A Fitness Verification of Time Series Models for Network Traffic Predictions 


Vol. 29,  No. 2, pp. 217-227, Feb.  2004


PDF
  Abstract

With a rapid growth in the Internet technology, the network traffic is increasing swiftly. As for the increase of traffic, it had a large influence on performance of a total network. Therefore, a traffic management became an important issue of network management. In this paper, we study a forecast plan of network traffic in order to analyze network traffic and to establish efficient correspondence. We use time series forecast models and determine fitness whether the model can forecast network traffic exactly. In order to predict a model, AR, MA, ARMA, and ARIMA must be applied. The suitable model can be found that can express the nature of traffic for the forecast among these models. We determines whether it is satisfied with stationary in the assumption step of the model. The stationary can get the results by using ACF(Auto Correlation Function) and PACF(Partial Auto Correlation Function). If the result of this function cannot satisfy then the forecast model is unsuitable. Therefore, we are going to get the correct model that is to satisfy stationary assumption. So, we proposes a way to classify in order to get time series materials to satisfy stationary. The correct Prediction method is managed traffic of a network with a way to be better than now. It is possible to manage traffic dynamically if it can be used.

  Statistics
Cumulative Counts from November, 2022
Multiple requests among the same browser session are counted as one view. If you mouse over a chart, the values of data points will be shown.


  Cite this article

[IEEE Style]

S. Jung, D. Kim, Y. Kwon, C. Kim, "A Fitness Verification of Time Series Models for Network Traffic Predictions," The Journal of Korean Institute of Communications and Information Sciences, vol. 29, no. 2, pp. 217-227, 2004. DOI: .

[ACM Style]

Sangjoon Jung, Dongju Kim, Younghun Kwon, and Chonggun Kim. 2004. A Fitness Verification of Time Series Models for Network Traffic Predictions. The Journal of Korean Institute of Communications and Information Sciences, 29, 2, (2004), 217-227. DOI: .

[KICS Style]

Sangjoon Jung, Dongju Kim, Younghun Kwon, Chonggun Kim, "A Fitness Verification of Time Series Models for Network Traffic Predictions," The Journal of Korean Institute of Communications and Information Sciences, vol. 29, no. 2, pp. 217-227, 2. 2004.

Search
(IN TITLE, AUTHOR, ABSTRACT,KEYWORDS)

Advanced Search ▼

POPULAR KEYWORDS
(TOP 10 KEYWORDS)

Recent Publications
(LAST 3 YEARS)






pISSN : 1226 - 4717
eISSN : 2287 - 3880