...
首页> 外文期刊>IEEE Signal Processing Magazine >Long-range dependence and heavy-tail modeling for teletraffic data
【24h】

Long-range dependence and heavy-tail modeling for teletraffic data

机译:远程交通数据依赖和重尾模型

获取原文
获取原文并翻译 | 示例
           

摘要

The analysis and modeling of computer network traffic is andaunting task considering the amount of available data. This is quitenobvious when considering the spatial dimension of the problem, since thennumber of interacting computers, gateways and switches can easily reachnseveral thousands, even in a local area network (LAN) setting. This isnalso true for the time dimension: Willinger and Paxson (see Ann.nStatist., vol.25, no.5, p.1856-66, 1997) cite the figures of 439 millionnpackets and 89 gigabytes of data for a single week record of thenactivity of a university gateway in 1995. The complexity of the problemnfurther increases when considering wide area network (WAN) data. Innlight of the above, it is clear that a notion of importance for modernnnetwork engineering is that of invariants, i.e., characteristics thatnare observed with some reproducibility and independently of the precisensettings of the network under consideration. In this tutorial article,nwe focus on two such invariants related to the time dimension of thenproblem, namely, long-range dependence, or self-similarity, andnheavy-tail marginal distributions
机译:考虑可用数据量,对计算机网络流量进行分析和建模是一项艰巨的任务。考虑到问题的空间范围时,这并不是很明显,因为这样一来,即使在局域网(LAN)设置中,交互的计算机,网关和交换机的数量也可以轻松达到数千个。时间维度也是如此:Willinger和Paxson(请参阅Ann.nStatist。,第25卷,第5期,p.1856-66,1997年)引用了单周记录的4.39亿个数据包和89 GB的数据。这是1995年大学网关的活动记录。考虑广域网(WAN)数据时,问题的复杂性进一步增加。综上所述,很显然,对于现代网络工程来说,一个重要的概念是不变性,即具有某些可再现性且与所考虑的网络的精确设置无关的特性。在本教程文章中,我们将重点讨论与问题的时间维相关的两个这样的不变量,即长距离依赖性或自相似性以及重尾边际分布

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号