NETWORK BURST MONITORING AND DETECTION BASED ON FRACTAL DIMENSION WITH ADAPTIVE TIME-SLOT MONITORING MECHANISM

Authors

Hsiao-Wen Tin, Department of Electrical Engineering, National Taiwan Ocean University, Keelung, Taiwan. R.O.C.
Shao-Wei Leu, Department of Electrical Engineering, National Taiwan Ocean University, Keelung, Taiwan. R.O.C.Follow
Shun-Hsyung Chang, Department of Microelectronics Engineering, National Kaohsiung Marine University, Kaohsiung, Taiwan, R.O.C.
Gene Eu Jan, Department of Electrical Engineering, National Taipei University, Taipei County, Taiwan, R.O.C.

Abstract

This study proposed an approach to measure the burstiness of network traffic based on fractal dimensions (FDs). By definition, burstiness is the degree of variation in network traffic. This study defined two types of FDs: (1) the FD of network traffic that describes the flow variation of network traffic, and (2) the FD of the range that describes the degree of flow dispersal. The proposed method uses an adaptive time-slot monitoring mechanism to monitor the network. The relevant FDs are derived from measurements obtained during each time slot in a monitoring window. This study conducted experiments using NS2 simulation data. The experimental results indicate that the proposed method can effectively measure the burstiness of network traffic. The method provides a meaningful way to describe the variation of network traffic and reduces monitoring overhead by using an adaptive time-slot monitoring mechanism.

Recommended Citation

Tin, Hsiao-Wen; Leu, Shao-Wei; Chang, Shun-Hsyung; and Jan, Gene Eu (2013) "NETWORK BURST MONITORING AND DETECTION BASED ON FRACTAL DIMENSION WITH ADAPTIVE TIME-SLOT MONITORING MECHANISM," Journal of Marine Science and Technology: Vol. 21: Iss. 6, Article 9.
DOI: 10.6119/JMST-013-0516-1
Available at: https://jmstt.ntou.edu.tw/journal/vol21/iss6/9