Priority-based Adaptive Shortest Path Routing for IP over LEO Satellite Networks

Designing an efficient routing algorithm for LEO satellite constellations is crucial for optimizing IP over Satellite (IPoS) network resources. Since there could be many shortest paths between two satellites, an efficient routing algorithm should provide better utilization of these paths. For this purpose, we propose a Priority-based Adaptive Routing (PAR) technique, which distributedly sets the shortest path through a destination. In this technique, a direction decision is made at each hop by a priority mechanism, depending on the past utilization and buffering information about the links. We further make some enhancements on PAR, and propose ePAR algorithm that also accounts for the contentions between packets with different source-destination pairs. We explore the performances of PAR and ePAR algorithms based on an extensive set of simulations, and compared their performances with static and adaptive routing techniques as a reference. Obtained results show that while the proposed PAR algorithm is promising for use in LEO satellite networks, ePAR algorithm may be more suitable for MEO satellite networks.