Graph-Based Lightweight Satellite Selection for Emergency Communication Scenarios

In disaster or defense scenarios, whether caused by natural disasters or intentional sabotage of terrestrial infrastructure, isolated information areas are often created due to the disruption of ground communication networks. In such scenarios, large volumes of data can be relayed to the outside world by establishing satellite links via Very Small Aperture Terminals (VSATs). The transmitted data typically include situational awareness information such as high-definition video and 3D mapping data, which are time-sensitive and require high uplink data rates. Additionally, due to the high mobility of Low Earth Orbit (LEO) satellites, VSATs must frequently switch the connection to a different satellite to maintain a communication link (i.e. handover). This introduces additional overhead and further complicates the challenge of efficient satellite selection in emergency scenarios. To address these challenges, this paper proposes a lightweight graph-based satellite selection method. Specifically, a cost function is constructed by jointly considering uplink data rate, delay, and HO overhead. The satellite selection problem is formulated as a path optimization problem over a time-expanded graph, which is efficiently solved using Dijkstra’s algorithm. Simulation results show that, compared with baseline satellite selection strategies and the MultiAttribute Decision Making (MADM)-based approach, the proposed method achieves superior performance in terms of average data rate, latency, and HO frequency. Moreover, in contrast to the Bellman-Ford path search algorithm, it significantly reduces computational complexity, enabling lightweight deployment in emergency scenarios while ensuring Quality of Service (QoS) guarantees.