End-to-end stochastic QoS performance under multi-layered satellite network

Abstract

To meet the growth of real-time and multimedia traffic, the next generation of satellite networks with a guarantee of quality of service (QoS) is indeed, urgent. In this paper, we support the multi-layered satellite network as the scenario, owing to dynamic topology and distinct classification of the generated traffics. We map the satellite network system into a tandem queuing model, which the purpose is devoted to use a mathematical tool for evaluating the performance bounds of per-flow end-to-end networks. For delay-sensitive traffics, we compare two different arrival models–Poisson process and self-similar process. Meanwhile, we apply traditional scheduling strategy to MEO nodes while considering link impairment between a pair of satellites. Finally, we analyze, in a numerically way, which parameters (and how they) influence the per-flow end-to-end performance bounds. Our analysis can be used as a reference to China’s future satellite topology and routing algorithms designed and, optimization given a network with performance requirements and constraints.