多接入边缘计算(MEC)可以在商品化平台上与开放无线接入网(O-RAN)一起实现,以提供低成本的部署并将服务更接近最终用户。本文提出了一种基于贝叶斯深度强化学习(RL)框架的联合O-RAN/MEC编排方法,该方法联合控制O-RAN功能拆分、跨地理分布平台的O-RAN/MEC服务所分配的资源及托管位置,以及每个O-RAN/MEC数据流的路由。其目标是在适应可能随时间变化的O-RAN/MEC需求和资源可用性的同时,最小化长期整体网络运营成本并最大化MEC性能指标。该编排问题被建模为马尔可夫决策过程(MDP)。然而,系统由多个基站(BS)组成,它们共享相同的资源并服务于异构需求,其参数之间存在非平凡的关系。因此,找到底层系统的精确模型是不切实际的,所建立的MDP具有大规模状态空间和多维离散动作。为了解决这种建模和维度问题,我们为解决方案框架提出了一种新颖的无模型RL代理。该代理基于Double Deep Q-network(DDQN)构建,以处理大规模状态空间,并结合了动作分支(一种动作分解方法),以线性增加的复杂度有效处理多维离散动作。此外,提出了一种基于汤普森采样的贝叶斯框架下的高效探索-利用策略,以提高学习性能并加速收敛。使用符合O-RAN的模型进行了基于轨迹的仿真。结果表明,我们的方法具有数据效率(即收敛更快),并且返回的奖励比其非贝叶斯版本高出32%。
英文摘要:
Multi-access Edge Computing (MEC) can be implemented together with Open Radio Access Network (O-RAN) over commodity platforms to offer low-cost deployment and bring the services closer to end-users. In this paper, a joint O-RAN/MEC orchestration using a Bayesian deep reinforcement learning (RL)-based framework is proposed that jointly controls the O-RAN functional splits, the allocated resources and hosting locations of the O-RAN/MEC services across geo-distributed platforms, and the routing for each O-RAN/MEC data flow. The goal is to minimize the long-term overall network operation cost and maximize the MEC performance criterion while adapting possibly time-varying O-RAN/MEC demands and resource availability. This orchestration problem is formulated as Markov decision process (MDP). However, the system consists of multiple BSs that share the same resources and serve heterogeneous demands, where their parameters have non-trivial relations. Consequently, finding the exact model of the underlying system is impractical, and the formulated MDP renders in a large state space with multi-dimensional discrete action. To address such modeling and dimensionality issues, a novel model-free RL agent is proposed for our solution framework. The agent is built from Double Deep Q-network (DDQN) that tackles the large state space and is then incorporated with action branching, an action decomposition method that effectively addresses the multi-dimensional discrete action with linear increase complexity. Further, an efficient exploration-exploitation strategy under a Bayesian framework using Thomson sampling is proposed to improve the learning performance and expedite its convergence. Trace-driven simulations are performed using an O-RAN-compliant model. The results show that our approach is data-efficient (i.e., converges faster) and increases the returned reward by 32\% than its non-Bayesian version.
Computation Offloading in Beyond 5G Networks: A Distributed Learning Framework and Applications
VTT Technical Research Centre of Finland(芬兰VTT技术研究中心);University of Electro-Communications(电气通信大学);Shizuoka University(静冈大学);University of Windsor(温莎大学);National Institute of Informatics(国立信息学研究所)
Intelligent Data-Driven Architectural Features Orchestration for Network Slicing
Universidade Federal de Viçosa (UFV)(维索萨联邦大学);Universidade Federal de Uberlândia (UFU)(乌贝兰迪亚联邦大学);Universidade do Minho(米尼奥大学);Universidade de São Paulo (USP)(圣保罗大学);Universidade Salvador (UNIFACS)(萨尔瓦多大学)
ORION: Intent-Aware Orchestration in Open RAN for SLA-Driven Network Management
National Telecommunications Institute (INATEL)(国家电信研究所);University of Vale do Rio dos Sinos (UNISINOS)(里奥多斯西诺斯谷大学);Federal Technological University of Paraná (UFTPR)(巴拉那联邦技术大学)
AIORA: An AI-Native Multi-Stakeholder Orchestration Architecture for 6G Continuum
Lenovo Deutschland GMBH(德国联想有限公司);Nederlandse Organisatie voor Toegepast Natuurwetenschappelijk Onderzoek TNO(荷兰应用科学研究组织TNO);Ruhr University Bochum(波鸿鲁尔大学)
随着计算能力逐渐成为数字经济时代的核心生产力,计算与网络融合(Computing and Network Convergence,CNC)的概念被提出并引起广泛关注,在CNC下,网络和计算资源可以根据用户的需求进行动态调度和分配。网络编排平面需要根据任务的属性灵活地将任务部署到合适的计算节点,并安排到达计算节点的路径。这是一个涉及资源调度和路径安排的编排问题。由于数控技术是一个相对较新的技术,本文对数控技术的研究和应用进行了综述。然后,我们设计了一种使用强化学习(RL)的CNC编排方法,这是第一次尝试,可以灵活地分配和调度计算资源和网络资源。其目标是高利润和低延迟。同时,我们使用多因素来确定优化目标,使编排策略优化的总体性能,从不同的方面,如成本,利润,延迟和系统负载在我们的实验。实验结果表明,与贪婪算法、随机选择算法和资源均衡算法相比,该方法具有更高的收益和更低的延迟。我们证明RL适合CNC编排。本文对RL在数控编排中的应用有一定的启示。
Agile Orchestration at Will: An Entire Smart Service-Based Security Architecture Towards 6G
National Engineering Research Center for Mobile Network Technologies, Beijing University of Posts and Telecommunications (BUPT), Beijing 100876, China and BUPT Shenzhen Institute, Shenzhen, China(国家移动网络技术工程研究中心,北京邮电大学(BUPT),北京100876,中国及BUPT深圳研究院,深圳,中国);China Telecom Corporation Limited Research Institute(中国电信股份有限公司研究院);Datang Mobile Communication Equipment Co. LTD(大唐移动通信设备有限公司);Singapore University of Technology and Design (SUTD)(新加坡科技设计大学(SUTD))
An Agentic AI Control Plane for 6G Network Slice Orchestration, Monitoring, and Trading
Florida International University, USA(佛罗里达国际大学);Center for Wireless Communications, University of Oulu, Finland(无线通信中心,奥卢大学);Accenture Technology Labs, Arlington, VA, USA(埃森哲技术实验室);Nanyang Technological University, Singapore(南洋理工大学)
Closed-Loop Decision-Focused Learning for User-Aware Cloud Orchestration under Uncertainty
School of Information Science and Engineering, Lanzhou University(兰州大学信息科学与工程学院);Division of Computer Science and Engineering, College of Engineering, Louisiana State University(路易斯安那州立大学工程学院计算机科学系);School of Artificial Intelligence, Shenzhen University(深圳大学人工智能学院)
Energy-Aware Agentic Orchestration for Multi-Site Service Deployment
Federal University of Viçosa (UFV)(维索萨联邦大学);Federal University of Uberlândia (UFU)(乌贝兰迪亚联邦大学);Instituto de Telecomunicações(电信研究所);Universidade de Aveiro (UA)(阿威罗大学);University of São Paulo (USP)(圣保罗大学);Universitat Politecnica de València (UPV)(瓦伦西亚理工大学);University of Minho (UMinho)(米尼奥大学)
AI-Native Orchestration in the 6G Continuum: Evolving Operator Platforms with Agentic AI
i2CAT Foundation(i2CAT基金会);Intracom Telecom(Intracom电信);Industrial Systems Institute (ISI)/Athena Research Center(工业系统研究所(ISI)/雅典研究中心);University of Patras(帕特雷大学)
Conflict-Aware Autonomous Resource Orchestration for 5G Edge Slicing Using Multi-Agent Systems and LLMs
Federal University of Rio Grande do Sul (UFRGS)(南里奥格兰德联邦大学);Federal University of Viçosa (UFV)(维索卡联邦大学);University of São Paulo (USP)(圣保罗大学);University of Minho (UMinho)(米尼奥大学)