Skip to main content

Deep learning based secure transmissions for the UAV-RIS assisted networks: Trajectory and phase shift optimization.

Li, J., Wang, D., Zhang, J., Alfarraj, O., He, Y., Al-Rubaye, S., Yu, K. and Mumtaz, S., 2024. Deep learning based secure transmissions for the UAV-RIS assisted networks: Trajectory and phase shift optimization. In: GLOBECOM 2024 - 2024 IEEE Global Communications Conference. New York, NY: IEEE, 1617-1622.

Full text available as:

[thumbnail of m16482-li final.pdf]
Preview
PDF
m16482-li final.pdf - Accepted Version
Available under License Creative Commons Attribution Non-commercial.

1MB

Official URL: https://globecom2024.ieee-globecom.org/

DOI: 10.1109/GLOBECOM52923.2024.10901660

Abstract

This paper investigates the secure transmissions in the Unmanned Aerial Vehicle (UAV) communication network facilitated by a Reconfigurable Intelligent Surface (RIS). In this network, the RIS acts as a relay, forwarding sensitive information to the legitimate receiver while preventing eavesdropping. We optimize the positions of the UAV at different time slots, which gives another degree to protect the privacy information. For the proposed network, a secrecy rate maximization problem is formulated. The non-convex problem is solved by optimizing the RIS’s phase shifts and UAV trajectory. The RIS phase shift optimization problem is converted into a series of subproblems, and a non-linear fractional programming approach is conceived to solve it. Furthermore, the first-order taylor expansion is employed to transform the UAV trajectory optimization into convex function, and then we use the deep Q-network (DQN) method to obtain the UAV’s trajectory. Simulation results show that the proposed scheme enhances the secrecy rate by 18.7% compared with the existing approaches.

Item Type:Book Section
ISBN:9798350351255
Additional Information:ISSN: 2576-6813
Uncontrolled Keywords:Unmanned aerial vehicle; reconfigurable intelligent surface; security
Group:Faculty of Science & Technology
ID Code:40300
Deposited By: Symplectic RT2
Deposited On:09 Sep 2024 12:40
Last Modified:23 Apr 2025 14:28

Downloads

Downloads per month over past year

More statistics for this item...
Repository Staff Only -