Modeling And Analyzing Of Traffic Behavior In Network

In recent years, with the rapid growing of users, the scale of network expands dramatically. At the same time, the services conveyed in the network present the trend of diversity and integration. The large-scale structure not only results the difficulty for measurement, but makes the state of network changing from time to time. There is still little effective approach to grasp the going of traffic flows and network performance.In order to meet users’ demand and improve the network performance, massive work has been done. It is mainly divided into three parts:1. some work focuses on the one or more targets to find the optimal solution according to the macro-characteristics of traffic flows.2. Some work study on the network topology to find the influence on network behavior.3. Other work provides network performance model.However, in our view, it is still inadequate to understand the network behaviors. In this article, we are starting with the micro-scenario, from bottom to up to gradually form the macro-case. The coupling interaction among traffic flows is originated from contradiction between user demands and network resources. We propose our traffic model to reveal the correlations between traffic flows in network. We claims that our model that incorporate hard technological constraints on router and link bandwidth and connectivity, together with abstract models of user demand and network performance, can successfully presents the coupling behaviors. According to our experiments, the network effects include diffusion, fluctuation, and superimposition. Furthermore, in the congestion state, the feedback mechanism results the worse of network performance, which deteriorates the quality of services for traffic flows. We name this phenomenon as’Matthew Effect’. The actual result is obtained by integrating the case with feedback mechanism and without feedback mechanism. Finally, the coupling model is further extended to get the potential energy model, which provides unified standard to analyze and evaluate the network performance.