期刊
COMPUTER NETWORKS
卷 188, 期 -, 页码 -出版社
ELSEVIER
DOI: 10.1016/j.comnet.2021.107829
关键词
Network slicing; Machine-Learning; Resource orchestration; 5G and beyond; OpenAirInterface OAI
类别
资金
- FUI SCORPION project, France [17/00464]
- CNRS PRESS project, France [07771]
- Lebanese University
- Azm & Saade'' Association
This research proposes a framework for orchestrating 5G network resources using machine learning techniques, aiming to improve resource utilization and network performance through predictive modeling and dynamic slicing methods. By utilizing experimental prototypes and real datasets, the study shows that ML-based approaches outperform traditional methods in terms of prediction accuracy and system utilization.
To efficiently serve heterogeneous demands in terms of data rate, reliability, latency and mobility, network operators must optimize the utilization of their infrastructure resources. In this context, we propose a framework to orchestrate resources for 5G networks by leveraging Machine Learning (ML) techniques. We start by classifying the demands for resources into groups in order to adequately serve them by dedicated logical virtual networks or Network Slices (NSs). To optimally implement these heterogeneous NSs that share the same infrastructure, we develop a new dynamic slicing approach of Physical Resource Blocks (PRBs). On first hand, we propose a predictive approach to achieve optimal slicing decisions of the PRBs from a limited resource pool. On second hand, we design an admission controller and a slice scheduler and formalize them as Knapsack problems. Finally, we design an adaptive resource manager by leveraging Deep Reinforcement Learning (DRL). Using our 5G experimental prototype based on OpenAirInterface (OAI), we generate a realistic dataset for evaluating ML based approaches as well as two baselines solutions (i.e. static slicing and uninformed random slicing-decisions). Simulation results show that using regression trees for both classification and prediction, coupled with the DRL-based adaptive resource manager, outperform alternative approaches in terms of prediction accuracy, resource smoothing, system utilization and network throughput.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据