New technologies bring opportunities to deploy AI and machine learning to the edge of the network, allowing edge devices to train simple models that can then be deployed in practice. Federated learning (FL) is a distributed machine learning technique to create a global model by learning from multiple decentralized edge clients. Although FL methods offer several advantages, including scalability and data privacy, they also introduce some risks and drawbacks in terms of computational complexity in the case of heterogeneous devices. Internet of Things (IoT) devices may have limited computing resources, poorer connection quality, or may use different operating systems. This paper provides an overview of the methods used in FL with a focus on edge devices with limited computational resources. This paper also presents FL frameworks that are currently popular and that provide communication between clients and servers. In this context, various topics are described, which include contributions and trends in the literature. This includes basic models and designs of system architecture, possibilities of application in practice, privacy and security, and resource management. Challenges related to the computational requirements of edge devices such as hardware heterogeneity, communication overload or limited resources of devices are discussed.

With the intent to further increase production efficiency while making human the centre of the processes, human-centric manufacturing focuses on concepts such as digital twins and human–machine collaboration. This paper presents enabling technologies and methods to facilitate the creation of human-centric applications powered by digital twins, also from the perspective of Industry 5.0. It analyses and reviews the state of relevant information resources about digital twins for human–machine applications with an emphasis on the human perspective, but also on their collaborated relationship and the possibilities of their applications. Finally, it presents the results of the review and expected future works of research in this area.

Over the past decade, the Internet of Things (IoT) has advanced rapidly. New technologies have been proposed and existing approaches optimised to meet user, society and industry requirements. However, as the complexity and heterogeneity of the traffic that flows through the networks are continuously growing, the innovation becomes difficult to achieve in both IoT and legacy networks. This article provides an overview of IoT application domains from a traffic characteristics perspective. Specifically, it identifies several groups of major IoT application use cases and discusses the exhibited traffic characteristics, used network technologies for implementation, and their feasibility as well as challenges. We stress that a key factor in future IoT development is network technologies and the way they handle and forward network traffic. The traffic characteristics emerging from this work can serve as a basis for future design proposals to develop more efficient solutions and improve the network technologies.