Dynamic Source–Jammer Role Assignment for Secure UAV Communications: Joint Trajectory and Power Optimization

Abstract

Unmanned aerial vehicle (UAV)-enabled communications provides a cost-effective option for next-generation wireless networks. Nevertheless, the inherently open characteristics of air-to-ground links leave such communications particularly susceptible to eavesdropping. Motivated by this vulnerability, this paper studies a secure UAV communication system supported by cooperative jamming from the perspective of physical layer security (PLS). We introduce a new mechanism in which UAVs dynamically alternate between the source and jammer roles according to time-varying channel conditions. A security optimization problem is then formulated to jointly determine the role assignments, UAV trajectories, and transmission power control, with the objective of maximizing the worst-case secrecy rate (WCSR). The formulated problem is highly non-convex because of its binary role assignment variables, coupled decision variables, and semi-infinite constraints. To address this problem, we develop an optimization framework termed OASIS, which partitions the original formulation into tractable subproblems and iteratively solves them through successive convex approximation (SCA) and the S-procedure. Under simulation scenarios with K=2 and K=4 potential eavesdroppers and a maximum location-uncertainty radius of 20 m, the proposed strategy achieves a higher WCSR than the evaluated fixed-role, partial-optimization, and dynamic non-robust benchmark schemes.