REVIEW OF ANOMALY DETECTION IN INDUSTRIAL IOT

Abstract

Over recent years, IoT has evolved into a paramount technology of the modern era. The data we share every day over this personified platform exposes one to a huge list of benefits. Now that it is possible to connect objects like home appliances, cars etc. to the internet, communication is possible between humans and things. The Internet has given us the comfort to control/automate the entire work whilst sitting in any part of the world. This network of “things” operates on huge amounts of real-time data. The data so collected is processed and is then used to make certain modelled decisions. This technology gives a vast implementation opportunity to all kinds of industries whether it be a product, service, or even security-based system. Although IoT technology can benefit humans in many possible ways, the risks involved cannot be ignored. The majority of risks involved are associated with the privacy and security of data. In this review paper, we discuss and deliver an organized literature review of the anomaly detection system with machine learning and deep learning models that help carry out the IoT process securely. On the basis of our study, academics and scholars can familiarize themselves with various methods and can use them to solve anomaly detection problem by developing novel techniques for predicting and detecting anomalies in Industrial IoT.