A Dynamic Network Simulation Method of Mega-Constellation Networks

Zhixiang Gao,Jun Cai,Xin Xu,Dongming Bian
DOI: https://doi.org/10.1109/ICCC56324.2022.10066008
2022-12-09
Abstract:Mega-constellation networks have attracted great attention in recent years. The main difficulty of the research lies in the lack of packet-level simulation tools. Toward this end, this paper discusses the problem of topology construction and packet-level dynamic simulation for mega-constellation networks. A data-driven fixed timestep dynamic simulation method is proposed. This method combines the characteristics of the fixed timestep static simulation and the data-driven dynamic simulation. It takes into account the timeliness and simulation efficiency simultaneously. Then a local-state congestion avoidance routing method is presented. It is evaluated by using the proposed simulation method. Simulation results show that the data-driven fixed timestep dynamic simulation method can accurately and effectively capture the topology dynamics and path changes of mega-constellations. It is a promising tool for the quantitative evaluation of the dynamic behavior and performance of mega-constellation networks. In addition, the optimized routing method can avoid congestion and greatly improve the performance of mega-constellation networks with low network local-state awareness overhead.
Engineering,Computer Science
What problem does this paper attempt to address?