Home > Published Issues > 2013 > Volume 8, No. 10, October 2013 >

A Simplified Traffic Generating Method for Network Reliability Based on Self-Similar Model

Yue Zhang2, Ning Huang1,2, Ning Hu3 , and Zhitao Wu2
1.Science & Technology Lab. on Reliability & Environmental Engineering, Beihang University, Beijing 100191, China
2.School of Reliability and Systems Engineering, Beihang University, Beijing 100191, China
3.The 5th Electronics Research Inst., the Ministry of Industry and Information Technology, Guangzhou 510610, China

Abstract—Similar traffic corresponding to the actual usage should be generated when doing the test or simulation for the network reliability. However, in the test or simulation for a large scale network, it is very difficult to generate the realistic network traffic, because it is complex and time-consuming to generate different traffic flows corresponding to businesses in the actual usage of the network. This paper proposes a simplified method of traffic generating based on the ON/OFF self-similar traffic model. We build the ON/OFF traffic model and infer the values of the model parameters. We generate the traffic based on the ON/OFF traffic model with the optimal parameters’ values and get the aggregate traffic which is similar to the target traffic. In the simplified traffic generating method, we solve the problem that the network traffic generating is too complex in the test or simulation for the large scale network and we save the time spent on the traffic generating. A case study shows that: the deviation between the aggregate traffic and target traffic is about 0.05, and compared with the actual traffic generating method, the time spent on traffic generating is reduced from 11s to 1s.

Index Terms—complex networks, traffic generating, self-similar, ON/OFF model

Cite: Yue Zhang, Ning Huang, Ning Hu, and Zhitao Wu, "A Simplified Traffic Generating Method for Network Reliability Based on Self-Similar Model," Journal of Communications, vol. 8, no. 10, pp. 629-636, 2013. doi: 10.12720/jcm.8.10.629-636