Detrended fluctuation analysis of IP-network traffic using a two-dimensional topology map
✍ Scribed by Masao Masugi
- Publisher
- Elsevier Science
- Year
- 2004
- Tongue
- English
- Weight
- 255 KB
- Volume
- 337
- Category
- Article
- ISSN
- 0378-4371
No coin nor oath required. For personal study only.
✦ Synopsis
This paper describes an analysis of IP-network tra c in terms of the time variation of self-similarity. To get a comprehensive view of network tra c conditions in analyzing the degree of long-range dependence (LRD) of IP-network tra c, this paper used a self-organizing scheme-based topology map, which provides a way to map high-dimensional data onto a lowdimensional domain. Also, in the LRD-based analysis, this paper employed detrended uctuation analysis (DFA), which is applicable to the analysis of long-range power-law correlations or LRD in apparently non-stationary time-series signals. Based on sequential measurements of IP-network tra c at a 100-Mbps point of interface between an access provider and the Internet, this paper derived corresponding values for the LRD-related parameter of the tra c. In training the topology map, this paper used three parameters: the value, average throughput, and a parameter that re ects the degree of non-stationarity for each measured data set. We visually conÿrmed that the tra c data could be projected onto the topology map in accordance with the tra c properties, resulting in a combined depiction of the e ects of the degree of LRD and other factors. The proposed method can deal with multi-dimensional parameters, projecting its results onto a two-dimensional space in which the positions of the projected data give us with an e ective depiction of network conditions at di erent times.