Optimization Of Traffic On Complex Networks

Since the scale-free and small-world properties were identified, complex networks havebeen applied with success to describe a wide range of natural and social systems in whichentities or people are connected by physical links or some abstract relationship. The increasingimportance of complex networks (such as the Internet, the power grid, the transportnetworks, and so on), upon which our society survives, calls for the need for high efficiencyin handling and delivering loads (information, energy, goods, and so on). The traffic on anetwork is one of the hot topics of recent research on complex networks.It is usually inexpensive and convenient by altering the routing protocol to enhancethe transmission efficiency of networks. In this thesis, some improved routing strategiesare proposed to enhance the traffic capacity of complex networks. Firstly, some previouslyproposed routing strategies are introduced. Based on their characteristics, some proposalsare given for the design and selection of routing strategies on complex networks.Based on the definition of betweenness centrality, the concept of expanding betweennessfor a certain route set is given, which gives an estimate of the traffic handled by thenode in the route set. Then an improved routing strategy is proposed to enhance the trafficefficiency of complex networks based on expanding betweenness centrality. Comparing resultsof previous routing strategies, such as the shortest path routing , the efficient routing,and the optimal routing based on betweenness, it is shown that the present improved routingperforms more effectively. The improved routing algorithm can be used effectively andconveniently in heterogeneous networks.Traffic dynamic is also investigated in weighted complex networks with limited packetdeliveringcapacity. An improved routing strategy is proposed to enhance the traffic capacityof complex networks based on expanding betweenness and capacity of links. Simulation resultsin two models with different capacity distributions indicate that the network capabilityin processing traffic is improved by the new routing strategy. Finally, a summary has been done for all discussion in the thesis. The research worksin further study on optimization of traffic in complex networks are presented.