A Critical Role for Federated Learning in IoT
Kaushal Kishor, Angeles Quezada, Satya Prakash Yadav · 2024
Artificial intelligence (AI) is increasing the use of intelligent services and applications, resulting in extensive integration of the Internet of Things (IoT) into all aspects of our lives. Conventional AI systems often involve centralized data collection and processing. Given the large size of IoT networks and the issues associated with data privacy, this strategy is unlikely to be useful in real-world scenarios. Federated learning (FL) is an AI technique that encourages cooperation and decentralization in the creation of intelligent IoT applications. AI may be taught on scattered IoT devices without having to exchange data. This article investigates the rise of FL on IoT networks. The research begins with a brief summary of advances in FL and the IoT, followed by an examination of their integration. We evaluate the IoT services offered by FL. We provide solutions for IoT data transfer, unloading, caching, attack detection, relocation, mobile crowdsensing, privacy, and security. The next stage will be to apply FL to large-scale IoT applications, including smart healthcare, smart transportation, unmanned aerial vehicles (UAVs), smart cities, and smart industries. This research also highlights noteworthy findings on FL-IoT services and applications. To end, we will discuss current issues and future research on this rapidly developing subject.