Grouping Enhanced Cooperative Covert Communications in RIS-Aided Multi-User Systems

In this paper, we aim at reconfigurable intelligent surfaces (RIS) aided multi-user communication systems including multiple user pairs, where an important target user pair needs to guarantee covertness in the presence of an adversarial warden with imperfect channel state information (CSI). By adopting the grouping technology, the user pairs in the same group share the same channel while different groups use orthogonal channels. Then, the mutual interference from the perspective of cooperation is considered to provide a covert cover for the target user pair. As a result, a grouping enhanced cooperative covert communications scheme with the aid of RIS is proposed. First, the analytical expressions of detection error probability, equivalent covert constraint based on Kullback-Leibler divergence and general user covert rate are derived. Then, to simultaneously guarantee the covertness and the robustness, the maximization of the worst-case sum rate with covert constraint is modeled as a non-convex optimization problem by jointly designing user group selection, transmission power allocation and RIS phase shift. To deal with the intractable problem consisting of mixed-integer programming, an alternative optimization algorithm that includes three subproblems is proposed. In each subproblem, fractional programming, relaxation variables and successive convex approximation methods are applied. Besides, S-procedure and general sign definition lemmas are applied to transform the CSI uncertainty into linear matrix inequalities. Finally, the simulation results demonstrate that the proposed cooperative covert communications scheme always outperforms the benchmark schemes with a better robustness.