BA Scale-free Network Model's Application And Extending

Recent years, along with the more and more deeper acquaintance of complex systems as well as the research of relatively independent individual's properties and structure on different layers of them and difficulties meeting when analyzing them, people more and more found the closed relations existing in their subsystems as well as all sorts of immanent closed relationships in which and the importance risen by which to the constituting of the whole system. So, assisted by the high computing capability of the computer, through observing some complex huge systems and their subsystems such as ecological system, internet system, social system, vital system and so on, people begin to attach importance to interaction and unity of them. Furthermore, the all above foment the complex networks research in lots of different scientific domains. Complex networks theory is such a result that issue from the observation to relative simple topology structure of networks in different domains. And, recent several years, many research show that many complex systems possess the scale-free property, that is, the mass of nodes have only a few connections but abundant connections are possessed by minority nodes. So, to deeply know about complex systems, it is very essential to study the scale-free networks.The discovery of scale-free property breaks through the restriction of random network model and which makes us know about many different kinds of structure of complex systems all obey some same fundamental laws, and also makes us know about it is impossible to study the current rules of system networks' structure. In addition, scale-free property also makes us can hold a deeper study to system structure base on topological properties of complex networks. The paper does a summary and broaden research to the BA model which is the typical representation of scale-free networks from the following several aspects:1. Scale-free network is a newest risen implement for system analyzing, so it hasn't be known about very well. Just for this reason, it is very necessary to present asummary of whose forming background , forming mechanism, range of application, statistical properties and applied value. What is more, it is also the theory fundamental to be on practical applying.2. Protein folding structure prediction is a cosmopolitan tickler. How to accurately apply energy function and how to choose the most proper conformation from hundreds of millions of conformations are all needed to be settled by a implement which can deal with large-scale problem high efficiently. But, so far, there isn't even one method whose prediction is enough precise. In this paper, we putforward a trailblazing study------exert complex networks theory to protein foldingstructure prediction and from which we have got a more precise result of coarse granulated model.3. The exiting typical representation of scale-free networks------the BA model isabstracted from many real networks. But, after some application, it is found the model has more severe defects. So, to make it satisfy topological structures of more real systems, the paper does some extending to BA scale-free network model's algorithm. And after a comparison of their behavior base on the characters of the static statistics, it can be easily found that the topological structure and development of the extending model is more closed to real networks. Besides, it can improve networks' robustness.