Published Paper Details:

Because Industrial Internet of Things (IIoT) devices are becoming increasingly integrated into contemporary manufacturing procedures, there has been an increase in the demand for machine learning models that are both reliable and secure inside these contexts. On the other hand, due to the sensitive nature of industrial data, a rigorous approach to the protection of their privacy is required. The purpose of this study is to offer a framework that is trustworthy and protects privacy, with the intention of facilitating machine learning applications in Industrial Internet of Things systems while simultaneously protecting sensitive information. The framework makes use of a variety of encryption methods, federated learning, and differential privacy in order to guarantee the secrecy of data, the correctness of models, and the protection of privacy. The performance of object data interchange may be improved by the utilisation of device-to-device (D2D) communication mechanisms, which can be utilised by the Internet of Things (IoT). It is the goal of Internet of Things networks to provide a vast array of services of a high quality, and a significant proportion of the devices that are responsible for providing these services are mobile. Wearables, sensors, drones, and smart cars are examples of devices that require ongoing communication despite their movement patterns. As a result, an Internet of Things design should take into consideration both Quality of Service (QoS) and mobility. By enabling devices to connect with one another directly, D2D makes it possible for them to exchange material and functionality, such as access to the Internet. In order to improve the performance of Internet of Things (IoT) communication and to provide better quality of service (QoS) for data exchange services between mobile Internet of Things devices, this article presents REMOS-IoT, which stands for a RElay and MObility Scheme. The effectiveness of the suggested architecture and algorithms was demonstrated through simulation-based testing, which demonstrated how the performance of electronic devices improved in a number of different circumstances.