Hybrid-RIS Empowered UAV-Assisted ISAC Systems: Transfer Learning-Based DRL
IEEE Transactions on Communications · 2025-09 · 73 (9) · 8314-8329
DOI: 10.1109/tcomm.2025.3548797 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Communications, 2025-09. Verfasst von Prajwalita Saikia und 5 weiteren Autor:innen. Die Arbeit wurde bisher 41-mal zitiert (Crossref) und nennt 69 Literaturangaben.
Abstract (Originalsprache)
In this paper, we consider a novel hybrid reconfigurable intelligent surface (HRIS) consisting of active as well as passive reflecting elements mounted on unmanned aerial vehicle (UAV). The aim is to improve air-to-ground communication by assisting multiple users, while detecting several low mobility targets. We formulate a sum-rate optimization problem that accounts for statistical channel estimation errors (SCEEs) to concurrently fine-tune both active and passive phase-shift matrices, UAV trajectory, and transmit beamformer for integrated sensing and communication (ISAC). Subsequently, we introduce a transfer learning based approach combining with deep deterministic policy gradient (DDPG) to enhance the overall data rate while minimizing the time it takes for users to transmit data. Additionally, we present an alternating optimization (AO) algorithm that employs a repetitive method to address the combinatorial nonconvex optimization problem and offers a solution that is very close to optimal. Finally, we showcase the superiority of the proposed scheme through Monte Carlo simulations. Also, we have compared the performance with perfect channel state information (CSI) counterpart. The outcomes of simulations confirm the theoretical analysis and demonstrate the efficiency of the proposed framework. Additionally, the results reveal the advantages of incorporating HRIS aided UAV assisted ISAC in improving the quality of both communication and sensing performance.
Zitieren
APA
Saikia, P., Jee, A., Singh, K., Huang, W., Boulogeorgos, A. A., & Tsiftsis, T. A. (2025). Hybrid-RIS Empowered UAV-Assisted ISAC Systems: Transfer Learning-Based DRL. IEEE Transactions on Communications, 73(9), 8314-8329. https://doi.org/10.1109/tcomm.2025.3548797
Vancouver
Saikia P, Jee A, Singh K, Huang W, Boulogeorgos AA, Tsiftsis TA. Hybrid-RIS Empowered UAV-Assisted ISAC Systems: Transfer Learning-Based DRL. IEEE Transactions on Communications. 2025;73(9):8314-8329. doi:10.1109/tcomm.2025.3548797
BibTeX
@article{Saikia2025,
author = {Saikia, Prajwalita and Jee, Anand and Singh, Keshav and Huang, Wan-Jen and Boulogeorgos, Alexandros-Apostolos A. and Tsiftsis, Theodoros A.},
title = {Hybrid-RIS Empowered UAV-Assisted ISAC Systems: Transfer Learning-Based DRL},
journal = {IEEE Transactions on Communications},
year = {2025},
volume = {73},
number = {9},
pages = {8314--8329},
doi = {10.1109/tcomm.2025.3548797}
}
EUVIMED