A New Topology Detecting Method And Its Application In Vector Network

With the rapid development of network technology, network technology is difficult to keep up with people’s requirements for faster, safer and more convenient network. Research on the new generation network technologies never ceases, and as a result,new network technologies also spring up like endless streams. In this case, vector network technology emerges as the times required.Compared with the networks which already exists, it has plenty of advantages such as QoS(Quality of Service), infinite scalability, supporting ability, strong safety and so on. In addition,thanks to the functional separation of switching and routing in the vector network, switching nodes only need to complete a simple function, thereby simplifying the equipment costs, reducing the network maintenance cost objectively.In vector network, data plane and control plane are separated. In order to reduce the costs of network equipments, functions of switching equipments in data plane are only confined to essential data-switching and the most simplified signaling-appending.Consequently,typical mechanism should be designed to acquire the topology structure according to the specialities mentioned above. In this dissertation,we come up with a method of ending systems initiatively probing for the sake of obtaining the network structure,of which principles are as following.Nodes in vector network answer simply when receiving detecting packets from specific ending system which is selected as a Leader(also called NALP,abbreviation of Node Anwsering on Leader Probing).Leader integrates information based on the feedback signals,and finally achieves the topology structure and routing process. What’s more,we also design the NALP protocol and stipulate the signal formation so as to accomplish the topology detection process. In addition, problem of detecting network topology which contains different kinds of simple switches has also been solved thanks to the combination of NALP and GSDM(guessing and scanning detection method).Last but not least,we realize the simulation system which detects the topology of middle-scaled network.The method proposed in this dissertation is based on the unique features of data transmitting of vector network, so it does not need the support of any additional network protocols. Hence the NALP method has incredible versatility,which also means that it has good portability (applicable to any vector network structure). At the end of this dissertation, the author designs a simulation system by programming with C++,and successfully acquires the topology information of different network structures, and thus fully proving accuracy and integrity of the NALP method.