Intelligent Decision-Making for Low-Latency Communication in MEC

Control requirement-driven connectivity optimization is pivotal in Multi-access Edge Computing (MEC) systems, which aim to provide low-latency and high-reliability communication services for connected vehicular networks. By using reinforcement learning as well as heuristic algorithms, we jointly measure the performance of different connectivity options (such as 4G, 5G and 802.11p), and manage the choice of the best technology by considering the requirements for vehicle control, in order to deliver the optimal communication service to the vehicles for every scenario, manoeuvre, and topology. This work is part of the SELF4COOP project.

Enhancing Edge Computing Efficiency and Connectivity Optimization/Dynamic Control Function Placement in Vehicular Networks: Optimization and Adaptation

Computing infrastructure monitoring and control function placement is essential for the dynamic orchestration of computing resources in MEC systems. The system measures the responsiveness of each computing resource (including MEC servers and on-vehicle computers) and solve optimization problems to dynamically orchestrate the placement of the connectivity optimization and maneuvering control functions on the resources that optimally serve the vehicles in real time. This work is part of the SELF4COOP project.

Analysis for speech recognition Using artificial intelligence methods

The aim of this project is to carry out a data collection campaign to have sample recordings of pathological voices, which are then analysed and characterized. This work is part of the RATTLE project.