发表机构
Nanjing University; China Academy of Space Technology（南京大学; 中国空间技术研究院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文提出一种面向GNSS星间网络的低复杂度星载路由与拓扑协同构建框架，通过向后传播投递信息避免端到端路径搜索，在北斗场景中实现与接触图路由相当性能且计算时间减少约1500倍。
AI 中文摘要
星间链路是全球导航卫星系统（GNSS）自主运行的关键使能技术。然而，由于GNSS星座拓扑频繁变化以及星载计算资源有限对路由计算造成的约束，星载路由仍面临巨大挑战。本文针对GNSS星间网络时隙化结构，开发了一种路由与拓扑控制框架。利用每颗卫星在一个时隙内至多与一个调度对端通信这一事实，所提出的路由方法在时隙序列上向后传播未来投递信息，并直接确定数据应存储还是转发。该结构避免了大规模端到端路径搜索，并使拓扑生成与路由计算能够并发进行。相同的投递信息进一步引入拓扑构建，使链路调度能够直接考虑端到端通信性能。采用主星自主定轨场景验证了这种拓扑-路由协同构建机制。在24小时、30颗卫星的北斗场景中，所提出的路由方法实现了与接触图路由相当的路由性能，同时将路由计算时间减少约1500倍。协同构建的拓扑进一步改善了全对全通信及两个主星相关流量方向，同时保持了强星间测距性能。这些结果填补了自主拓扑规划与自主路由计算之间的关键空白，为未来GNSS更完整的星载自主组网提供支持。
英文摘要
Inter-satellite links constitute a critical enabler for the autonomous operation of global navigation satellite systems (GNSSs).Nevertheless, onboard routing still poses substantial challenges, stemming from the frequent topology variations of GNSS constellations and the constraint of limited onboard computational resources for routing calculation. This paper develops a routing and topology-control framework tailored to the time-slotted structure of GNSS inter-satellite networks. By exploiting the fact that each satellite communicates with at most one scheduled peer in a slot, the proposed routing method propagates future delivery information backward over the slot sequence and directly determines whether data should be stored or forwarded. This structure avoids massive end-to-end path searches and enables topology generation and routing computation to proceed concurrently. The same delivery information is further introduced into topology construction so that link scheduling can directly account for end-to-end communication performance. A master-satellite autonomous orbit-determination scenario is used to demonstrate this topology--routing co-construction mechanism. In a 24-h, 30-satellite BeiDou scenario, the proposed routing method achieves routing performance comparable to Contact Graph Routing while reducing routing computation time by approximately 1500 times. The co-constructed topology further improves all-to-all communication and both master-related traffic directions while preserving strong inter-satellite ranging performance. These results fill a critical gap between autonomous topology planning and autonomous routing computation, supporting more complete onboard autonomous networking for future GNSSs.
Comments11 pages, 5 figs
