Abstract: Traffic models are crucial to network management and performance evaluation of communication networks. The modeling theory from a view of applied mathematics and the applications of traffic models from the point of view of networking gain wide interests of applied mathematicians, applied statisticians, physicists, and the scientists in electrical engineering and computer science.
There are two main categories in traffic modeling. One is statistical modeling based on random processes, in particular, fractal time series. The other is deterministic modeling by using bounded models. On the one hand, statistical models are useful to reveal the statistical properties of aggregated traffic, such as self-similarity, long-range dependence, and heavy-tailed distributions. On the other hand, deterministic models, instead of a statistical description, can be used to well characterize traffic at connection level with a set of bounds on the packet generation process for the purpose of guaranteed quality of service, such as guaranteed end-to-end delay for a specific connection. The theory and practice of both types of traffic models are desired.
Owing to the large impact upon the performance of communication network systems, two concepts are particularly discussed. One is the local irregularity of traffic, which may be characterized by the fractal dimension with the model of fractal time series or represented by the traffic burstiness with a bounded model. The other is the global persistence of traffic, which may be explained from the point of view of the long-range dependence if the model of fractal time series is used or interpreted as an average rate when one makes use of a bounded model.
This plenary lecture will give a talk with respect to the statistical modeling of traffic based on fractal time series and the deterministically bounded models. Applications of two types of models will also be mentioned. This speech was supported in part by the National Natural Science Foundation of China (NSFC) under the project grant numbers 60573125 and 60873264.
 



Brief Biography of the Speaker:
Short Biography of the Speaker: Ming Li, Ph.D. (City University of Hong Kong), is a professor at East China Normal University, PR. China. He was with the School of Computing, National University of Singapore, before joining East China Normal University in 2004. His research areas relate to applied statistics and signal processing with the recent interests in fractal time series and time-frequency analysis, computer science currently focusing on network traffic modeling and network security, and measurement & control in the aspects of error analysis and optimal control. He has published over 90 papers in international journals and international conferences in those areas. His research is supported in part by the National Natural Science Foundation of China (NSFC) under the project grant numbers 60573125 and 60873264.
Dr. Ming Li is an editorial member of Journal of Universal Computer Science, the guest editor (with Pierre Borgnat) of Telecommunication Systems by Springer for the Special Issue on Traffic Modeling, Its Computations and Applications in 2008, and the guest co-editor (with Carlo Cattani and Cristian Toma) of Mathematical Problems in Engineering by Hindawi Publishing Corporation for the Special Issue on Short Range Phenomena: Modeling, Computational Aspects and Applications in 2008. Ming Li is also the editor-in-chief of two international journals, International Journal of Electronics and Computers, and International Journal of Engineering and Interdisciplinary Mathematics.