Enhancing MAC Security Model with Meta-policy Approach Using an Intelligent Anomaly Based HIDS

Abstract

This paper exhibits a new structure based on a meta-policy approach using an intelligent anomaly based host based intrusion detection system, These access control decisions are decided in agreement with a meta-policy. The combination of policy and host based intrusion detection system approach depends on operating systems integrating with mandatory access control. The proposed structure controls a extended set of attacks and provides the management of the system policy, compared to other existing access control approaches and policy based host based intrusion detection system techniques. Items are given about methods used for the meta-policy, and implementation of the structure. We used genetic algorithm to try to select what was the best set of features and chose a Bayesian network, that is trained by using the selected system log on operating system.