基于模块化的动态贝叶斯网络的安全态势评估模型

摘要

To solve the security situation evaluation problem of poor timeliness caused by reconstruc‐tion of the network information system ,a situation assessment model based on modular dynamic Bayesian networks is established by combining hierarchical network with modular network .Firstly ,a conditional independence test is used to determine whether there is a connection between nodes and dynamic Bayesian network modules applied for subsystems are constructed .Then ,an overall dynamic Bayesian network with several modules and independent nodes is built .The network probability pa‐rameters are obtained by utilizing the constrained recursive algorithm .Finally ,simulation tests vali‐date the model′s correctness .%针对由网络信息系统结构变动引起的安全态势评估时效性差的问题，结合层次网络和模块网络的优点，建立了基于模块化动态贝叶斯网络的态势评估模型。首先，通过判断节点间是否存在有向连接，建立子系统的动态贝叶斯网络模块单元；然后，由若干模块单元和独立节点构建整体的动态贝叶斯网络，并利用约束递归算法学习网络概率参数；最后，通过仿真测试验证该模型的正确性。