Full-Space UAV Trajectory Design for STAR-RIS-Assisted Secure Ground–Air Communications

In this paper, we study simultaneous transmitting and reflecting reconfigurable intelligent surface (STAR-RIS)-assisted secure ground-air communications. Ground nodes upload data to an unmanned aerial vehicle (UAV) with the assistance of a STAR-RIS, while facing eavesdropping threats targeting confidential information from a secure node. Planning UAV trajectories in a full space instead of only on the single side of a STAR-RIS is helpful to make the most of UAV flexibility. However, the transmission rates of ground nodes is hard to express because cascaded channels vary with the UAV’s location relative to the STAR-RIS. To address this issue, we propose a unified expression of the rate incorporating the UAV-location-dependent cascaded channels. Based on the proposed expression, we maximize the average secrecy rate (ASR) of a secure node while ensuring the rate requirements of a regular node via robust optimization of STAR-RIS coefficients and UAV trajectory. Specifically, we consider the imperfect eavesdropping channel state information. Due to the non-convex structure, the original mixed integer nonlinear programming problem is decomposed into two subproblems. Semidefinite relaxation, S-Procedure, and successive convex approximation techniques are used to tackle the non-convex objective function and constraints. Finally, the two subproblems are solved iteratively. Numerical simulations reveal that the proposed scheme achieves the full-space flight of the UAV and outperforms the benchmark schemes in terms of ASR.