Q-GERT survivability assessment of LEO satellite constellation

Abstract

Due to the complexity of space environment and the difficulty of satellite maintenance in outer space, survivability has become one of the most important problems of LEO satellite constellation. Network survivability is the ability of a system to complete communication missions in case of encountering accidents, failures or attacks. A survivable LEO satellite constellation can guarantee a certain mission success rate when some satellites fail, but the completed performance depends on the architecture designed, resources owned and route adopted by the satellite constellation. The existing satellite network survivability evaluation methods can not accurately describe the impact of constellation design parameters, limited resources and access algorithm on survivability. To overcome the above shortcomings, this paper proposes a new survivability assessment model with resource constraints. We exploit the queue graphical evaluation and review technology random network with feedback to characterize resource limitations and describe the dynamic random mission transfer process of LEO satellite constellation. Based on the time, jitter and consumed resources produced by different link arc activities when completing missions, combined with queuing birth and death technology, we derive the mission cost and network utility function, aiming at reflecting the completion ability of LEO satellite constellation for communication mission after facing fault or strike. The results show that increasing the orbit height does not significantly improve the network survivability; rerouting load balancing access strategy can better share the mission of failed satellite; LEO satellite constellation has weak survivability towards GCF; user retry after mission failure will aggravate delay.