Decentralized robust secure beamforming for IRS-assisted integrated satellite–terrestrial networks

Abstract

Integrated satellite–terrestrial networks (ISTNs) are considered a promising framework to provide seamless connectivity for future mobile communication. However, due to the spectrum sharing and heterogeneous architecture, interference management and secrecy transmission have become urgent issues in ISTNs. This paper proposes a decentralized robust beamforming (BF) scheme to enhance the secrecy performance of an intelligent reflecting surface (IRS)-assisted ISTN. In this framework, the satellite and terrestrial networks provide multicast services to satellite users (SUs) and terrestrial users (TUs), respectively, while sharing the same spectrum resources. To further improve communication quality and secrecy performance, an IRS is deployed on an unmanned aerial vehicle (UAV) to assist terrestrial transmissions and mitigate interference. To this end, we formulate an optimization problem to minimize the total transmit power of the ISTN while satisfying the achievable rate constraints of TUs and the achievable secrecy rate constraints of SUs. Given the non-convex nature of the problem and the availability of only imperfect channel state information (CSI), we decompose it into two parallel subproblems by introducing interference-related auxiliary variables. Furthermore, we leverage the triangle inequality and Hölder’s inequality to address channel uncertainty, employ Lagrange duality to iteratively update the auxiliary variables, and develop a decentralized robust algorithm for BF optimization. Finally, simulation results validate that the proposed scheme achieves superior secrecy performance while significantly reducing signaling overhead and computational complexity compared to conventional centralized BF schemes.