信息网络安全 ›› 2020, Vol. 20 ›› Issue (7): 53-59.doi: 10.3969/j.issn.1671-1122.2020.07.006

• 技术研究 • 上一篇    下一篇

面向云平台虚拟层的安全态势评估关键技术研究

余晴1, 郑崇辉2, 杜晔3()   

  1. 1.北京交通大学计算机与信息技术学院,北京 100044
    2.中国科学院大学杭州高等研究院,杭州 310024
    3.北京交通大学国家保密学院,北京 100044
  • 收稿日期:2020-04-10 出版日期:2020-07-10 发布日期:2020-08-13
  • 通讯作者: 杜晔 E-mail:ydu@bjtu.edu.cn
  • 作者简介:余晴(1995—),女,福建,硕士研究生,主要研究方向为云安全、态势感知等|郑崇辉(1970—),男,北京,研究员,博士,主要研究方向为保密管理、信息安全等|杜晔(1978—),男,黑龙江,教授,博士,主要研究方向为保密技术、网络攻防等
  • 基金资助:
    国家自然科学基金(61672092)

Research on Key Technologies of Security Situation Assessment for the Virtual Layer of Cloud Platform

YU Qing1, ZHENG Chonghui2, DU Ye3()   

  1. 1. School of Computer and Information Technology, Beijing Jiaotong University, Beijing 100044, China
    2. Hangzhou Institute For Advanced Study, UCAS, Hangzhou 310024, China
    3. National School of Secrecy, Beijing Jiaotong University, Beijing 100044, China
  • Received:2020-04-10 Online:2020-07-10 Published:2020-08-13
  • Contact: Ye DU E-mail:ydu@bjtu.edu.cn

摘要:

随着针对云平台的攻击和破坏行为的日益增多,云平台的安全保障机制也由传统的被动防护转为主动防御。态势评估是一种可主动分析评估云平台当前安全风险状态的方法,是态势感知全过程的关键一环。文章面向云平台中大量部署的虚拟机,在分析提取虚拟层安全态势评估要素的基础上,提出一种改进的自适应遗传模拟退火算法OAGSAA,并应用于BP神经网络,可有效对云平台虚拟层安全状态进行分析评估。仿真实验结果显示,该方法具有较高预测准确率和收敛速度,并可避免陷入局部最小值。

关键词: 云平台, 虚拟层, 态势评估, 自适应遗传模拟退火算法, BP神经网络

Abstract:

With the increasing attacks and destructions against cloud platforms, the security guarantee mechanism of cloud platforms has also changed from traditional passive protection to active defense. Situation assessment is a method that can actively analyze and evaluate the current security risk status of the cloud platform, and is a key part of the whole process of situation awareness. This paper aims at a large number of virtual machines deployed in cloud platforms, on the basis of analysis and extraction of virtual layer security situation assessment elements, an improved adaptive genetic simulated annealing algorithm OAGSAA is proposed, and applied to BP neural network, which can effectively analyze and evaluate the security status of the virtual layer. Simulation experiment results show that the method has higher prediction accuracy and convergence speed, and can avoid falling into the local minimum.

Key words: cloud platform, virtual layer, situation assessment, adaptive genetic simulated annealing algorithm, BP neural network

中图分类号: