iot-security-anomaly-based-intrusion-detection-system

This is a research paper based on anomaly based intrusion detection systems used in iot systems. This surveys similar technologies used along with details of system proposed by Shadi A et. al. This system helps in automatically identifying suspicious IOT devices connected to the network. It consist of the training phase where a profile of normal behaviors is built and testing phase where current traffic is classified as attack or normal with the profile created in the training phase. Machine learning ensemble model has been used, including several classifiers including J48, Meta Pagging, Random Forest, REPTree, AdaBoostM1, Decision Stump and Naïve Bayes. It is trained on the popular dataset of NSL-KDD.