摘要
为了采用有物理意义的参数,在多个时间尺度下对现代网络业务流进行建模,考察了网络流量长相关(LRD)的基本特征,提出了业务流中满足的几个基本定律,揭示了在流量中描述LRD定量关系的参数cγ(或cf)、流量聚集级m、平均到达率l之间的基本关系,并通过实际测量的数据加以验证.所提出的定律表明:(1)流量确实存在LRD行为;(2)平均到达率(TCP和IP流等)与参数cγ(或cf)存在一定的统计定量关系.
In order to model the modern network traffics in multi-time scales which enables the parameters to have physics meanings, the basic characteristics of the long-range dependence （LRD） in network traffics are investiga- ted, and several essential laws required by modern network traffics are proposed. The relationships among the parameters such as cγ or cf that describes the quantitative LRD, the traffic aggregated level m and the average arrival rate λin long-range dependence traffic are then revealed, followed by verification with real measured data. The pro- posed laws include the opinions that there exists LRD in network traffics indeed, certain statistically-quantitative relationships between the average arrival rate involving the TCP and IP levels, etc. and the parameter cγ or cf.
出处
《华南理工大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2007年第5期54-59,共6页
Journal of South China University of Technology(Natural Science Edition)
基金
广东省自然科学基金资助项目(3139)
关键词
因特网
业务流
自相似
长相关
流量模型
Internet
network traffic
self-similarity
long-range dependence
traffic model
