基于韧性度的低轨卫星通信网络抗毁性度量及优化

低轨卫星通信网络的抗毁性是描述网络安全可靠的有效工具,在网络体系结构设计和路由策略等领域得到了广泛的应用。根据低轨卫星通信网络中卫星在轨道平面内移动,需要不断进行切换的特点,从建立抗毁性测度模型以及网络抗毁性优化两个角度来评估和提高网络抗毁性,提出一种基于韧性度的低轨卫星通信网络抗毁性度量方法。通过对移动模型以及切换模型的结构分析,对每种结构以一定概率出现的低轨卫星通信网络,应用韧性度函数,求得网络在某个时刻及某一段时间段内的抗毁性,并针对切换模型的不足之处进行优化,用赋权韧性度来体现优化的效果,得到了优化后的网络抗毁性。以铱星系统为应用实例进行仿真,结果表明:任意时刻网络的抗毁性跟拓扑结构的韧性度值有关,并且是一种线性关系,即随着韧性度的增加,其抗毁性也增加。通过对铱星通信系统切换模型的优化,网络的抗毁性与平均抗毁性都得到了提升,说明本文所构建模型的有效性和实用性。

Measurement and Optimization of Invulnerability of Low-orbit Satellite Communication Networks Based on Resilience

The invulnerability of low-orbit satellite communication networks is an effective tool to describe network security and reliability. It has been widely used in network architecture design and routing strategies. According to the characteristics of satellites moving in the orbital plane in the low-orbit satellite communication network, it is necessary to continuously switch. From the establishment of the invulnerability measurement model and the network invulnerability optimization, the network invulnerability is evaluated and improved. A measure of the invulnerability of low-orbit satellite communication networks based on resilience is proposed. Through the structural analysis of the mobile model and the switch model, the resilience degree function is applied to the low-orbit satellite communication network with a certain probability for each structure, and the invulnerability of the network at a certain moment and a certain period of time is obtained. The deficiencies of the switching model are optimized, and the weighted resilience is used to reflect the optimization effect, and the optimized network invulnerability is obtained. Simulation with Iridium system as an application example shows that the invulnerability of the network is related to the resilience value of the topology at any time, and it is a linear relationship, that is, as the resilience increases, the invulnerability increases. Through the optimization of the switching model of the Iridium communication system, the invulnerability and average invulnerability of the network are improved, which shows the validity and practicability of the model constructed in this paper.