Generating traffic time series based on generalized cauchy process


Li, Ming and Lim, S. C. and Feng, Huamin (2007) Generating traffic time series based on generalized cauchy process. In: Computational Science – ICCS 2007. Lecture Notes in Computer Science (4487). Springer Berlin Heidelberg, pp. 374-381. ISBN 978-3-540-72583-1

[img] Text
Restricted to Repository staff only

Download (230kB)


Generating traffic time series (traffic for short) is important in networking, e.g., simulating the Internet. In this aspect, it is desired to generate a time series according to a given correlation structure that may well reflect the statistics of real traffic. Recent research of traffic modeling exhibits that traffic is well modeled by a type of Gaussian process called the generalized Cauchy (GC) process indexed by two parameters that separately characterize the self-similarity (SS), which is local property described by fractal dimension D, and long-range dependence (LRD), which is a global feature that can be measured by the Hurst parameter H, instead of using the linear relationship D = 2 − H as that used in traditional traffic model with a single parameter such as fractional Gaussian noise (FGN). This paper presents a computational method to generate series based on the correlation form of GC process indexed by 2 parameters. Hence, the present model can be used to simulate realizations that flexibly capture the fractal phenomena of real traffic for both short-term lags and long-term lags.

Item Type: Book Section
Additional Information: Book Subtitle: 7th International Conference, Beijing, China, May 27 - 30, 2007, Proceedings, Part I
Subjects: T Technology > T Technology (General)
Divisions: Faculty of Engineering (FOE)
Depositing User: Ms Rosnani Abd Wahab
Date Deposited: 04 Feb 2014 04:23
Last Modified: 04 Feb 2014 04:23


Downloads per month over past year

View ItemEdit (login required)