A Survey on Fault Detection in Industrial IoT: A Machine Learning Approach with Emphasis on Federated Learning and Intrusion Detection Systems

Abstract In recent years, the Internet of Things (IoT) has received a lot of attention and research. The concept of Industrial IoT (IIoT) has emerged from the con- vergence of information technology (IT) and industrial automation and control systems. The increasing number of disjointed IoT networks deployed in many industrial sectors has exposed vulnerabilities leading to security incidents, jeopar- dizing the overall security of IIoT systems. This paper provides a comprehensive survey, analyzing and comparing current technologies for securing IIoT networks. Researchers have developed various detection strategies supported by machine learning (ML) approaches. Federated Learning (FL) offers lower latency and pre- serves privacy, emerging as a promising distributed ML paradigm that enhances detection performance. Several challenges and recommendations are defined in the context of intrusion detection systems (IDS) as a security monitoring mechanism.