STAR-RIS-Aided Full-Space Covert Communications: Resisting the Position Randomness of Eavesdropper

This paper investigates a full-space covert communication (CC) scheme aided by a simultaneously transmitting and reflecting reconfigurable intelligent surface (STAR-RIS), which can resist the position randomness of the eavesdropper (PRE). In contrast to prevailing STAR-RIS assisted CC schemes assuming an eavesdropper with fixed position, the proposed CC scheme considers an eavesdropper which randomly locate at both sides of the STAR-RIS, leading to 360° eavesdropping risks. To restrict the eavesdropper’s ability to detect the CCs, the covert user is designed with two antennas working in full-duplex mode. One of the antennas is employed to receive the desired covert messages, while the other produces jamming signals at various power levels to impede the eavesdropper’s detection. Except the covert user, the base station (BS) also needs to serve a public user, which has certain quality of service (QoS) requirement. In order to construct a robust covert constraint, we analyze and derive a closed-form formula for the eavesdropper’s minimum detection error probability (DEP) in the worst-case situation. Subsequently, an optimization problem is established to maximize the covert rate of the system, through joint optimizing the bandwidth allocation, the active beamforming of the BS, and the passive beamforming of STAR-RIS, while adhering to the covert constraint and QoS requirement for the public user. An iterative algorithm is provided to tackle this non-convex optimization problem, utilizing the semi-definite relaxation (SDR) method and the augmented Lagrange technique. The simulation results demonstrate that the proposed STAR-RIS assisted CC scheme exhibits superior performance in resisting the full-space eavesdropping compared to other benchmark schemes, thereby validating the efficacy of the proposed scheme.