网络安全防御态势有效预测仿真

发布时间：2018-08-02 13:41

【摘要】：对网络安全防御态势的有效预测,能够保障网络的安全稳定的运行状态。对安全防御态势预测,需要计算出当前的网络防御系数和态势变化值,来完成有效预测。传统方法融合D-S证据理论,得到安全可信度区间和相应防御系数,但忽略了对防御态势的变化趋势的分析,导致预测结果不准确。提出基于模糊动态贝叶斯网络安全防御态势预测模型。将模糊理论与叶贝斯理论相融合得到在当前防御态势下发生攻击的后验概率,统计网络历史防御的数据样本,组建历史网络安全防御态势下发生攻击时间序列,计算出当前的网络安全防御态势变化值、攻击发生概率、网络防御系数;组建网络安全防御态势的评估矩阵,采用遗传算法进行网络安全态势预测模型的优化,依据优化结果组建网络安全防御态势预测模型。仿真结果表明,所提模型预测精确度高,为保障网络安全稳定地运行提供了有力的依据。
[Abstract]:The effective prediction of the network security defense situation can ensure the safe and stable running state of the network. For the prediction of security defense situation, it is necessary to calculate the current network defense coefficient and situation change value in order to complete the effective prediction. The traditional method combines D-S evidence theory to obtain the interval of security reliability and the corresponding defense coefficient, but the analysis of the trend of the change of defense situation is ignored, which leads to the inaccuracy of the prediction results. A security defense situation prediction model based on fuzzy dynamic Bayesian network is proposed. By combining fuzzy theory with Yebeth theory, the posteriori probability of attack under the current defense situation is obtained, the data samples of network historical defense are counted, and the attack time series under the situation of historical network security defense is established. Calculate the current network security defense situation change value, attack occurrence probability, network defense coefficient, set up the network security defense situation evaluation matrix, use genetic algorithm to optimize the network security situation prediction model, According to the optimization results, the network security defense situation prediction model is established. The simulation results show that the proposed model has high prediction accuracy and provides a powerful basis for the safe and stable operation of the network.
【作者单位】：内蒙古财经大学计算机学院;
【基金】：国家自然科学基金(61563038)
【分类号】：TP393.08

【相似文献】