基于半马尔可夫过程的虚拟网络生存性模型  

Survivability Model of Virtual Network Based on Semi-Markov Process

作　　者：苏玉泽 孟相如[1] 康巧燕[1] 韩晓阳 SU Yu-ze;MENG Xiang-ru;KANG Qiao-yan;HAN Xiao-yang(Information and Navigation College,Air Force Engineering University,Xi'an 710077,China)

机构地区：[1]空军工程大学信息与导航学院

出　　处：《科学技术与工程》2019年第24期260-267,共8页Science Technology and Engineering

基　　金：国家自然科学基金(61871313)资助

摘　　要：针对虚拟网络生存性缺乏准确的定量描述,提出了一种基于半马尔可夫过程(semi-Markov process,SMP)的虚拟网络生存性模型。根据虚拟网络运行特征,重点引入虚拟网络重构状态以及认知状态,构建一种优化的虚拟网络状态转移图,并引入半马尔可夫理论,建立虚拟网络状态转移概率方程,构建基于SMP的虚拟网络生存性模型,最后通过仿真实验验证了虚拟网络生存性对各参量的敏感度以及不同类型虚拟网络的生存性特点。仿真表明,降低网络部件发生故障的概率,提高虚拟网络攻击识别概率和虚拟网络重构成功概率可以显著提高虚拟网络生存性;在映射阶段进行资源备份可以显著增强军事虚拟网络生存性。Aiming at the problem that there was lack of accurate quantitative description for virtual network survivability,the survivability model of virtual network was proposed based on semi-Markov process.According to the operation characteristic of virtual network,an optimized state transition diagram of virtual network was constructed based on the introduction of reconfiguration state and cognitive state.The theory of semi-Markov was introduced to establish the probability equation of system state and the survivability model of virtual network based on SMP was proposed.The sensitivity of the virtual network survivability to each parameter and the survivability characteristic of different virtual networks were simulated.Simulation experiments show that decreasing the network failure probability or increasing the attack recognition probability and reconfiguration probability can improve the virtual network survivability significantly.The resource backup in embedding stage can enhance the survivability of military virtual network.

关 键 词：虚拟网络 生存性模型 半马尔可夫过程 重构策略 认知状态 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]