Statistical CSI-based covert beamforming for UAV-assisted SAGINs

Abstract

Covert communication in space-air-ground integrated networks (SAGINs) faces severe challenges due to aerial platform mobility and the impracticality of acquiring instantaneous channel state information (CSI). To address this, we investigate a multi-cell SAGIN architecture where mobile unmanned aerial vehicle (UAV) relays assist low Earth orbit (LEO) satellite transmissions, proposing a covert beamforming framework relying solely on statistical CSI. Specifically, we formulate a sum ergodic rate maximization problem for legitimate users, constrained by an expected Kullback-Leibler divergence metric to guarantee transmission concealment against terrestrial wardens. To tackle this highly non-convex problem coupled with spatial interference, we transform the intractable covertness constraint via successive convex approximation. Subsequently, we develop an efficient iterative algorithm based on the weighted minimum mean square error criterion to jointly optimize the UAV transmit beamformers. Simulation results demonstrate that the proposed SCA-WMMSE algorithm converges rapidly, achieving a tight performance gap approaching the instantaneous CSI upper bound. Furthermore, evaluations reveal that location-aware relay selection significantly outperforms random baseline strategies, underscoring the efficacy of spatial coordination and global interference management for maximizing covert rate in the integrated networks.