Optimising Decision Making on Communication Systems: The Federated Learning Approach
Konstantinos D. Stergiou, Kostas E. Psannis, Μanos Roumeliotis, George Kokkonis, Yutaka Ishibashi · 2021 IEEE 9th International Conference on Information, Communication and Networks (ICICN) · 2021
We provide a survey of different categories of communication systems to which conventional decision-making schemes which are aware of the goals and constraints of each system are enhanced Federated Learning algorithms. Formulating new Federated Learning-oriented strategies key design aspects and open research issues (e.g. resource optimization, reliability) are presented from an empirical research view. Controlled experiments at the level of simulation or real-life data sets indicate that FL-based communication schemes become context- and mobility- aware with the mission of optimizing the decisions of each design. Also, empirical research results present a goodness of fit of FL schemes into hybrid, multilayer communication systems as they are emerging in mobile-cloud networks, 5G, and IOT.