Design of a Simulation Framework to Evaluate Trust Models for Collaborative Intrusion Detection

Different trust models have been developed for dealing with possible dishonest behavior and attacks from malicious peer Intrusion Detection Systems (IDSs) in a collaborative Intrusion Detection Network (IDN). For evaluating and comparing these models, this paper introduces a simulation framework that incorporates different components namely expertise model, deception model, attack model, and evaluation metrics. The proposed framework offers flexibility for users to adjust the simulation parameters according to their needs. We then compare three existing trust models in this domain to demonstrate the effectiveness of our framework when used in analyzing their efficiency, robustness and scalability.