一种自适应抽样方法在网络管理中的应用研究  

Application Research on Self-adaptive Sampling Method in Network Management

作　　者：要趁红[1,2] 裴昌幸[1] 王民[2] 

机构地区：[1]西安电子科技大学综合业务网理论及关键技术国家重点实验室,西安710071 [2]西安建筑科技大学信息与控制工程学院,西安710055

出　　处：《计算机工程》2013年第5期1-4,共4页Computer Engineering

基　　金：国家自然科学基金资助项目(61072067);高等学校学科创新引智计划基金资助项目(B08038)

摘　　要：在智能网络管理系统中,传统的系统抽样方法无法适应对现代信息网络的实时监测要求。为克服该问题,提出一种模糊自适应抽样方法。以模糊控制原理为基础,通过设定的隶属函数和模糊规则动态调整抽样间隔,在尽可能减小对网络时延和带宽影响的情况下进行异常探测和网络瓶颈检测。实验结果表明,在相同的误差条件下,该方法比系统抽样方法所需样本数少69%,若采用相同的样本数,其抽样误差比系统抽样方法低54%。Regarding the intelligent network management system, the traditional systematic sampling method is unable to monitor to the modem information network in real-time. In order to solve this problem, this paper presents a fuzzy self-adaptive sampling method. Based on the fuzzy control principle, the method can dynamically adjust the sampling interval using the set of membership functions and fuzzy rules, and it can detect the network anomaly and network bottlenecks in the minimize of affecting network latency and network bandwidth. Experimental results show that this method is less than the system sampling method 69% based on the same error condition, and the errors of fuzzy adaptive sampling method are less than the system sampling method 54% in the equal sample count.

关 键 词：网络管理系统 流量监测 流量预测 模糊控制 自适应抽样 系统抽样 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]