Application of the Digital Twin Technology in Novel Edge Intelligent Computing Systems

New generation networks are expected to represent a concrete support to novel edge intelligence computing environments, enabling a novel plethora of disruptive delay-sensitive applications. In such a context, the advent of the digital twin (DT) paradigm has given new opportunities to Edge Intelligent systems and enabled new applications. This chapter presents an AI-empowered DT framework for a novel Edge Intelligent ecosystem according to the emerging paradigm of an integrated ground–air multiaccess edge computing infrastructure. In particular, we discuss the case of an unmanned aerial vehicle-aided multiaccess edge computing (UAV-MEC) system, where a set of DTs provide decisions to lower the edge computation congestion at the ground edge nodes on the basis of an appropriate approach. A three-dimensional matching game has been presented to perform the planning of the tasks offloading, channels allocation, and assignment of the UAV support to the congested ground edge nodes. Then, an architectural overview about the DT network has been detailed, considering a DT as a composition of elementary cyber entities. Finally, extensive experimental simulations are provided to highlight the advantages of the proposed framework, in terms of task completion delay and lower system congestion.