Intrusion Detection in IoT using Machine Learning (CNN Focal)

IoT has become popular since the number of devices connected to the Internet increased tremendously. It is highly regarded because of its ability to integrate new capabilities for various automation and interconnectivity functions. At the same time, the proliferation of IoT devices has created some weaknesses that allow attacks to succeed on IoT networks. This paper focuses research of intrusion detection systems of IoT Network using various machine learning techniques. In order to mitigate intrusions in real-time, supervised and unsupervised algorithms, amongst others, are analyzed in the framework of the research. The task entails building a resilient architecture that incorporates network traffic patterns, behaviors of devices and anomalous patterns to secure IoT solutions. The architecture is tested on benchmark datasets and its performance in minimizing false alarms and improving detection system performance is reported in the paper. All in all, the authors are certain now that implementation of IoT in various smart environments is secured and will advance secure interconnected devices.