A Distributed Survivable Routing Algorithm for Mega-Constellations With Inclined Orbits

• Published in: IEEE access Vol. 8; pp. 219199 - 219213
• Main Authors: Qi, Xiaoxin, Zhang, Bing, Qiu, Zhiliang
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 2020; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Alliances; Delays; Flooding; Inclined constellation; Low earth orbit satellites; Low earth orbits; mega-constellation; Network topologies; Network topology; Orbits; Routing; Satellite constellations; Satellites; Signaling; Survivability; Topology
• ISSN: 2169-3536; 2169-3536
• DOI: 10.1109/ACCESS.2020.3041346

Mega-constellations consisting of hundreds to thousands of low-earth-orbit (LEO) satellites are an attractive solution for providing global ubiquitous network access. Due to good coverage properties for populated areas, inclined orbits are gaining popularity among commercial constellations. A scalable routing algorithm with survivability plays a key role in such systems. In this paper, we propose a distributed survivable routing algorithm for mega-constellations with inclined orbits. First, the special topology characteristic of inclined constellations is identified and formalized. Based on the topology characterization, a basic X-Y routing algorithm is presented to determine multiple primary and secondary paths towards each destination utilizing the regularity of the network topology with minimal computation overhead. Then, a failure recovery mechanism which consists of a restricted flooding mechanism and a pre-detour mechanism is proposed to reduce end-to-end delay and signaling overhead in case of link failures. Besides, a partial-record loop avoidance mechanism is proposed to deal with routing loops with minimal overhead. Finally, a vector-based next hop selection mechanism is proposed to facilitate the selection of next hop while incorporating various criteria. The performance of the proposed routing algorithm is evaluated through simulation on the Starlink constellation. Simulation results show that our proposal achieves scalability by reducing signaling overhead and provides better quality of service in terms of end-to-end delay under link failures.