In recent years, the extension of network function virtualization (NFV) concept to the edge of mobile networks has allowed the deployment of virtualized Radio Access Networks (vRANs), where RAN functions are decoupled from specific-purpose hardware, thus permitting mobile operators to mitigate vendor lock-in and to gain in flexibility and time-to-market when deploying RANs. Since several vRANs functions need to be offloaded to dedicated Hardware Accelerators (HAs) to meet tight time constraints, it is crucial for mobile operators to share efficiently the HAs among multiple Base Stations with the aim of minimizing cost while keeping certain performance requirements. By focusing on Forward Error Correction (FEC) LDPC Decoding as 5G vRAN function, we propose a Reinforcement Learning based algorithm that controls vRAN radio resources by applying radio scheduling policies together with a scheduler responsible for HAs sharing among multiple Base Stations, with the final goal of optimizing cost-performance behavior of the vRAN.
Leonardo Lo Schiavo is a Ph.D. student of the Networks Data Science Group at IMDEA Networks Institute since May 2020. He received his Bachelor’s Degree in Computer Science Engineering from the University of Catania in 2016 and a double Master’s Degree in Communications and Computer Networks Engineering from Politecnico di Torino and Politecnico di Milano in 2018, holding the research for his Master’s Thesis in the Multi-Access Edge Computing (MEC) team at Telecom Italia in Turin, where he focused on developing MEC solutions for safety applications in vehicular networks. His current main research interests include Network Functions Virtualization (NFV), RAN virtualization, Deep/Reinforcement Learning.
This event will be conducted in English