IEEE Network · 2020 · 157 citations · 13 references
ComputingComputer ScienceComputer SystemsArtificial IntelligenceEdge ComputingMachine LearningDeep LearningEdge Artificial IntelligenceLatency OptimizationMobile Edge ComputingArchitecture-level OptimizationDeep Reinforcement LearningEdge IntelligenceEnd-to-end DelayEnd-to-end Delay AnalysisEnd-to-end Delay OptimizationMobile Edge Computing ArchitectureMulti-tier Architecture
Future 6G networks rely on URLLC to support mission‑critical applications demanding strict delay and reliability, yet current solutions are largely theoretical and not directly deployable. The paper aims to review data‑driven deep learning and reinforcement learning for URLLC and to propose a multi‑level architecture that integrates device, edge, and cloud intelligence to overcome identified challenges. The proposed architecture merges theoretical models with real‑world data to train DNNs, employs deep transfer learning to adapt pre‑trained networks to non‑stationary conditions, and uses federated learning to enhance efficiency under limited device and edge computing resources. Experimental and simulation results demonstrate the feasibility of the architecture, and the authors outline future research directions.
In future 6th generation networks, URLLC will lay the foundation for emerging mission-critical applications that have stringent requirements on end-to-end delay and reliability. Existing works on URLLC are mainly based on theoretical models and assumptions. The model-based solutions provide useful insights, but cannot be directly implemented in practice. In this article, we first summarize how to apply data-driven supervised deep learning and deep reinforcement learning in URLLC, and discuss some open problems of these methods. To address these open problems, we develop a multi-level architecture that enables device intelligence, edge intelligence, and cloud intelligence for URLLC. The basic idea is to merge theoretical models and realworld data in analyzing the latency and reliability and training deep neural networks (DNNs). Deep transfer learning is adopted in the architecture to fine-tune the pre-trained DNNs in non-stationary networks. Further considering that the computing capacity at each user and each mobile edge computing server is limited, federated learning is applied to improve the learning efficiency. Finally, we provide some experimental and simulation results and discuss some future directions.
13
Communication-Efficient Learning of Deep Networks from Decentralized Data
H. Brendan McMahan, Eider Moore, Daniel Ramage et al. · arXiv (Cornell University) · 2016 · 5.2K citations · Full text
Client-Edge-Cloud Hierarchical Federated Learning
Lumin Liu, Jun Zhang, Shenghui Song et al. · 2020 · 886 citations