摘要
基于云模型的虚拟设备业已成为非法访问和恶意攻击的常见目标,虚拟设备的大数据安全防护至关重要。本文对云计算的攻击检测模型及虚拟设备的安全分析方法进行研究,建立了虚拟设备的攻击检测模型,针对属性特征分别给出了逻辑回归分类器的训练过程、边缘概率有效转换为条件概率的过程以及贝叶斯网络因子图的生成过程,提出了基于逻辑回归与置信传播的数据安全分析(Data Security Analytics,DSA)算法。仿真实验结果表明,随着虚拟机数量的增加,DSA算法的中值检测时间均保持在0.06 ms左右,体现出该算法良好的性能。
The virtual device based on cloud model has already become a common target to carry out illegal access and malicious attack,so it is very important for the security protection of the big data of the virtual machines.In this paper,the attack detection model of cloud computing and the security analysis method of virtual machines are studied deeply.First,an attack detection model of virtual machines is established.Secondly,we respectively give the training process of the logistic regression classifier(LRC),the process of the effective transfer of the edge probability to the conditional probability and the factor graphs of Bayesian network.A novel algorithm called Data Security Analytics(DSA)based on LRC and the belief propagation(BP)is proposed.The simulation results show that with the increase of the number of virtual machines,the median detection time of DSA algorithm is about 0.06 milliseconds,which reflects the good performance of the algorithm.
作者
李兵
LI Bing(College of Applied Engineering,Henan University of Science and Technology,Sanmenxia,Henan,472000,China)
出处
《石河子大学学报（自然科学版）》
CAS
北大核心
2020年第3期390-396,共7页
Journal of Shihezi University(Natural Science)
基金
河南省重点科技攻关基金项目(14210221042)
河南省教育厅科学技术研究重点基础研究计划基金项目(14A520048)。
关键词
虚拟设备
网络防御
安全分析方法
攻击检测
逻辑回归
置信传播
Virtual machines
Network defense
Security analysis method
Attack detection
Logistic regression
Belief propagation
