A Comprehensive Review of Multi-Resource Anomaly Detection Using Federated AI and IoT
https://doi.org/10.65900/JAIEITC.2026.v01i01.002
Keywords:
anomaly detection, artificial intelligence, deep learning, energy monitoring, federated learning, industrial IoT, internet of things, multi-resource monitoring, smart buildings, water monitoringAbstract
The integration of Artificial Intelligence (AI), Internet of Things (IoT), and Federated Learning (FL) has enabled advanced and privacy-preserving anomaly detection for smart resource management. Modern infrastructures generate large volumes of sensor data related to energy consumption, water usage, and electrical device operation. Identifying abnormal patterns is essential for improving efficiency, reducing resource wastage, enhancing equipment reliability, and supporting sustainability. Conventional centralized approaches require data to be transmitted to cloud servers, creating privacy, communication, and scalability challenges. Federated Learning enables collaborative model training across distributed devices without sharing raw data. Combined with techniques such as Deep Learning, LSTM, Autoencoders, CNNs, Transformers, and Explainable AI, FL provides promising solutions for accurate and secure anomaly detection. This survey examines twenty recent studies covering energy systems, water management, smart grids, Industrial IoT, battery storage, distributed energy resources, and intelligent buildings. The studies are analysed based on their methodologies, datasets, AI techniques, benefits, limitations, and research gaps. The review indicates that most existing approaches address individual resource domains, while integrated monitoring of energy, water, and electrical device usage remains limited. This gap motivates the development of a unified Federated AI and IoT-based framework for secure, scalable, and real-time multi-resource anomaly detection.
