A Federated Learning Framework for Mobile Edge Computing Networks

Recently, the continuous growth of smart devices needing processing has led to move storage and computation from cloud to the network edges, giving rise to the edge computing paradigm. Due to the limited capacity of edge computing nodes, the presence of popular applications in the edge nodes results in significant improvements on users satisfaction and services accomplishment. However, the high variability in the content requests makes prediction demand not trivial and, typically, the majority of the classical prediction approaches require the gathering of personal users information at a central unit, giving rise to many users privacy issues. In this context, federated learning gained attention as solution to perform learning procedures from data disseminated across multiple users, keeping the sensitive data protected. This paper applies the federated learning to the applications demand prediction problem, to accurately forecast the more popular applications types in the network. The proposed framework reaches high accuracy levels on the predicted applications demand, by aggregating in a global and weighted model the feedback received by users, after their local training process. The validity of the proposed approach is verified by performing a virtual machine replica copies and comparison with the alternative forecasting approach based on chaos theory.