Evolution Analysis Of Internet Macroscopic Topology Of IPv4 And IPv6 And Its Modeling

Internet is atypical complex network,and that exploring its inner law has been concemed.However,it has a large scale and complex structure,it is not enough to improve technology of network.We should research topology of overall Internet.and explore the general rules of Internet.Only has a more in-depth understanding of macroscopic network topology,we can find the evolution mechanism,and then modeling.It has a great significance to design and implementation of the next generation Internet,and promoting performance of Internet.Firstly,the thesis introduces and analyzes basic characteristics and advanced features of Internet.Analyzing the four basic models of complex network.Basic characteristics of every model are computed.To be a theoretical foundation for modeling Internet.Secondly,to define that measurement of Internet topology and deduction index.the thesis contrast the IPv4 to IPv6 data that obtain from the CAIDA database and select san-us node from 2010 to 2014.and found the metabolic that a common law:the characteristics of period growth.To define metabolism according to the common law.The metabolism is a breakthrough.To contract two networks through growth of network,structure of node degree and connection preference.Then find the common law.Metabolic is generated by the large number of nodes in the network,and the network is period growth.It is not related with the death nodes and stable nodes.Analyzing the structure of node degree found that maximum degree of network is very small The degree of new nodes is small.and found the leaf nodes are about 40%-50%of all nodes.It determines the changes of network,To find the birth nodes prefer to connect node of large degree,through analyzing the preferential attachment.New edges are two forms that new node to old node and old node to old node.and it is ignored that new node to new node.Finally,contrasting the classical model of Internet,and find their respective characteristics.I adopt the PFP model as the original model accord to characteristics of the model and the analysis results,and combined with real network to improve it.To inheritance new node and new edges of inner interactive growth,and improve new nodes and new edges growth.To make it according with evolution of real network.To contrast real network,PFP model and PG-PFP model that found PG-PFP is the most accord with evolution law of real network.