SRv6 dynamic performance optimization method based on machine learning

Tian Dongyang; Wang Shuofan; Qi Huang; Wang Kai

doi:10.11959/j.issn.1000-0801.DXKX250630

您当前的位置：

首页 >

文章列表页 >

SRv6 dynamic performance optimization method based on machine learning

Research and Development | 更新时间：2026-07-21

- SRv6 dynamic performance optimization method based on machine learning
- Telecommunications Science Vol. 42, Issue 6, Pages: 188-202(2026)
- 作者机构：
  
  中国移动通信集团江苏有限公司，江苏南京 210029
- 作者简介：
- 基金信息：
- DOI：10.11959/j.issn.1000-0801.DXKX250630
  CLC： TP393.08;TN915.05
- Received：28 October 2025，
  
  Revised：2026-02-03，
  
  Accepted：10 February 2026，
  
  Online First：21 July 2026，
  
  Published：20 June 2026
- 稿件说明：
移动端阅览
田东阳,王硕凡,祁璜等.基于机器学习的SRv6动态性能优化方法研究[J].电信科学,2026,42(06):188-202.

Tian Dongyang,Wang Shuofan,Qi Huang,et al.SRv6 dynamic performance optimization method based on machine learning[J].Telecommunications Science,2026,42(06):188-202.
田东阳,王硕凡,祁璜等.基于机器学习的SRv6动态性能优化方法研究[J].电信科学,2026,42(06):188-202. DOI： 10.11959/j.issn.1000-0801.DXKX250630.

Tian Dongyang,Wang Shuofan,Qi Huang,et al.SRv6 dynamic performance optimization method based on machine learning[J].Telecommunications Science,2026,42(06):188-202. DOI： 10.11959/j.issn.1000-0801.DXKX250630.

摘要

随着互联网和通信技术的迅速发展，段路由IPv6（segment routing over IPv6，SRv6）作为一项革新性网络技术，简化了网络结构，提高了网络拓扑的灵活性和可编程性，可满足更多新业务的多种需求。而传统的静态策略无法实时响应庞大且复杂的网络状态变化，同时网络流量难以实时监测和预测，且网络优化通常需要在多个相互冲突的目标之间权衡。为此，提出了一种基于时空混合神经网络与增强型强化学习相结合的SRv6动态优化框架，利用长短期记忆（long short-term memory，LSTM）网络和图卷积网络（graph convolutional network，GCN）捕捉网络流量的时空特征，进一步通过特征融合得到网络性能指标。再使用增强型Q学习算法与环境交互，获得最优策略。实验结果表明，所提算法较基本Q学习的算法速度有所提升，且保持较高的带宽利用率，为提升SRv6效率提供了技术参考。

Abstract

With the rapid development of the Internet and communication technology

segment routing over IPv6(SRv6)

as an innovative network technology

simplifies network architecture and enhances the flexibility and programmability of the network topology

thereby meeting the diverse demands of various new services. In contrast

traditional static policies are unable to respond in real-time to the vast and complex changes in the network states

while network traffic is difficult to monitor and predict in real-time

and network optimization often requires trade-offs among multiple conflicting objectives. To address these problems

an SRv6 dynamic optimization framework that combined spatio-temporal mixed neural networks with enhanced reinforcement learning was proposed. It utilized long short-term memory (LSTM) and graph convolutional network (GCN) to capture the spatio-temporal features of network traffic

subsequently merging these features to derive network performance metrics. The enhanced Q-learning algorithm was then employed to interact with the environment

obtaining the optimal strategy. Experimental results show that compared with the basic Q-learning

this approach shows improved learning speed while maintaining a high bandwidth utilization rate

providing a technical reference for enhancing the efficiency of SRv6.

关键词

Keywords

references

Filsfils C , Dukes D , Previdi S , et al . Segment routing architecture [J ] . IEEE Communications Magazine , 2015 , 53 ( 7 ): 38 - 46 .

Polverini M , Cianfrani A , Caiazzi T , et al . SRv6 meets DetNet: a new behavior for low latency and high reliability [J ] . IEEE Journal on Selected Areas in Communications , 2025 , 43 ( 2 ): 448 - 458 .

Liang B C , Li F L , Zheng N G , et al . AutoSRv6: configuration synthesis for segment routing over IPv6 [J ] . IEEE Journal on Selected Areas in Communications , 2025 , 43 ( 2 ): 473 - 483 .

赵肖河 , 宋丽华 , 王晨 . 基于SRv6的卫星网络业务承载策略 [J ] . 无线电通信技术 , 2025 , 51 ( 2 ): 302 - 310 .

Zhao X H , Song L H , Wang C . Traffic carrying strategy for satellite networks based on SRv6 [J ] . Radio Communications Technology , 2025 , 51 ( 2 ): 302 - 310 .

Cianfrani A , Listanti M , Polverini M . Incremental deployment of segment routing into an ISP network: a traffic engineering perspective [J ] . IEEE/ACM Transactions on Networking , 2017 , 25 ( 5 ): 3146 - 3160 .

Wu Y T , Zhou J H . Dynamic service function chaining orchestration in a multi-domain: a heuristic approach based on SRv6 [J ] . Sensors , 2021 , 21 ( 19 ): 6563 .

Hao P , Yang F . SRv6 strategy based on network state and behavior prediction in 5G network [C ] // Proceedings of the 2023 9th International Conference on Computer and Communications (ICCC) . Piscataway : IEEE Press , 2023 : 1044 - 1048 .

Kreutz D , Ramos F M V , Veríssimo P E , et al . Software-defined networking: a comprehensive survey [J ] . Proceedings of the IEEE , 2015 , 103 ( 1 ): 14 - 76 .

Rusek K , Suárez-Varela J , Almasan P , et al . RouteNet: leveraging graph neural networks for network modeling and optimization in SDN [J ] . IEEE Journal on Selected Areas in Communications , 2020 , 38 ( 10 ): 2260 - 2270 .

Chen B , Zhu D , Wang Y W , et al . An approach to combine the power of deep reinforcement learning with a graph neural network for routing optimization [J ] . Electronics , 2022 , 11 ( 3 ): 368 .

Gunavathie M A , Umamaheswari S . Traffic-aware optimal routing in software defined networks by predicting traffic using neural network [J ] . Expert Systems with Applications , 2024 , 239 : 122415 .

Yalda K , Hamad D J , Tapus N . Network traffic prediction performance UsingLSTM [J ] . Romanian Journal of Information Science and Technology , 2024 , 27 ( 3/4 ): 336 - 347 .

王语乔 . 基于机器学习的城域网流量预测与业务重构技术研究 [D ] . 北京 : 北京邮电大学 , 2019 .

Wang Y Q . Traffic prediction and service reconstruction in metropolitan area networks based on machine learning [D ] . Beijing : Beijing University of Posts and Telecommunications , 2019 .

程伟强 , 李振斌 , 徐洪磊 , 等 . G-SRv6: 面向IPv6网络的增强型分段路由技术 [J ] . 电信科学 , 2020 , 36 ( 10 ): 12 - 20 .

Cheng W Q , Li Z B , Xu H L , et al . G-SRv6: enhanced segment routing technology for IPv6 networks [J ] . Telecommunications Science , 2020 , 36 ( 10 ): 12 - 20 .

Loreti P , Mayer A , Lungaroni P , et al . SRv6-PM: a cloud-native architecture for performance monitoring of SRv6 networks [J ] . IEEE Transactions on Network and Service Management , 2021 , 18 ( 1 ): 611 - 626 .

Bakhare S , Mohod D S W . Evaluating the performance and challenges of machine learning models in network anomaly detection [J ] . International Journal of Scientific Research in Science, Engineering and Technology , 2024 , 11 ( 3 ): 42 - 52 .

Shalini K S . Decentralized machine learning for dynamic resource optimization in wireless networks using reinforcement learning [J ] . Journal of Electrical Systems , 2024 , 20 ( 5 s): 2025 - 2033 .

Eppstein D . Finding the k shortest paths [J ] . SIAM Journal on Computing , 1998 , 28 ( 2 ): 652 - 673 .

Katz D , Kompella K , Yeung D . Traffic engineering extensions to OSPF [S ] . RFC 3630, 2003 .

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Large language model for cloud-network configuration audit and its application in IP network

Distributed reinforcement learning-based AUV 3D underwater current target tracking control algorithm

Efficient packet transmission over long-fat lossy links:research and development

Multipath intelligent transmission strategy for UAM dynamic networks

Intelligent matching task offloading scheme in green computing power networks

Related Author

XIE Liqin

ZHU Wenhui

SHU Dong

DING Yu

ZHANG Zheng

LIN Wentong

SU Yishan

LI Yun

Related Institution

China Telecom Co., Ltd. Jiangsu Branch

School of Electrical and Information Engineering, Tianjin University

Guangxi Engineering Research Center of Intelligent Vision and Collaborative Robotics

Guangxi Key Laboratory of ZHIYU Humanoid Robots

Engineering Research Center of Multi-Modal Information Intelligent Sensing, Processing and Application, Guangxi Universities

⁰