Wireless radio networks for industrial applications

The increasing complexity and dynamic nature of industrial applications and Use Cases (UCs) pose significant challenges. While wired solutions meet stringent communication and adaptability requirements, they fail to accomplish mobility. Conversely, wireless technologies support mobility but often lack the required level of adaptability and robustness. These constraints underscore the need for innovative wireless technologies capable of addressing the evolving demands of industrial UCs. To address such limitations, this thesis investigates next-generation wireless radio networks for Industrial Internet of Things (IIoT) applications. Within this framework, the analysis assesses the potential of several key enabling technologies envisioned for the Sixth-Generation (6G) and their integration across the protocol stack, with the goal of enhancing production processes. At the physical layer, the examination encompasses high-frequency bands, including the Terahertz (THz) spectrum, assessing their performance compared to current wireless technologies relying on sub-6 GHz and millimeter Waves (mmWaves). The exploration continues with the investigation of Machine Learning techniques for channel prediction and concludes by assessing strengths and limitations of Reconfigurable Intelligent Surfaces (RISs) to extend coverage in challenging industrial environments. At the data link layer, adaptive Medium Access Control (MAC) protocols are proposed to satisfy a wide range of heterogeneous, stringent, evolving, and conflicting requirements. The study spans classical optimization strategies and Artificial Intelligence (AI)-based solutions to develop adaptable, multi-goal MAC frameworks for diverse industrial UCs (e.g., production, pick-and-place operations, and transportation). At the application layer, two Digital Twin (DT) frameworks are developed for network and industrial operation optimization, enabling simulation, prediction, and optimization through the integration of Reinforcement Learning algorithms into a safe virtual environment. The findings of this research demonstrate the potential of several 6G key enabling technologies to enhance production processes and operational efficiency within IIoT scenarios, aligned with the vision of Industry 5.0.