A novel complex network model for Internet traffic flows

Nowadays, complex network theory has been widely used in different fields. Many models were proposed for different networks such as Internet or World Trade Web. In this paper, we focus on Internet traffic flows and construct flow graphs from complex network perspective. We propose a novel complex network model for the flow graphs. Besides growth and preferential attachment characteristics of BA model, our model also considers three different kinds of edge growth: between two new nodes, between the new one and the old one, between two old nodes. The probabilities of the three kinds of edge added are changing when the graph grows in our model. We analyze the theoretical degree distribution of our model and obtain the power law exponent, which is adjustable. Our simulations show that the degree distribution of our model is close to the theoretical analysis, and our model matches the flow graphs well.