A Dual Threshold Handover Scheme for Improved Service Stability in Large-Scale Optical Satellite Networks

Optical satellite networks (OSNs), comprising thousands of low-earth orbit (LEO) satellites interconnected via laser links, will play a pivotal role in establishing seamless and high-speed global connectivity. However, large-scale LEO satellites pose significant challenges for mobility management within OSNs, including high satellite velocities, frequent handovers due to rapid orbital changes, and the dynamic nature of inter-satellite optical links. This study investigates handover management of high-velocity satellites in LEO networks and proposes a seamless, user-centric handover scheme under the software-defined satellite network (SDSN) architecture. In this process, a dual-threshold, graph-based conditional handover scheme is introduced to enhance handover efficiency and performance. Dynamic satellite nodes, segmented by time slots, are then mapped to vertices on a graph, wherein an analytic hierarchy process (AHP) is employed to determine edge weights. A dual-threshold graph (DualT-G) is then used to identify optimal satellites and handover timing to satisfy quality of service (QoS) requirements. In addition, this study examines the root causes of optical path variations across access satellites and introduces a systematic satellite encoding scheme. This strategy enables the handover selection process to efficiently estimate hop counts and optical transmission forwarding times between user-accessed satellites, thereby reducing delays. Simulated results showed the proposed handover scheme significantly reduced connection delays in Walker-Delta satellite optical networks and markedly improved the continuity of satellite network services. Specifically, for the simulated Starlink constellation, end-to-end propagation delays were reduced by 90% compared to joint threshold methods, with the fluctuation range reduced by nearly 90%.