Skip to search boxSkip to navigationSkip to main content

A graph theoretical approach to network vulnerability analysis and countermeasures

Research Output: Contribution to journal Article Peer-review

Abstract

Computer networks are certainly vulnerable as long as they deliver services from different machines. An attack graph is a security model representing the chains of vulnerability exploits in a network displays the ways an attacker can compromise a network or host. A number of researchers have admitted attack graph visual complications and a large amount of source data must be assembled to accurately build an attack graph, the difficulty scaling to large, enterprise-size networks with tens of thousands of hosts and the lack comprehensive understanding. Information on vulnerabilities is present in public vulnerability databases, such as the National Vulnerability Database and Nessus. But current attack graph tools are reserved to only limited attributes. The automatic formation of vulnerability information has been troublesome and vulnerability descriptions were created by hand or based on limited information. Much vulnerability has still not been discov-ered and many others without patches or solutions Our approach to developing a cost metric exploits the Markov’s model using combinations well known vulnerabilities (the Common Vulnerability Scoring System, CVSS) and Risk Assessment Values (RAV) and using ranking algorithms (similar to V. Mehta et al. 2006 and kijsanayothin, 2010) but instead of using vulnerabilities. For each host we have developed a cost rank Markov’s model reducing the complexity in the attack graph, representing the network topology and dipping the problem of visibility.

Publication Information

Output type

Research Output: Contribution to journal Article Peer-review

Original language

English

Pages from-to (Number of pages)

Pages .13-18

Journal (Volume, Issue Number)

International Journal of Computer Applications (Volume 1)

Publication milestones

  • Published - 01/01/2011

Publication status

Published - 01/01/2011

ISSN

0975-8887

External Publication IDs

  • handle.net: 10547/270777

Access to documents