基于NS2的网络拥塞控制算法实验仿真与分析  

Experimental simulation and analysis of network congestion control algorithm based on NS2

作　　者：张奎 ZHANG Kui(School of Computer Science and Technology,Kashi University,Kashi 844006,China)

机构地区：[1]喀什大学计算机科学与技术学院,新疆喀什844006

出　　处：《青海师范大学学报（自然科学版）》2021年第2期36-41,共6页Journal of Qinghai Normal University(Natural Science Edition)

基　　金：新疆高校科研计划项目(XJEDU2019Y041);喀什大学校内课题(2020)2738.

摘　　要：针对网络资源需求大于可用资源引发的网络拥塞,进而导致网络服务质量(Quality of Service,QoS)下降的问题,在分析网络拥塞控制算法的基础上,利用NS2仿真软件设计并仿真了不同类型的网络拓扑.仿真结果表明Vegas算法在同构型网络环境下具有高平均吞吐量和低时延、无丢包性能,Tahoe、Reno、Newreno和Sack算法性能表现一致;而在异构型网络中由于竞争机制,Vegas算法性能相对较差,但其他算法会周期性的遭遇网络拥塞以及网络稳定性较差的情况.该结果为不同环境下的网络拥塞算法选择提供了依据.Aiming at the problem of network congestion caused by network resource demand greater than available resources,which leads to the decline of network quality of service(QoS),based on the analysis of network congestion control algorithms,NS2 simulation software was used to design and simulate different types of networks Topology.The simulation results show that the Vegas algorithm has high average throughput,low latency,and no packet loss in the homogeneous network environment.The performance of the Tahoe,Reno,Newreno,and Sack algorithms are consistent.However,due to the competition mechanism in heterogeneous networks,Vegas algorithm performance was relatively poor,but other algorithms will periodically encounter network congestion and poor network stability.The results provide a basis for network congestion algorithm selection in different environments.

关 键 词：拥塞控制 同构型网络 异构型网络 实验仿真 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]