Issues of Internet of Things (IoT) and an Intrusion Detection System for IoT Using Machine Learning Paradigm

Abstract

Internet of Things (IoT), Internet of Everything or Internet of Intelligent Things are few of the buzzwords that we hear almost every day in the Information Technology sector. The whole world is entering a new era where everything will be communicating with each other by means of IoT. IoT aims to bring everything under a single communication infrastructure, giving us the power to control anything from anywhere. Powerful data analytic tools are being deployed to collect and analyze data from those connected devices to make a drastic change in our daily life. Researchers around the globe has made a projection that the quantity of associated IoT gadgets will be just about 100 billion and the financial impact will be about $11 trillion before the finish of 2025. However, there are a significant number of challenges that has already grabbed the attention besides few other emerging challenges. As the number of IoT devices increases, new security and privacy challenges will be introduced. In this paper, we have discussed various issues of IoT ecosystem. Then, we have built an IoT smart-home and developed an adaptive intrusion detection system. The system employs Machine Learning strategies and is capable of detecting simple form of Denial of Service attacks. We have applied proposed method on our IoT test bed with commercially available IoT devices and we have shown that the proposed method is successful in identifying network scanning probing.