Towards Robust Federated Learning in IoT Systems: A Review of Fault-Tolerance Strategies

With the increasing number of devices being connected to the internet, the Internet of Things (IoT) has emerged as a promising technology to enhance the efficiency of various domains, including healthcare, transportation, and manufacturing. However, the reliability of IoT systems poses a significant challenge due to their distributed nature, heterogeneous devices, and dynamic environments. In this context, computational intelligence techniques have been employed to address the reliability issues of IoT systems. This article presents a comprehensive review of recent research on the application of computational intelligence techniques for enhancing the reliability of IoT systems. It provides an overview of IoT systems and the challenges they face concerning reliability. Also, discusses the principles and methodologies of computational intelligence techniques, including machine learning, deep learning, and evolutionary algorithms, and their potential applications in IoT reliability. Furthermore, this review work highlights several case studies that demonstrate the effectiveness of computational intelligence techniques in addressing various reliability issues in IoT systems, such as fault diagnosis, fault tolerance, and fault prediction. The case studies include the use of machine learning for fault detection in smart grids, deep learning for predictive maintenance in manufacturing, and the evolutionary algorithms for optimization of sensor placement in structural health monitoring.