An Improved Monte Carlo Graph Search Algorithm for Optimal Attack Path Analysis

Hui Xie, Kun Lv, Changzhen Hu

科研成果: 书/报告/会议事项章节会议稿件同行评审

摘要

The problem of optimal attack path analysis is one of the hotspots in network security. Many methods are available to calculate an optimal attack path, such as Q-learning algorithm, heuristic algorithms, etc. But most of them have shortcomings. Some methods can lead to the problem of path loss, and some methods render the result un-comprehensive. This article proposes an improved Monte Carlo Graph Search algorithm (IMCGS) to calculate optimal attack paths in target network. IMCGS can avoid the problem of path loss and get comprehensive results quickly. IMCGS is divided into two steps: selection and backpropagation, which is used to calculate optimal attack paths. A weight vector containing priority, host connection number, CVSS value is proposed for every host in an attack path. This vector is used to calculate the evaluation value, the total CVSS value and the average CVSS value of a path in the target network. Result for a sample test network is presented to demonstrate the capabilities of the proposed algorithm to generate optimal attack paths in one single run. The results obtained by IMCGS show good performance and are compared with Ant Colony Optimization Algorithm (ACO) and k-zero attack graph.

源语言英语
主期刊名Proceedings - 17th IEEE International Conference on Trust, Security and Privacy in Computing and Communications and 12th IEEE International Conference on Big Data Science and Engineering, Trustcom/BigDataSE 2018
出版商Institute of Electrical and Electronics Engineers Inc.
307-315
页数9
ISBN(印刷版)9781538643877
DOI
出版状态已出版 - 5 9月 2018
活动17th IEEE International Conference on Trust, Security and Privacy in Computing and Communications and 12th IEEE International Conference on Big Data Science and Engineering, Trustcom/BigDataSE 2018 - New York, 美国
期限: 31 7月 20183 8月 2018

出版系列

姓名Proceedings - 17th IEEE International Conference on Trust, Security and Privacy in Computing and Communications and 12th IEEE International Conference on Big Data Science and Engineering, Trustcom/BigDataSE 2018

会议

会议17th IEEE International Conference on Trust, Security and Privacy in Computing and Communications and 12th IEEE International Conference on Big Data Science and Engineering, Trustcom/BigDataSE 2018
国家/地区美国
New York
时期31/07/183/08/18

指纹

探究 'An Improved Monte Carlo Graph Search Algorithm for Optimal Attack Path Analysis' 的科研主题。它们共同构成独一无二的指纹。

引用此