multiple Access-Enabled Radar Antijamming Against Multimode Jammers via Partially Observable Reinforcement Learning. IEEE Transactions on Aerospace and Electronic Systems. 2026;62:19551-19567. doi:10.1109/taes.2026.3743295

BibTeX
@article{Gao2026,
  author = {Gao, Yayun and Yuan, Ye and Mao, Yongfei and Yi, Wei and Kong, Lingjiang},
  title = {Multiple Access-Enabled Radar Antijamming Against Multimode Jammers via Partially Observable Reinforcement Learning},
  journal = {IEEE Transactions on Aerospace and Electronic Systems},
  year = {2026},
  volume = {62},
  pages = {19551--19567},
  doi = {10.1109/taes.2026.3743295}
}
Kostenlos starten →