The Application Of System Integrity And Key Points Identification In Complex Networks

The system with complex structure and mechanism exists widely in our world. From a macro perspective, we can think the workings of nature or social development as a complex system, and based on this perspective,we can study the the complex inter-contact of nature and the law of system development of human society. The complex system can be described by the complex network which is abstracted into various forms, so it would be more convenient to conduct deeper and straightforward research on complex systems, such to promote the scientific research on the complex system. Therefore, it is important that we abstract the complex systems into complex networks, and use mathematical tools to study the complex network, and then apply the merits of related research to our society.Complex systems and complexity science has become one of today's most popular and most practical research tools.In this paper, the application of the system integrity and the definition of key points in complex networks is studied. First the integrity of navigation satellite system is defined, as well as the definition key points in navigation satellite network integrity. Secondly, this paper put forward a method to identify the key points of the integrity of navigation satellite based on grey correlation degree analysis. In the framework of the key points of the integrity of the navigation satellite, firstly the method of maximum correlation is applied to screen each evaluation index, and then the key points of the network is identified by using grey correlation analysis. Compared with the traditional algorithm, this method can effectively identify the key points in the network, and objectively reflect the importance of nodes in the network.Then, according to the software defined network control layer scalability problem, this paper proposes a method to solve the stack structure of software defined network based on complex network and combined with the integrity and the key points are the shortest routing strategy based on this method, reflected by adding node weights in calculating the route to the data layer of key nodes of key nodes. Compared to the classical Dijkstra shortest path routing algorithm, a network topology in the given conditions, if the number is far less than the given topology boundary node, routing strategy is proposed in this paper in terms of time overhead.