Robust secure beamforming algorithm for intelligent reflecting surface-assisted satellite-terrestrial integrated networks

Abstract

For intelligent reflection surface (IRS) assisted satellite-terrestrial integrated network (STIN), a robust secure beamforming method based on the imperfect channel state information (CSI) of eavesdroppers is proposed. First of all, considering that the satellite uses spot beam technology to serve the earth station, while the ground base station (BS) serves multiple ground users through multicast technology, and in the case of frequency spectrum shared between the two networks, a joint optimization problem is formulated to minimize the total transmit power of STIN while satisfying the quality of service constraints of the BS users and the achievable secrecy rate constraints of the earth station. Secondly, in order to solve this non-convex problem, the upper bound and the lower bound of the eavesdroppers’ output signal-to-noise ratio under the condition of imperfect CSI are derived by employing triangle inequality together with Holder inequality. Then, a joint optimization scheme of robust BF and power control based on semi-positive definite programming and penalty function method is proposed to realize the secure and reliable transmission in STIN. Finally, the simulation results verify the effectiveness and superiority of the proposed algorithm.