服务功能链中基于机器学习的QoE评估与预测

doi:10.19678/j.issn.1000-3428.0060387

计算机工程 ›› 2022, Vol. 48 ›› Issue (1): 163-169. doi: 10.19678/j.issn.1000-3428.0060387

服务功能链中基于机器学习的QoE评估与预测

赵季红^1,2, 张文娟¹, 乔琳琳¹, 张梦雪¹

1. 西安邮电大学通信与信息工程学院, 西安 710121;
2. 西安交通大学电子信息工程学院, 西安 710049

收稿日期:2020-12-24 修回日期:2021-01-25 发布日期:2022-01-04
作者简介:赵季红(1963-),女,教授、博士、博士生导师,主研方向为带宽通信网、新一代无线移动互联网;张文娟、乔琳琳、张梦雪,硕士研究生。
基金资助:
国家自然科学基金（61531013）；国家重点研发计划重点专项（2018YFB1800300）。

QoE Evaluation and Prediction Based on Machine Learning in Service Function Chain

ZHAO Jihong^1,2, ZHANG Wenjuan¹, QIAO Linlin¹, ZHANG Mengxue¹

1. School of Communication and Information Engineering, Xi'an University of Post and Telecommunications, Xi'an 710121, China;
2. School of Electronic and Information Engineering, Xi'an Jiaotong University, Xi'an 710049, China

Received:2020-12-24 Revised:2021-01-25 Published:2022-01-04

摘要/Abstract

摘要： 在软件定义网络与网络功能虚拟化协同的网络架构下，只考虑单个服务质量（QoS）指标的服务功能链部署无法满足用户的多业务体验需求。提出一种基于机器学习的服务功能链部署模型。基于层次分析法构造MPNQ2算法以建立QoS与体验质量（QoE）的映射关系，得出影响QoE的网络参数并评估其影响权重。在此基础上，利用具备较强综合学习和泛化能力的随机森林模型对服务功能链的QoE进行预测。实验结果表明，与梯度提升决策树、线性判别分析等机器学习模型相比，随机森林模型为预测QoE的最佳模型，同时在影响QoE的网络参数中，丢包率对服务功能链的部署影响最大。

关键词: 软件定义网络, 网络功能虚拟化, 服务功能链, 机器学习, 体验质量

Abstract: Under the Software Defined Network(SDN)/Network Function Virtualization(NFV) cooperative network architecture, the service function chain deployment which only considers a single Quality of Service(QoS) index can not meet the user experience demands for multiple services.To address the problem, a Quality of Experience(QoE) evaluation and prediction model based on machine learning is proposed.By using Analytic Hierarchy Process(AHP), a MPNQ2 algorithm is constructed to establish the mapping relationship between QoS and QoE, and the network parameters influencing QoE and their influence weights are pointed out.Then the QoE of service function chains is predicted by using the strong comprehensive learning ability and generalization ability of random forest.The experimental results show that compared with gradient boosting decision tree and other machine learning models, random forest is the best model to predict QoE.In addition, the packet loss rate has the greatest impact on the deployment of service function chains.

Key words: Software-Defined Network(SDN), Network Function Virtualization(NFV), Service Function Chain(SFC), machine learning, Quality of Experience(QoE)

中图分类号:

TP393

赵季红, 张文娟, 乔琳琳, 张梦雪. 服务功能链中基于机器学习的QoE评估与预测[J]. 计算机工程, 2022, 48(1): 163-169.

ZHAO Jihong, ZHANG Wenjuan, QIAO Linlin, ZHANG Mengxue. QoE Evaluation and Prediction Based on Machine Learning in Service Function Chain[J]. Computer Engineering, 2022, 48(1): 163-169.

https://www.ecice06.com/CN/Y2022/V48/I1/163

图/表 12

20220108122728

20220108122733

20220108122737

20220108122742

20220108122747

20220108122751

20220108122755

20220108122800

20220108122804

20220108122808

20220108122812

20220108122816

参考文献

[1] PRAJAPATI A, SAKADASARIYA A, PATEL J.Software defined network:future of networking[C]//Proceedings of the 2nd International Conference on Inventive Systems and Control.Washington D.C., USA:IEEE Press, 2018:1351-1354.
[2] HERRERA J G, BOTERO J F.Resource allocation in NFV:a comprehensive survey[J].IEEE Transactions on Network and Service Management, 2016, 13(3):518-532.
[3] ABBAS B, Mubashir HUSAIN R, ISRAAT H, et al.Software-Defined Networking (SDN) and Network Function Virtualization (NFV) for a hyperconnected world:challenges, applications, and major advancements[J].Network and System Management, 2020, 28(3):433-435.
[4] ADELl N T, JUNGMIN S, QING C, et al.ElasticSFC:auto-scaling techniques for elastic service function chaining in network functions virtualization-based clouds[J].System and Software, 2019, 152(2):108-119.
[5] SON J M, HE T Z, RAJKUMAR B.CloudSimSDN-NFV:modeling and simulation of network function virtualization and service function chaining in edge computing environments[J].Software-Practice & Experience, 2019, 49(12):1748-1764.
[6] 彭峰.边缘计算网络与QoE感知资源调度研究[D].北京:北京邮电大学, 2019. PENG F.Research on edge computing network and QoE sensing resource scheduling[D].Beijing:Beijing University of Posts and Telecommunications, 2019.(in Chinese)
[7] SUBRAMANYA T, HARUTYUNYAN D, RIGGIO R.Machine learning-driven service function chain placement and scaling in MEC-enabled 5G networks[J].Computer Networks, 2020, 166(15):1-16.
[8] GS A, QLB C, YONG J A, et al.A four-stage adaptive scheduling scheme for service function chain in NFV-sciencedirect[EB/OL].[2020-11-07].https://www.researchgate.net/publication/340904617_A_Four-Stage_Adaptive_Scheduling_Scheme_for_Service_Function_Chain_in_NFV.
[9] KIM S, HAN Y, PARK S.An energy-aware service function chaining and reconfiguration algorithm in NFV[C]//Proceedings of the 1st IEEE International Workshops on Foundations and Applications of Self Systems.Washington D.C., USA:IEEE Press, 2016:54-59.
[10] KIM S, PARK S, KIM Y, et al.VNF-EQ:dynamic placement of virtual network functions for energy efficiency and QoS guarantee in NFV[J].Cluster.Comput.2017, 20(3):2107-2117.
[11] XI C, LI Z, ZHANG Y, et al.Reinforcement learning⁃based QoS/QoE-aware service function chaining in software-driven 5G slices[J].Transactions on Emerging Telecommunications Technologies, 2018, 29(11):477-491.
[12] WANG I C, WEN H P, CHAO H J.Improving quality of experience of service-chain deployment for multiple users[C]//Proceedings of the 26th IEEE/ACM International Symposium on Quality of Service.Washington D.C., USA:IEEE Press, 2018:1-6.
[13] HE W.Markov approximation method for optimal service orchestration in IoT network[J].IEEE Access, 2019, 7:38-49
[14] JIN Q.ScaRL:service function chain allocation based on reinforcement learning in mobile edge computing[C]//Proceedings of the 7th International Conference on Advanced Cloud and Big Data.Washington D.C., USA:IEEE Press, 2019:327-332.
[15] 孙超.软件定义网络中流媒体QoE控制技术研究[D].济南:山东大学, 2018. SUN C.Research on QoE control technology of streaming media in software-defined network[D].Jinan:Shandong University, 2018.(in Chinese)
[16] CASAS P, WASSERMANN S.Improving QoE prediction in mobile video through machine learning[EB/OL].[2020-11-07].https://www.researchgate.net/publication/321075693_Improving_QoE_Prediction_in_Mobile_Video_through_Machine_Learning.
[17] 唐伦, 赵培培, 赵国繁, 等.基于深度信念网络资源需求预测的虚拟网络功能动态迁移算法[J].电子与信息学报, 2019, 41(6):1397-1404. TANG L, ZHAO P P, ZHAO G F, et al.Virtual network function dynamic migration algorithm based on deep belief network resource demand prediction[J].Journal of Electronics and Information Technology, 2019, 41(6):1397-1404.(in Chinese)
[18] 张晨.基于OpenFlow的服务链研究与设计[D].北京:北京邮电大学, 2017. ZHANG C.Research and design of service chain based on OpenFlow[D].Beijing:Beijing University of Posts and Telecommunications, 2017.(in Chinese)
[19] 张若晨.基于OpenvSwitch的代理虚拟交换机在SDN网络中的实现与应用[D].广州:华南理工大学, 2016. ZHANG R C.Implementation and application of openvswitch-based virtual switch in SDN network[D].Guangzhou:South China University of Technology, 2016.(in Chinese)
[20] BANOVIĆ-ĆURGUZ N, ILIŠEVIĆ D.Mapping of QoS/QoE in 5G networks[EB/OL].[2020-11-07].https://www.researchgate.net/publication/334417694_Mapping_of_QoSQoE_in_5G_Networks.
[21] SADREDDINI Z.A novel cancellation protection service in online reservation system[J].IEEE Access, 2020, 99:1-11.
[22] LAN H, PAN Y A Crowdsourcing quality prediction model based on random forests[C]//Proceedings of the 18th International Conference on Computer and Information Science.Washington D.C., USA:IEEE Press, 2019:315-319.
[23] GUO Y, ZHOU Y, HU X, et al.Research on recommendation of insurance products based on random forest[EB/OL].[2020-11-07].https://www.researchgate.net/publication/338360719_Research_on_Recommendation_of_Insurance_Products_Based_on_Random_Forest.
[24] BEN L.WBQoEMS:web browsing QoE monitoring system based on prediction algorithms[J].International Journal of Communication Systems.2019, 32(13):30-48.

选择文件类型/文献管理软件名称

选择包含的内容

服务功能链中基于机器学习的QoE评估与预测

QoE Evaluation and Prediction Based on Machine Learning in Service Function Chain

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 12

参考文献

相关文章 15

编辑推荐

Metrics

本文评价

[1]	崔允贺, 赵建朋, 杨鸿珍, 李显超. 基于曲线决策融合的SDN饱和攻击检测方法[J]. 计算机工程, 2024, 50(8): 123-132.
[2]	徐明亮, 李芳媛, 马浩然, 何飞. 大规模神经记录的峰电位聚类算法(特邀)[J]. 计算机工程, 2024, 50(6): 1-34.
[3]	李永飞, 李铭洋, 常鑫, 曹可欣. 基于可解释性深度学习的物联网水质监测数据异常检测[J]. 计算机工程, 2024, 50(6): 179-187.
[4]	孙毅, 王会梅, 鲜明, 向航. Kubeflow异构算力调度策略研究[J]. 计算机工程, 2024, 50(2): 25-32.
[5]	郭雷, 荆山, 魏亮, 赵川. 基于软件定义网络的Crossfire攻击防御方法[J]. 计算机工程, 2024, 50(10): 216-227.
[6]	张冠莹, 伊鹏, 李丹, 朱棣, 毛明. 面向大规模网络的服务功能链部署方法[J]. 计算机工程, 2023, 49(8): 122-129.
[7]	刘向举, 赵犇, 方贤进, 徐杨洋. SDN中基于过程优化的动态负载均衡策略[J]. 计算机工程, 2023, 49(8): 137-145.
[8]	赵徐炎, 崔允贺, 钱清, 郭春, 申国伟. 基于虚拟扩展网络的虚拟机及路由路径联合部署[J]. 计算机工程, 2023, 49(8): 154-162.
[9]	陈治旭, 靳雁霞, 芦烨, 杨晶, 刘亚变, 史志儒. 基于子图卷积神经网络的多精度服装建模方法[J]. 计算机工程, 2023, 49(4): 174-181.
[10]	陈何雄, 罗宇薇, 韦云凯, 郭威, 杭菲璐, 何映军, 杨宁. 基于联邦学习的SDN异常流量协同检测技术[J]. 计算机工程, 2023, 49(3): 168-176.
[11]	刘宇宸, 唐菀, 杨喜敏, 张艳. 引入区块链的SDN路由一致性检测[J]. 计算机工程, 2023, 49(11): 160-168.
[12]	刘金硕, 詹岱依, 邓娟, 王丽娜. 基于深度神经网络和联邦学习的网络入侵检测[J]. 计算机工程, 2023, 49(1): 15-21,30.
[13]	葛昕, 邹福泰, 郭万达, 谭越, 李林森. 社交僵尸网络发展综述[J]. 计算机工程, 2022, 48(8): 12-24.
[14]	俞莎莎, 牛保宁. 基于交易不可信度的比特币非法交易检测[J]. 计算机工程, 2022, 48(8): 166-172.
[15]	金海波, 赵欣越. 共形预测框架下的高可靠入侵检测算法[J]. 计算机工程, 2022, 48(7): 130-140.

模态框（Modal）标题

选择文件类型/文献管理软件名称

选择包含的内容

服务功能链中基于机器学习的QoE评估与预测

QoE Evaluation and Prediction Based on Machine Learning in Service Function Chain

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 12

参考文献

相关文章 15

编辑推荐

Metrics

本文评价