Adaptive Snapshot-Based Elastic Route Reconstruction for Distributed LEO Satellite Networks

Abstract

The distributed low Earth orbit (LEO) satellite networks have been considered a promising solution to provide robust broadband ubiquitous access. However, the constantly changing topology introduces unacceptable overheads and performance to route computation. This article first introduces the adaptive snapshot into distributed LEO satellite networks and defines a path delay increment (PDI) to measure the performance loss relative to the shortest path. Consequently, the problem of route reconstruction is formulated as an optimization problem of joint routing and snapshot dividing, with the maximum PDI limitation, to minimize the number of snapshots. Next, we propose an iterative elastic route reconstruction algorithm to obtain a suboptimal solution with polynomial complexity while solving the problem of outdated route computation under intensive failures. Furthermore, a snapshot synchronization mechanism is developed to support the proposed routing algorithm on the distributed framework. Simulation results show that the proposed routing scheme can effectively trade off the average delay, packet loss rate, and storage and computational overhead.