Building NIDS for IoT Network using Ensemble Approach

The term ‘IoT’ is a trendy word in current era due to its usage in day to day live of human beings. Its applications are widespread across many areas such as home, agriculture, health care, transportation, etc., Due to the vast development of IoT services, cyber-attacks against IoT application and devices also increased. Thus, Security and Privacy are two major concerns in IoT environment. Building a Network Intrusion Detection System (NIDS) is one of the ways for securing the IoT system against cyber-attacks. So, an Ensemble based Network Intrusion Detection System using machine learning algorithms has been proposed to detect botnet attacks against HTTP and DNS protocol in network traffic. The goal of this proposed work is to higher the detection rate and decrementing the false positive rate. UNSW_NBl5 dataset was used for evaluating the performance of the proposed NDS. In this work, an Io T environment architecture using fog node proposed and the performance of various classifiers included in the framework compared with the ensemble methodbased on detection rate and false positive rate.