基于机器学习的网络拥塞控制算法研究  

Research on Network Congestion Control Algorithms Based on Machine Learning

作　　者：单彩云 SHAN Caiyun(Communication University of China,Nanjing,Nanjing Jiangsu 211172,China)

机构地区：[1]南京传媒学院,江苏南京211172

出　　处：《信息与电脑》2024年第14期44-46,共3页Information & Computer

摘　　要：随着网络规模的扩大及其复杂化,网络拥塞问题加剧,传统的网络拥塞控制算法应对乏力。基于此,本文对基于机器学习的网络拥塞控制策略进行了探索,利用预测与优化能力精准识别和调整拥塞状态,旨在提升算法性能和用户体验。文章在综述了基于机器学习的网络拥塞控制算法研究现状后,提出了新的算法优化策略,阐述了核心思想、模型构建的思路,最后还对未来研究方向进行了展望,旨在为网络拥塞控制方向的研究者提供参考。With the expansion and complexity of network scale,the problem of network congestion has intensified,and traditional network congestion control algorithms are powerless to cope.Based on this,this article explores machine learning based network congestion control,using prediction and optimization capabilities to accurately identify and adjust congestion states,aiming to improve algorithm performance and user experience.After summarizing the current research status of machine learning based network congestion control algorithms,this article proposes new algorithm optimization strategies,elaborates on the core ideas and model construction ideas,and finally looks forward to future research directions,providing some reference for researchers in the field of network congestion control.

关 键 词：机器学习 网络拥塞控制 算法设计 性能优化 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]