Intrusion Detection Using Hidden Markov Model and XGBoost Algorithm

Abstract

Security has been widely concerned and recognized as a critical issue in wireless communication networks recently, because the openness of the wireless medium allows unintended receivers i. e. intruders to potentially eavesdrop on the transmitted messages. Unauthorized access by an intruder can be monitored by Intrusion detection system. Machine learning algorithms such as Hidden Markov Model and Extreme gradient boost algorithm can be used for intrusion detection based on CICIDS dataset. Based on dataset, algorithms create classifiers of signatures of particular attack. These trained classifiers are tested against user data for intrusion detection. System reports attack in network. Here, XGBoost classifier gives higher accuracy compared to HMM classifier.