计算机科学

计算机科学 ›› 2025, Vol. 52 ›› Issue (3): 349-358.doi: 10.11896/jsjkx.240800067

• 计算机网络 • 上一篇    下一篇

基于异构图神经网络的网络切片端到端时延估计

胡海峰1, 朱漪雯2, 赵海涛3   

  1. 1 南京邮电大学通信与信息工程学院 南京 210003
    2 南京邮电大学波特兰学院 南京 210023
    3 南京邮电大学物联网学院 南京 210003
  • 收稿日期:2024-08-13 修回日期:2024-11-21 出版日期:2025-03-15 发布日期:2025-03-07
  • 通讯作者: 胡海峰(huhf@njupt.edu.cn)
  • 基金资助:
    国家自然科学基金(62371245)

Network Slicing End-to-end Latency Prediction Based on Heterogeneous Graph Neural Network

HU Haifeng1, ZHU Yiwen2, ZHAO Haitao3   

  1. 1 College of Telecommunications & Information Engineering,Nanjing University of Posts and Telecommunications,Nanjing 210003,China
    2 Portland Institute,Nanjing University of Posts and Telecommunications,Nanjing 210023,China
    3 College of Internet of Things,Nanjing University of Posts and Telecommunications,Nanjing 210003,China
  • Received:2024-08-13 Revised:2024-11-21 Online:2025-03-15 Published:2025-03-07
  • About author:HU Haifeng,born in 1973,Ph.D,professor.His mian research interests include deep learning for communication networks and graph representation lear-ning.
  • Supported by:
    National Natural Science Foundation of China(62371245).

PDF (PC)

179

摘要： 端到端时延作为网络切片重要的性能指标,在切片部署中因受到网络拓扑、流量模型和调度策略等影响,很难通过建模方式进行准确预测。为了解决上述问题,提出基于异构图神经网络的网络切片时延预测(Heterogeneous Graph Neural Network-Based Network Slicing Latency Prediction,HGNN)算法。首先,构建了切片-队列-链路的分层异构图,实现了切片的分层特征表达。然后,针对分层图中切片、队列和链路3种类型节点的属性特点,使用异构图神经网络挖掘拓扑动态变化、边特征信息和长依赖关系等和切片相关的底层特征,即分别选用GraphSAGE图神经网络、EGRET图神经网络和门控循环单元GRU来提取切片、队列和链路特征。同时,利用基于异构图神经网络的深度回归实现了网络切片特征表达的更新迭代和切片时延的准确预测。最后,通过构建基于OMNeT++的不同拓扑结构、流量模型和调度策略的切片数据库,验证了HGNN在实际网络场景下对切片端到端时延预测的有效性,并通过对比多种基于图深度学习的切片时延预测算法,进一步验证了HGNN在时延预测准确度和泛化性方面的优越性。

关键词: 网络切片, 异构图神经网络, 时延预测, 深度回归

Abstract: End-to-end latency,as a crucial performance metric for network slicing,is difficult to predict accurately via modeling due to the influences of network topology,traffic model,and scheduling policies.To tackle the above issues,we propose a heterogeneous graph neural network-based network slicing latency prediction(HGNN) algorithm,where the hierarchical heterogeneous graph of slice-queue-link is constructed to implement the hierarchical feature representation of the slice.Then,considering the attribute characteristics of three types of nodes in the hierarchical graph,i.e.slices,queues,and links,a heterogeneous graph neural network is presented to extract the underlying slice-related features such as topological dynamic changes,edge feature information,and long dependency relationships.Specifically,the graph neural network GraphSAGE,the graph neural network EGRET,and gated recurrent unit GRU are respectively adopted to extract the features of slices,queues,and link.Meanwhile,the iterative update of network slice feature representation and accurate prediction of slice latency are achieved using deep regression based on the heterogeneous graph neural network.Finally,a slice database with various topologies,traffic models,and scheduling policies is constructed using OMNeT++,and the effectiveness of HGNN in predicting slice end-to-end latency is validated on this database.Additionally,by comparing with other graph deep learning-based slice latency prediction methods,the superiority of HGNN in terms of prediction accuracy and generalization is further verified.

Key words: Network slicing, Heterogeneous graph neural network, Latency prediction, Deep regression

中图分类号: 

  • TN929
[1]WANG R,ZHANG K L.Survey of Network slicing[J].Journal of Nanjing University of Posts and Telecommunication(Natural Science Edition),2018,38(5):19-27.
[2]HEGDE P,MEENA S M.A survey on 5G Network Slicing-Epitome and opportunities for a novice[C]//2021 12th International Conference on Computing Communication and Networking Technologies(ICCCNT).2021:1-5.
[3]JIMÉNEZ M B,FERNÁNDEZ D,RIVADENEIRA J E,et al.A Survey of the Main Security Issues and Solutions for the SDN Architecture[J].IEEE Access,2021:122016-122038.
[4]ZOURE M,AHMED T,RÉVEILLÈRE L.Network ServicesAnomalies in NFV:Survey,Taxonomy,and Verification Me-thods[J].IEEE Transactions on Network and Service Mana-gement,2022,19(2):1567-1584.
[5]WANG S Y,YUEN C,NI W,et al.Multiagent Deep Reinforcement Learning for Cost- and Delay-Sensitive Virtual Network Function Placement and Routing[J].IEEE Transactions on Communications,2022,70(8):5208-5224.
[6]SATTER D,MATRAWY A.Optimal Slice Allocation in 5GCore Networks[J].IEEE Networking Letters,2019,1(2):48-51.
[7]WANG H Z,WU Y L,MIN G Y,et al.A Graph Neural Network-based Digital Twin for Network Slicing Management[J].IEEE Transactions on Industrial Informatics,2022,18(2):1367-1376.
[8]WU Y W,ZHANG K,ZHANG Y.Digital Twin Networks:ASurvey[J].IEEE Internet of Things Journal,2021,8(18):13789-13804.
[9]HAMILTON W L,YING R,LESKOVEC J.Inductive representation learning on largegraphs[C]//Proceedings of the 31st International Conference on Neural Information Processing Systems(NIPS’17).2017:1025-1035.
[10]MAHBUB S,BAYZID M S.EGRET:edge aggregated graph attention networks and transfer learning improve protein-protein interaction site prediction[J].Briefings in Bioinformatics,2022,23(2):bbab578.
[11]ANGGONO G,MOORS T.A Flow-Level Extension toOMNeT++ for Long Simulations of Large Networks[J].IEEE Communications Letters,2017,21(3):496-499.
[12]GUO S S,LU B B,WEN M W,et al.Customized 5G and Beyond Private Networks with Integrated URLLC,eMBB,mMTC,and Positioning for Industrial Verticals[J].IEEE Communications Standards Magazine,2022,6(1):52-57.
[13]WU Z X,YOU Y Z,LIU C C,et al.Machine Learning Based 5G Network Slicing Management and Classification[C]//2024 International Conference on Artificial Intelligence in Information and Communication(ICAIIC).Osaka,2024:371-375.
[14]MOVVA N D,ISHIGAKI G.Neural Network-based BlockingPrediction for Dynamic Network Slicing[C]//2024 33rd International Conference on Computer Communications and Networks(ICCCN).Kailua-Kona,HI,USA,2024:1-6.
[15]LI H,KONG Z X,CHEN Y W,et al.Slice-Based Service Function Chain Embedding for End-to-End Network Slice Deploy-ment[J].IEEE Transactions on Network and Service Management,2023,20(3):3652-3672.
[16]DAI M,SUN G,YU H F,et al.Maximize the Long-Term Average Revenue of Network Slice Provider via Admission Control Among Heterogeneous Slices[J].IEEE/ACM Transactions on Networking,2024,32(1):745-760.
[17]DONG J Y,GAO S T,LU H J,et al.Joint Optimization of Resource Allocation and Tasks Scheduling in Network Slicing Enabled Internet of Vehicles[C]//IEEE 8th International Confe-rence on Computer and Communications(ICCC).2022:552-556.
[18]GONG Y,WEI Y F,FENG Z Y,et al.Resource Allocation for Integrated Sensing and Communication in Digital Twin Enabled Internet of Vehicles[J].IEEE Transactions on Vehicular Technology,2023,72(4):4510-4524.
[19]TANG L,DU Y C,LIU Q H,et al.Digital-Twin-Assisted Resource Allocation for Network Slicing in Industry 4.0 and Beyond Using Distributed Deep Reinforcement Learning[J].IEEE Internet of Things Journal,2023,10(19):16989-17006.
[20]CHIANG Y,HSU C H,CHEN G H,et al.Deep Q-Learning-Based Dynamic Network Slicing and Task Offloading in Edge Network[J].IEEE Transactions on Network and Service Mana-gement,2023,20(1):369-384.
[21]XIONG Z P,WANG D Y,LIU X H,et al.Pushing the Boundaries of Molecular Representation for Drug Discovery with the Graph Attention Mechanism[J].Journal of Medicinal Chemistry,2020,63(16):8749-8760.
[22]HUANG Z H,WANG Z C,CHEN S.Sub-6GHz Assisted mmWave Hybrid Beamforming with Heterogeneous Graph Neural Network[J].IEEE Transactions on Communications,2024,72(11):6917-6928.
[23]MARWANI M,KADDOUM G.Graph Neural Networks Approach for Joint Wireless Power Control and Spectrum Allocation[J].IEEE Transactions on Machine Learning in Communications and Networking,2024,2:717-732.
[24]TAN Y W,LIU J J,WANG J D.5G End-to-End Slice Embedding Based on Heterogeneous Graph Neural Network and Reinforcement Learning[J].IEEE Transactions on Cognitive Communications and Networking,2024,10(3):1119-1131.
[25]YANG S,LI F,TRAJANOVSK S,et al.Recent Advances of Resource Allocation in Network Function Virtualization[J].IEEE Transactions on Parallel and Distributed Systems,2021,32(2):295-314.
[26]CHEN W K,LIU Y F,DOMENICO A D,et al,Optimal Network Slicing for Service-Oriented Networks With Flexible Routing and Guaranteed E2E Latency[J] IEEE Transactions on Network and Service Management,2021,18(4):4337-4352.
[27]ZHANG X Q,WANG T,JIANG R H,et al.Multi-AttentionConvolutional Neural Network for Video Deblurring[J].IEEE Transactions on Circuits and Systems for Video Technology,2022,32(4):1986-1997.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市渝北区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发