基于小波的网络连接层流量模型

洪飞

基于小波的网络连接层流量模型

洪飞

北京航空航天大学计算机学院, 北京 100083

基金项目: 国家863基金资助项目(2001AA112110)

详细信息

中图分类号: TP 393.02
计量
- 文章访问数: 2450
- HTML全文浏览量: 127
- PDF下载量: 967
- 被引次数: 0
出版历程
- 收稿日期: 2004-07-06
- 网络出版日期: 2005-11-30

Connection-level network traffic model based on wavelet

Hong Fei

School of Computer Science and Technology, Beijing University of Aeronautics and Astronautics, Beijing 100083, China

摘要

摘要: 最近研究表明网络流量的统计特性在不同尺度上表现出很大的差异,由大量具有高斯性的网络流量和少量的突发性网络流量组成.为了更好地理解这种复杂的网络流量统计特性,引入多尺度分析工具小波,在多尺度分析的基础上提出一种新的基于小波的网络连接层流量模型,有机地结合了独立小波模型和多重分形小波模型的优点,即保持了独立小波模型在不同尺度对高斯性方差的拟合,同时保持了多重分形小波模型对突发性的拟合.真实网络流量的仿真结果表明这个模型能够灵活地适应网络流量在不同尺度下所表现出的高斯特性和突发行为.
- 小波 /
- 高斯 /
- 多重分形 /
- 流量模型 /
- 连接层
Abstract: Recent research shows network traffic exhibits drastically different statistics according to scales, including one component holding most traffic and being mostly Gaussian and the other absorbing virtually all the small scale bursts. For better understanding of this phenomenon, wavelet for multiscale analysis was introduced, a novel traffic model based on wavelet was proposed. Independent wavelet model was proved to be an ideal model for the Gaussian component, while the bursty component find a better match with the multifractal wavelet model, this novel traffic model combined the merits of independent wavelet model and multifractal wavelet model. Simulation results with the real traffic show this model is flexible and parsimonious to accommodate Gaussian as well as bursty behavior on different scales.
- wavelet /
- Gaussian /
- multifractal /
- traffic model /
- connection level

HTML全文

参考文献(1)

[1] Willinger W, Taqqu M, Sherman R, et al. Self similarity through high-variability:statistical analysis of ethernet LAN traffic at the source level[J]. IEEE/ACM Trans Networking, 1997, 5(1):71~86 [2] Riedi R H, Crouse M S, Ribiero V, et al. A multifractal wavelet model with application to TCP network traffic[J]. IEEE Trans Information Theory, 1999, 45(3):992~1018 [3] Erramilli A, Narayan O, Willinger W. Experimental queueing analysis with long-range dependent traffic[J]. IEEE/ACM Trans Networking, 1996, 4(2) :209~223 [4] Brichet F, Roberts J, Simonian A, et al. Heavy traffic analysis of a fluid queue fed by a superposition of ON/OFF sources[J]. Queueing System, 1996, 23:197~215 [5] Sarvotham S, Riedi R H, Baraniuk R. Connection level analysis and modeling of network traffic. In:Proc IEEE/ACM Network Measurement Workshop. San Francisco:ACM Press, 2001.99~103 [6] Sarvotham S, Wang Xin, Riedi R H, et al. Additive and multiplicative mixture trees for network traffic modeling. In:Proceeding of Acoustics, Speech, and Signal Processing, 2002. Orland:IEEE Press,2002.4040~4043 [7] Ma Sheng, Ji Chuanyi. Modeling video traffic in the wavelet domain. In:Proc of 17th Annual IEEE Conf on Comp Comm, INFOCOM. San Francisco:IEEE Press, 1998. 201~208 [8] Tian Xusheng, Wu Jie, Ji Chuanyi. A unified framework for understanding network traffic using independent wavelet models. In:Proc of 21th Annual IEEE Conf on Comp Comm INFOCOM. New York:IEEE Press, 2002.446~454 [9] Tian Xusheng. Heterogenous network traffic:understanding and modeling. Troy, New York:Rensselaer Polytechnic Institute, 2002 [10] Ma Sheng, Ji Chuanji. Modeling heterogeneous network traffic in wavelet domain[J]. IEEE Trans Networking, 2001, 9(5):634~649 [11] Daubechies. Ten lectures on wavelets[M]. New York:SIAM Press, 1992 [12] WAND network research group. Auckland-2 network traces. http://wand.cs.waikato. ac.nz/ wand/ wits/auck/4/, 2004

施引文献

资源附件(0)

访问统计