基于模糊控制的自适应瓶颈带宽测量与定位采样方法  

ADAPTIVE BOTTLENECK BANDWIDTH MEASUREMENT AND LOCATION METHOD BASED ON FUZZY CONTROL

作　　者：贾民政[1] 朱元忠[1,2] 张仙妮[1] 

机构地区：[1]北京工业职业技术学院信息工程系,北京100042 [2]中国矿业大学(北京)计算机科学与技术系,北京100083

出　　处：《计算机应用与软件》2013年第3期229-232,共4页Computer Applications and Software

基　　金：2012北京工业职业技术学院科研课题(bgzyky201249)

摘　　要：主动测量中探测包的发送方式即采样方法直接影响了测量的准确性。介绍常见的采样方法并对它们进行了分析。针对目前现有采样方法存在的不足,重点研究瓶颈带宽测量与定位中的采样方法,使用消融包列作为探测包,提出一种基于模糊控制的自适应采样方法。该方法能够根据网络流量的变化自适应地调整探测包的发送时间间隔。通过模拟试验对该方法进行了验证,并与其他方法进行了比较,结果表明该方法在降低测量噪声、提高测量准确性方面有较大优势。Sending way of probe package in active measurement,i.e.the sampling method directly affects the measurement accuracy.Common sampling methods are introduced and analysed in this paper.In light of the shortcomings of current sampling methods,we emphatically study a sampling method in bottleneck bandwidth measurement and location,and put forward a fuzzy control-based adaptive sampling method by using meltable packet train column as the probe packet.This method can adaptively adjust the probe packets sending interval according to network traffic variations.The method is verified through simulation experiments,and compared with other methods.Results show that this method has great advantage in reducing the measurement noise and improving the measurement accuracy.

关 键 词：模糊控制 自适应采样 瓶颈带宽 网络测量 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]