Federated Learning for Industrial Internet of Things in Future\n Industries

The Industrial Internet of Things (IIoT) offers promising opportunities to\ntransform the operation of industrial systems and becomes a key enabler for\nfuture industries. Recently, artificial intelligence (AI) has been widely\nutilized for realizing intelligent IIoT applications where AI techniques\nrequire centralized data collection and processing. However, this is not always\nfeasible in realistic scenarios due to the high scalability of modern IIoT\nnetworks and growing industrial data confidentiality. Federated Learning (FL),\nas an emerging collaborative AI approach, is particularly attractive for\nintelligent IIoT networks by coordinating multiple IIoT devices and machines to\nperform AI training at the network edge while helping protect user privacy. In\nthis article, we provide a detailed overview and discussions of the emerging\napplications of FL in key IIoT services and applications. A case study is also\nprovided to demonstrate the feasibility of FL in IIoT. Finally, we highlight a\nrange of interesting open research topics that need to be addressed for the\nfull realization of FL-IIoT in industries.\n