互联网 qkzz.net
全刊杂志网:首页 > 女性 > 文章正文
刊社推荐

包对探测统计模型与测量精度分析


□ 刘 俊 张大方 金军航

   (湖南大学 a.计算机与通信学院;b.软件学院, 长沙 410082)
  摘 要:包对模型(PGM)是探测可用带宽的主要方法。该方法存在以下问题,影响其推广应用:模型没有精确的数学模型描述,原有模型假设流量包长看做无限小,与实际网络情况不符合;没有定量方法计算探测精度、样本容量,给测量带来不便;没有分析互联网数据包突发性对测量带来的影响。为解决以上问题,通过对包对模型进行数学建模的方法,证明了模型正确性;定量分析了测量精度;确定了样本数计算公式;分析了数据包突发性对探测造成的影响;最后对以上分析进行了仿真实验验证,提高了可用带宽测试技术的精度和可信度。
  
  关键词:计算机网络;可用带宽探测;包对模型
  中图分类号:TP393.06文献标志码:A
  文章编号:1001-3695(2009)04-1438-05
  
  Statistical modeling and accuracy analysis of available bandwidth measurement
  
  LIU Juna,ZHANG Da-fangb,JIN Jun-hangb
  (a.School of Computer & Communication,b.School of Software, Hunan University, Changsha 410082, China)
  Abstract:Available bandwidth measurement is widely applied in a dependable network.This paper proposed probe gap model(PGM) as a lightweight and fast available bandwidth estimation method. But a big problem of PGM was its poor accuracy in certain network conditions. The accuracy could’t be trusted. An important assumption behind the PGM was that the model should be statistical robust even the cross traffic follows do NOT follows fluid model. But no verification could prove this assumption. These problems made tools based on PGM could’t be widely used.Constructed a statistical model for PGM. Based on this insight,presented mathematical verification and accuracy analysis for PGM. And also use modeling, measurements to characterize the interaction between probing packets and bursting network traffic.NS simulation shows the modeling and the accuracy analysis is successful. By the model,can know how much can trust the probing results. ......
很抱歉,暂无全文,若需要阅读全文或喜欢本刊物请联系《计算机应用研究》杂志社购买。
欢迎作者提供全文,请点击编辑
分享:
 

了解更多资讯,请关注“木兰百花园”
分享:
 
精彩图文


关键字
支持中国杂志产业发展,请购买、订阅纸质杂志,欢迎杂志社提供过刊、样刊及电子版。
关于我们 | 网站声明 | 刊社管理 | 网站地图 | 联系方式 | 中图分类法 | RSS 2.0订阅 | IP查询
全刊杂志赏析网 2017