【摘要】 针对网络攻击事件难以实时准确反映，导致光网络安全态势预测效果较差的问题，提出一种基于贝叶斯攻击图的光网络安全态势预测方法。利用漏洞扫描软件对网络中的缺陷进行扫描，获得初始数据，基于相关数据建立贝叶斯攻击图对网络中的不确定参数分析，计算各个节点的先验概率和后验概率，确定各个节点的危险等级，设定遗传算法的遗传算子与迭代次数，对特征数据降维处理，采用随机森林方法选择最优特征子集，对局部条件概率计算，实现光网络安全态势预测。实验结果表明，设计方法能够在短时间内就实现光网络安全态势的准确预测，满足方法设计需求。更多还原

【Abstract】 Aiming at the problem that network attack events are difficult to accurately reflect in real time, which leads to poor prediction effect of optical network security situation, this paper proposes a method of optical network security situation prediction based on Bayesian attack graph. The vulnerability scanning software is used to scan the defects in the network, obtain the initial data, establish a Bayesian attack graph based on the relevant data, analyze the uncertain parameters in the network, calculate the prior probability and posterior probability of each node, determine the risk level of each node, set the genetic operator and iteration times of the genetic algorithm, reduce the dimension of the characteristic data, and select the best feature subset by using the random forest method, The local conditional probability is calculated to realize the security situation prediction of optical network. The experimental results show that the design method can accurately predict the security situation of optical network in a short time and meet the design requirements of the method.更多还原