电信科学 ›› 2022, Vol. 38 ›› Issue (8): 86-100.doi: 10.11959/j.issn.1000-0801.2022172

• 研究与开发 • 上一篇    下一篇

基于LSTM流量预测的路由规划和切换

诸葛斌, 王林超, 宋杨, 邵瑜, 董黎刚, 蒋献   

  1. 浙江工商大学信息与电子工程学院(萨塞克斯人工智能学院),浙江 杭州 310018
  • 修回日期:2022-05-27 出版日期:2022-08-20 发布日期:2022-08-01
  • 作者简介:诸葛斌(1976- ),男,博士,浙江工商大学信息与电子工程学院教授,浙江省新世纪151 人才工程第三层次培养人员,主要研究方向为网络与通信技术、互联网技术和网络安全
    王林超(1996- ),男,浙江工商大学硕士生,主要研究方向为软件定义网络
    宋杨(1996- ),男,浙江工商大学硕士生,主要研究方向为软件定义网络
    邵瑜(1994- ),男,浙江工商大学硕士生,主要研究方向为软件定义网络
    董黎刚(1973- ),男,博士,浙江工商大学信息与电子工程学院党委书记、教授、硕士生导师,主要研究方向为新一代网络和分布式系统
    蒋献(1988- ),男,浙江工商大学实验师,主要从事数字电路实验和模拟电路实验工作
  • 基金资助:
    国家自然科学基金资助项目(61871468);浙江省重点研发计划项目(2020C01079);浙江省重点研发计划项目(2021C01036);浙江省新型网络标准与应用技术重点实验室(2013E10012)

Routing planning and handoff based on LSTM traffic prediction

Bin ZHUGE, Linchao WANG, Yang SONG, Yu SHAO, Ligang DONG, Xian JIANG   

  1. School of Information and Electronic Engineering (Sussex Artificial Intelligence Institute), Zhejiang Gongshang University, Hangzhou 310018, China
  • Revised:2022-05-27 Online:2022-08-20 Published:2022-08-01
  • Supported by:
    The National Natural Science Foundation of China(61871468);The Key Research and Development Program of Zhejiang under Grant(2020C01079);The Key Research and Development Program of Zhejiang under Grant(2021C01036);The Zhejiang Provincial Key Laboratory of New Network Standards and Technologies(2013E10012)

摘要:

摘 要:随着近年网络技术的不断发展和演进,用户对网络资源的需求体量越来越大,内容越来越复杂,传统的网络架构很难满足当前灵活的网络需求。软件定义网络的革命性技术思想赋予了网络可编程性和可演进性,实现网络资源更灵活的管理。为了提高网络资源利用率,提出了基于LSTM模型的流量预测,为路由规划提供理论依据。通过提前预测网络流量,在流量激增之前进行防御措施保证网络的安全性和稳定性。实现基于流量预测的路由规划、无损路径切换等功能。

关键词: 软件定义网络, 流量预测, 路由规划, 路径切换, LSTM模型

Abstract:

With the continuous development and evolution of network technology in recent years, the user’s demand for network resources is becoming larger and larger, and the content is becoming more and more complex.The traditional network architecture is difficult to meet the current flexible network needs.The revolutionary technology of SDN endows the network with programm ability and evolvability, and realizes more flexible management of network resources.In order to improve the utilization of network resources, a traffic prediction agent based on LSTM model was proposed, which provides a theoretical basis for routing planning.By predicting the network traffic in advance, we can take defensive measures before the traffic surge to ensure the security and stability of the network.The function of route planning and lossless path switching based on traffic prediction was realized.

Key words: software defined network, traffic prediction, routing planning, paths witching, LSTM model

中图分类号: 

No Suggested Reading articles found!