Adding contextual information to intrusion detection systems using fuzzy cognitive maps
conference contributionposted on 18.04.2016, 13:53 by Francisco Aparicio-Navarro, Kostas Kyriakopoulos, David J. Parish, Jonathon Chambers
In the last few years there has been considerable increase in the efficiency of Intrusion Detection Systems (IDSs). However, networks are still the victim of attacks. As the complexity of these attacks keeps increasing, new and more robust detection mechanisms need to be developed. The next generation of IDSs should be designed incorporating reasoning engines supported by contextual information about the network, cognitive information from the network users and situational awareness to improve their detection results. In this paper, we propose the use of a Fuzzy Cognitive Map (FCM) in conjunction with an IDS to incorporate contextual information into the detection process. We have evaluated the use of FCMs to adjust the Basic Probability Assignment (BPA) values defined prior to the data fusion process, which is crucial for the IDS that we have developed. The results that we present verify that FCMs can improve the efficiency of our IDS by reducing the number of false alarms, while not affecting the number of correct detections.
This work was supported by the Engineering and Physical Sciences Research Council (EPSRC) Grant number EP/K014307/1 and the MOD University Defence Research Collaboration in Signal Processing.
- Mechanical, Electrical and Manufacturing Engineering