《计算机应用研究》|Application Research of Computers

基于特征选择的虚拟化系统语义鸿沟桥接研究

Research on bridging semantic gap in virtualization system based on feature selection

免费全文下载 (已被下载 次)  
获取PDF全文
作者 娄睿,蒋烈辉
机构 1.信息工程大学 四院,郑州 450002;2.数学工程与先进计算国家重点实验室,郑州 450002
统计 摘要被查看 次,已被下载
文章编号 1001-3695(2019)05-053-1533-07
DOI 10.19734/j.issn.1001-3695.2017.12.0761
摘要 虚拟化系统的强隔离性质在为安全机制部署提供可靠环境的同时,也引入了语义鸿沟问题。针对现有研究普遍依赖的软件体系结构信息、数据结构和控制流容易被窜改,采用的检测算法在客户机状态识别方面效率较低等问题,设计了特征构造和窗口标记的方法对虚拟机数据进行预处理,以满足实施数据挖掘的必要条件,建立了基于特征选择的虚拟化系统语义鸿沟桥接模型,能够仅依赖硬件体系结构数据构建虚拟机执行模式并进行安全检测。实验结果表明,所设计的系统模型能够筛选出关键的虚拟机特征,并有效地识别出客户机异常行为,提高语义鸿沟的桥接效率,为处理语义鸿沟问题提供了一种可行方案。
关键词 虚拟化安全; 语义鸿沟; 机器学习; 特征构造; 特征选择; 异常检测
基金项目 国家自然科学基金资助项目(61572516)
河南省科技攻关项目(162102210033)
本文URL http://www.arocmag.com/article/01-2019-05-053.html
英文标题 Research on bridging semantic gap in virtualization system based on feature selection
作者英文名 Lou Rui, Jiang Liehui
机构英文名 1.Fourth Department,Information Engineering University,Zhengzhou 450002,China;2.State Key Laboratory of Mathematical Engineering & Advanced Computing,Zhengzhou 450002,China
英文摘要 The strong isolation property of the virtualization system introduces the semantic gap problem while providing a relia-ble environment for the deploying the security mechanism. Current research generally relies on the information of software architecture which is not reliable, for the data structures and control flows are easy to be illegally manipulated. And the detection algorithm employed in related research has the low efficiency in identification of guest state. For these problems, this paper designed the feature construction and window mark to preprocess the captured data so as to meet the necessary conditions of carrying out data mining, and then established the semantic gap bridging model of virtualization system based on feature selection, which could build the execution mode of virtual machine and carried out the security detection only relying on the hardware architecture data. Test results show that proposed model can screen out the key features of virtual machine and effectively identify the abnormal behavior of guest system, which lead to the efficiency improvement of bridging semantic gap. This scheme provides a feasible solution for dealing with the problem of semantic gap.
英文关键词 virtualization security; semantic gap; machine learning; feature construction; feature selection; anomaly detection
参考文献 查看稿件参考文献
 
收稿日期 2017/12/18
修回日期 2018/1/29
页码 1533-1539
中图分类号 TP309
文献标志码 A