多状态空间信息网络拓扑生成优化算法

依据空间信息网络（SIN）高动态性的特点，并考虑卫星工作的多状态特性，兼顾星间通信时延和拓扑抗毁性的要求，研究了多状态下空间信息网络拓扑生成及动态优化的问题。根据卫星星座的周期性，建立了一种卫星网络的拓扑周期表。综合卫星的可视性和连接度等约束条件，以网络平均和最大时延作为通信性能的优化目标，建立拓扑的多目标优化模型。提出一种改进的多目标模拟退火（IMOSA）算法，求解全局时延最优的卫星拓扑，并在考虑多状态情况下对链路进行优化，以满足网络高动态性。最后基于具有66颗低轨（LEO）的铱星星座进行仿真，研究表明：针对多状态条件下的铱星星座，该算法最大化减小了通信时延，得到抗毁性良好的拓扑结构，通信性能较之原有静态拓扑明显得到改善。

Topological generation and optimization method in multi-state space information network

The topological generation and optimization of Space Information Network (SIN) in multi-state is studied based on the high dynamicity in SIN, the multi-state characteristics of satellites, and the requirements of communication time-delay and topological invulnerability. According to the periodicity of satellite constellation, a periodic table of satellite constellation topologies is established. Considering the constraints of satellite visibility and connectivity degree, a multi-objective optimization model of network topology is set up, with network average and maximal delay as optimization targets. An Improved Multi-Objective Simulated Annealing (IMOSA) algorithm is proposed and used for solving the structure of global optimization topology. Considering the multi-state characteristics, the time-delay of topology is optimized to satisfy the high dynamicity. The final simulation based on the Iridium Constellation with 66 Low Earth Orbit (LEO) satellites in multi-state situation shows that, the algorithm can maximize the decreasing of communication delay and obtain favorable invulnerability topology structures. Compared with the previous static typology, the communication ability of topology generated by the algorithm improved significantly.