Traffic Optimization and Congestion Control Methods in SDN Networks Based on Artificial Intelligence
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
万方数据
文章深入研究基于强化学习的流量优化与拥塞控制方法在软件定义网络(Software Defined Network,SDN)中的应用.首先,详细阐述SDN网络的架构与原理.SDN网络的灵活性和可编程性为网络管理提供了全新的范式.其次,提出了一种基于强化学习的流量优化与拥塞控制方法,通过建模状态、动作、奖励等要素,实现网络流量智能调整.最后,在Mininet仿真环境中进行了实验验证.通过监测吞吐量、延迟、拥塞情况等性能指标,验证所提方法的有效性.实验结果表明,在网络性能方面,所提方法相较于传统方法取得了显著改善,具备更好的适应性和优化能力.
In this paper,the application of traffic optimization and congestion control method based on reinforcement learning in Software Defined Network(SDN)is deeply studied.Firstly,the architecture and principle of SDN network are elaborated in detail.The flexibility and programmability of SDN network provide a new paradigm for network management.Secondly,a method of traffic optimization and congestion control based on reinforcement learning is proposed,which realizes intelligent adjustment of network traffic by modeling elements such as state,action and reward.Finally,the experiment is carried out in Mininet simulation environment.The effectiveness of the proposed method is verified by monitoring performance indicators such as throughput,delay and congestion.The experimental results show that,in terms of network performance,the proposed method has significantly improved compared with the traditional method,and has better adaptability and optimization ability.
Software Defined Network(SDN)strengthen learningnetwork optimizationartificial intelligence
NETL
NSTL
万方数据
