Vulnerability Analysis in Attack Graphs Using Conditional Probability
Pankaj Chejara1, Urvashi Garg2, Gurpreet Singh3
1Pankaj Chejara, Department of Computer Science, Sharda University, Noida, India.
2Urvashi Garg, Department of Computer Science, Lovely Professional University, Jalandhar, India.
3Gurpreet Singh, Department of Computer Science, Lovely Professional University, Jalandhar, India.
Manuscript received on November 25, 2013. | Revised Manuscript received on December 15, 2013. | Manuscript published on December 30, 2013. | PP: 18-21 | Volume-3, Issue-2, December 2013. | Retrieval Number: B2335123213/2013©BEIESP
Open Access | Ethics and Policies | Cite
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Computer networks have become an essential part of almost every organization. These organizations spend a lot of time and money to secure their networks from intruders and attackers. As the need of computers increased, need for network security increased correspondingly. Attackers are always trying to find weakness in network which can be used to break into the network known as vulnerability. So network administrator needs to patch vulnerabilities to thwart attacker from achieving their goal. As new vulnerability are discovered daily, it is very hard to patch every vulnerability in network but if riskier vulnerabilities get patched, risk level can be reduced significantly. Vulnerability score gives insight into the behavior of vulnerability. These scores make security analyst’s work easier to some extent. But these scores do not include collective effect of vulnerabilities. A number of vulnerability scanners are available, which provide complete vulnerability details about host. These vulnerability details give analyst a good idea about to which extent the network security can be compromised, but does not give complete view of network vulnerability. Attack graph provides solution to this problem. Attack graph is set of nodes and edges where node represents attacker’s state and edge represent possible transition among attacker’s state. This technique gives path that can be followed by attacker to gain network’s resources. In the network attack graph depict how vulnerability affect network in conjunction with other vulnerabilities. Some vulnerability may not be riskier alone but when chained with some other, it can compromise the security of network. These attack graphs are important security tools to find out such vulnerabilities also. In this paper, we have developed an technique to provide scores to each path in attack graph so as to analyze, which path is to be patched first to remove the risk of attack. These scores are based on conditional probability method.
Keywords: Attack Graphs, Attack Model, Vulnerability Score, Attack Sequence.