Authors:
Hanwen Zhang
1
;
Wenyong Wang
1
;
Lisheng Huang
1
;
Junrui Wu
1
;
Fengjun Zhang
2
and
Kai Shi
2
Affiliations:
1
School of Computer Science and Engineering, University of Electronic Science and Technology of China, Chengdu, China
;
2
The 30thInstitute of CETC, China Electronics Technology Cyber Security Co., Ltd, Chengdu, China
Keyword(s):
Cybersecurity, Attack Graph, Intrusion Detection, Virtual Networks.
Abstract:
Securing a computer networking system requires the ability to gather and organise information about potential vulnerabilities existing in the system. One way of utilising the information above is to generate an attack graph of all possible attack paths. Current attack graph generation methods reach scalability issue with the growth of network devices and links, and one solution is to correlate attack graph with intrusion detection systems. However, correlation techniques are rarely studied especially on generating attack graphs on virtual computer networks, as correlations are inflexible to be integrated to existing attack graph generators. Previously we proposed mAGG, an attack graph generation framework on virtual networkings; and LSAFID, an intrusion detection system based on doc-word. In this paper, we propose a new method for correlating intrusion detection algorithm for attack graph generation on virtual networkings. Our new proposed method is flexible in network architectures
and functionalities, and shortens the scale of generated attack graph.
(More)