Secure Precoding for Satellite NOMA-Aided Integrated Sensing and Communication

Satellite Internet of Things (IoT) network plays a crucial role in providing global coverage. To improve the efficient utilization of spectral and hardware resources in the satellite communications, integrated sensing and communication (ISAC) has attracted considerable attention. In addition, the greater broadcast nature of the satellite-terrestrial integrated network makes it extremely vulnerable to illegal eavesdropper (Eve). In this paper, we adopt non-orthogonal multiple access (NOMA) to support more users for the ISAC of a low earth orbit satellite system, with well-designed precoding according to whether the channel state information (CSI) of the Eve is perfect to ensure security. First, a joint precoding optimization problem is proposed to maximize the sum secrecy rate (SSR) of multiple users by considering the perfect CSI of the Eve. Then, we formulate a secure precoding optimization problem based on the imperfect CSI of the Eve, aiming to maximize the SSR of multiple users via artificial jamming. To address the non-convexity of the optimization problems, we transform them into the convex ones based on successive convex approximation, which includes Taylor’s approximation, arithmetic-geometric mean inequality, and linear matrix inequality. In addition, semi-definite relaxation and iterative penalty function methods are respectively used to optimize the secure precoding problems in the two cases: perfect CSI and imperfect CSI. Simulation results show that the proposed NOMA-ISAC scheme improves the SSR compared to the traditional time division multiple access while ensuring the sensing performance.