Study Of Information Congestion And Cascading Failure In Complex Communication Networks

A large number of complex systems in nature can be described by networks. In recent years the rising complex networks truly reflectes some important topology structure and statistical properties in the real world complex systems, and have become an important means of studying realistic networks. Since the pioneering work published by Watts and Barabasi, more and more attention has been focused on complex networks. With the rapid development of communication networks, large-scale communication networks with power-law degree distributions such as the Internet and World Wide Web have become some indispensable tools in our lives. In recent years, research of congestion control as well as cascading failures in communication networks has been a hot area which is concerned by domestic and foreign researchers. Consequently, physicists and engineers have paid more attention to a variety of dynamic processes occurred in the network structure which include the packet routing and congestion in complex communication networks, cascading failures and the robustness in complex networks. In the information circulation systems, such as the Internet, World Wide Web and global air networks, one problem which can not be ignored is how does the traffic dynamics affect the evolution of network structure and vice versa? Some of the previous works are mainly focused on how to drive the evolution of network structures with the growing of traffic, the other works mainly studied how different kinds of networks'topology structures can affect the traffic dynamics occurred on them. This paper mainly studies the information congestion and cascading failure in complex communication networks, which includes the following works.Firstly, we studied the information transmission on free-scale networks. Based on the local structural routing strategy, we proposed a new routing strategy integrating the degree and the idle degree of nodes, and studied the data transmission on BA scale-free network with using this strategy. A large number of numerical simulation shows that the BA scale free network will achieve its maximal communication capacity, the smallest average transmission time and the smallest network load withα=0.5. In comparison with the local routing strategy, this strategy can dramatically improve the network capacity and reduce the average transmission time of the data packets. Secondly, based on the routing policy with only considering the weight information on weighted networks and by introducing the dynamic free-degree of nodes, we proposed a new routing strategy which combines the weight information and dynamic idle degree of a node. Data transmission on BBV weighted networks was subsequently studied. A large number of numerical simulations shows that whenα=1, the network achieves the best performance with the maximal communication capacity, the smaller average transmission time and the smallest load. In comparison with the local link weights routing strategy, our strategy can dramatically improve the network capacity as well as reduce the network load.Finally, combined with the information transmission of complex network, we studied the cascading failure on a BA scale-free network after some max-degree nodes attacked when the network is in the free-flow state, where we use the local structural information routing strategy with giving the priority to those nodes with big degree. Simulation shows that when the network is in the free-flow state, both the robustness and the average efficiency of the network reduce with the increase of the packet generation rate. In critical congestion state, when 10 percent big nodes are attacked, about 50 percent nodes will become failure through cascading and when 20 percent big nodes are attacked, the network's efficiency will become zero. In the free-flow state with light load, attacking 40-50 percent big nodes will make the whole network collapse.