中文摘要：

针对态势感知模型无法准确捕捉网络系统信息变化的弊端,提出了基于模糊动态贝叶斯网络的态势评估模型。通过对态势要素的状态进行模糊和概率化处理,构建了态势感知和态势评估两个子模型。模型归一化各级态势要素的状态并权值相加,按值划分态势区间,推演态势等级,形成态势处置方案为网络空间防御提供了辅助决策。将概率及观测数据输入模型中进行仿真评估,并与静态贝叶斯网络模型和基于Hopfield神经网络的评估模型进行结果对比,实验结果表明,基于模糊动态贝叶斯网络模型的评估结果综合了更多节点关系和观测信息,在对抗中能逐渐适应对抗强度,防御水平逐渐提高,随着对抗时间的延续,能较好地反映网络防御作战态势的变化规律,同时能有效对目标网络威胁等级进行分类,对态势结果和趋向进行准确分析预估。

英文摘要：

For situational awareness model can not accurately capture network information changes drawbacks, a situation assessment model based on fuzzy dynamic Bias network was proposed in this paper. Two submodels of situation awareness and situation assessment were constructed by using the fuzzy and probabilistic processing of situation factors. The state of the model was normalized at all levels of state and the weight was added. According to the value of the situation, the situation was divided, and the situation of the deduction was deduced. Probability and observation data were input to establish a model for simulation and evaluation, and with the static Bayesian network model and evaluation model based on Hopfield neural network by comparing the simulation results. Experimental results showed that： more comprehensive relationship between the nodes and observations based on information to assess the results of fuzzy dynamic Bayesian network model could gradually adapt to the confrontation against strength, and gradually increased the level of defense. With the continuation of the fight against time, which could better reflect variation combat cyber defense posture, while effectively target network threat level to classify the results and trends of the trend estimate for accurate analysis.