EdgeBOL: A Bayesian Learning Approach for the Joint Orchestration of vRANs and Mobile Edge AI
This article proposed a Bayesian learning framework for jointly configuring the service and the Radio Access Network (RAN), aiming at minimisation of the total energy consumption while respecting accuracy and latency service requirements. Using a fully-fledged prototype with a software-defined base station (vBS) and a GPU-enabled edge server, it profiles a typical video analytics service and identifies new performance trade-offs and optimization opportunities. The associated findings suggest that this approach effectively adapts to different hardware platforms and service requirements, and outperforms state-of-the-art benchmarks based on neural networks.
Jose A. Ayala-Romero, et al. (NEC)
IEEE/ACM Transactions on Networking