Analysis of communication network performance from a complex network perspective

In this paper we study the performance of communication networks from a network science perspective. Our analysis and simulation results reveal the effects of network structure, resource allocation and routing algorithm on the communication performance. Performance parameters, including packet drop rate, time delay, and critical generation rate, are considered. For efficient data transmission, the traffic load should be as uniformly distributed as possible in the network and the average distance of it should be short. We propose to use the node usage probability, which depends on both the network topology and routing algorithm, to characterize the traffic load distribution, and show that resource allocation based on the node usage probability outperforms the uniform and degree-based allocation scheme. On the basis of the proposal analysis and routing algorithms, we compare the performances of regular networks, scale-free networks, random networks, and the Internet constructed at the autonomous system (AS) level. Results from this study provide important insights into the relationship between the structural properties of communication networks and their performances.