| Nowadays, the high development of information network, especially the computer network, takes great convenience to people of the world. At the same time, the developing network also put forward a lot of challenge, exhibiting the expanding of the network size and the complexity of the network structure. Therefore, the traditional network fault management system is not satisfying requirement of development since the single function, complex operation and low-level efficiency. It becomes the focus of the researchers' attention that how to assure the network to work fleetly, safely and stably. In this thesis, the mobile agent technology is introduced into the fault management field. Through the researches on the key technology, it is possible for the network fault detection and diagnosed to be efficient, real-time and accurate.The primary researches and the achievements of this thesis include:(1) A network fault management system model based on mobile agent has been designed. The function of the component and the working principle of the module are minutely illuminated.(2) The network fault diagnosis method based on support vector machine (SVM) is proposed. The SVM is embeded into the diagnose module of the mobile agent. With the strong analysis and decision ability of the SVM, the diagnosis module can realize the distributing fault detection and distributing fault diagnosis simultaneously. The flexibility and intelligence of the management system is improved further.(3) The network fault management system based on mobile agent is implemented. The contrast between the network fault management system based on SNMP and mobile agent is specifically researched in theoretical analysis and experimental test. Compared to the SNMP model, the network fault management system based on mobile agent can manage the network more efficient, real-time and exactly, especially in the complex network since the activity, intelligence and flexibility of mobile agent. |
PDF Full Text Request