电子政务网的网络安全评估技术研究

发布时间：2018-07-25 10:38

【摘要】：随着经济和互联网技术的快发发展,电子政务在政府信息化管理改革中应运而生,伴随而来的是电子政务网的安全性保障。网络安全评估技术针对网络系统的漏洞检测,网络态势感知,攻击入侵检测等方面具有优势,现有的网络安全评估技术在处理确定性数据时有较好的评估效果,针对不确定性信息态势评估则不尽人意。信息网络系统由于其结构的复杂性,物理设备多样性产生的数据量大,数据信息密度低,给安全评估带来了困难,一方面安全评估需要保证准确性,另一方面安全评估要有实时性,然而,现有的安全评估方法要么通过牺牲准确性达到高效性的目的,要么降低效率获得高准确性。因此,网络安全评估方法中满足评估结果准确性和实时性成为一种挑战。针对上述挑战,本文提出了一种基于灰色关联分析与D-S证据理论的网络安全评方法。首先,由于不同评估指标对于网络态势的影响并不相同,利用层次法来确定不同评估指标的权重。其次,考虑到不同指标对不同等级的隶属度差别,确定了各评估指标对评估等级的隶属度函数,定义了放大加权隶属度函数来降低整体不确定性。再者,D-S证据理论中基本概率分配是一个难以确定的问题,为了减少人为主观因素对基本概率分配的差异性影响,提出利用灰色关联分析法来解决基本概率分配问题,然后通过D-S证据理论合成规则对各Mass函数进行合成,逐步降低不确定性,评估结果根据置信函数大小确定。通过实验表明,该方法降低了评估结果的不确定性,提高了准确性和效率。网络安全态势预测是在网络安全评估的基础上对未来网络安全态势的估计,对于网络安全态势的把握有指导性作用。网络安全态势具有突发性、波动性,因此网络安全态势势预测方法要有良好的处理非线性问题的能力,对此,提出了基于GM(1,1)幂模型的网络安全态势预测方法,GM(1,1)幂模型具有优秀的处理非线性问题的能力,对于样本数量要求不高,实用性较好,最后通过实验对比,本文所用的网络安全态势预测方法较具有较好的预测精度和较低的误差率。
[Abstract]:With the rapid development of economy and Internet technology, e-government emerged as the times require in the government information management reform, accompanied by the security of e-government network. The network security assessment technology has advantages in the aspects of vulnerability detection, network situation awareness, attack intrusion detection and so on. The existing network security assessment technology has a good evaluation effect when dealing with deterministic data. The situation assessment of uncertain information is unsatisfactory. Because of the complexity of its structure, the large amount of data produced by the diversity of physical equipment and the low density of data information, the information network system brings difficulties to the security assessment. On the one hand, the security assessment needs to ensure the accuracy. On the other hand, security assessment should be real-time. However, the existing security assessment methods can achieve high efficiency at the expense of accuracy, or reduce efficiency to achieve high accuracy. Therefore, it is a challenge to satisfy the accuracy and real-time of network security evaluation. In view of the above challenges, this paper presents a network security evaluation method based on grey correlation analysis and D-S evidence theory. Firstly, because the influence of different evaluation indexes on network situation is different, the hierarchical method is used to determine the weight of different evaluation indicators. Secondly, considering the difference of membership degree of different indexes to different grades, the membership function of each evaluation index to evaluation grade is determined, and the enlarged weighted membership function is defined to reduce the overall uncertainty. Furthermore, in D-S evidence theory, the basic probability distribution is a difficult problem. In order to reduce the difference of artificial subjective factors on the basic probability distribution, the grey correlation analysis method is proposed to solve the basic probability distribution problem. Then each Mass function is synthesized by D-S evidence theory synthesis rule, and the uncertainty is reduced step by step. The evaluation results are determined according to the size of the confidence function. Experiments show that this method reduces the uncertainty of the evaluation results and improves the accuracy and efficiency. The prediction of network security situation is based on the evaluation of network security, and it can guide the grasp of network security situation. The network security situation is sudden and fluctuating, so the network security situation forecasting method should have a good ability to deal with nonlinear problems. In this paper, a method of network security situation prediction based on GM (1K1) power model is presented. The power model has excellent ability to deal with nonlinear problems, and it has low requirement for sample size and good practicability. Finally, it is compared by experiments. The network security situation prediction method used in this paper has better prediction accuracy and lower error rate.
【学位授予单位】：河南工业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP393.08

【参考文献】