A Comprehensive Review on Intrusion Detection in Edge-Based IoT Using Machine Learning

Smart environment is the need of today’s world. Smart environment means smart in every field like smart gadgets, smart cities, smart vehicles, smart healthcare systems and many more. The main aim of smart environment is to provide quality life and easiness to people and this can be achieved with the help of Internet of Things (IoT). Internet of Things is the web of devices that are connected with the help of Internet and smart in nature. As IoT is totally dependent on Internet, security and privacy is the primary concern in it. Traditional approaches to combat security and privacy threats are not applicable to IoT as these devices have smaller storage capacity, less computation capability and they are battery operated. So there is a key requirement to develop a smart intrusion detection system (IDS) that can work efficiently in IoT environment. IDS can be signature-based (SBID), anomaly-based (ABID) or hybrid in nature. There is also a major concern about latency in IoT which is not desirable in real-time applications. To overcome this latency issue edge computing came into existence. Machine learning is one of the promising approaches to implement IDS. The aim of the present research study is to provide a deep insight into different models based on machine learning to detect intrusion in edge-based IoT networks.