Font Size: a A A

Research And Implementation Of Traffic Fault Diagnosis Expert System

Posted on:2014-02-01Degree:MasterType:Thesis
Country:ChinaCandidate:L L ZhangFull Text:PDF
GTID:2248330398951269Subject:Computer application technology
Abstract/Summary:PDF Full Text Request
Network optimization has become an important method to improve the networkperformance and enhance the user’s perception in the field of mobile communication. As oneof the important branches, the traffic failure analysis has been required to study more anddeeply. However, in the case of the large amount of data, the existing analysis method hasbeen difficult to meet the demand.The method of the traffic fault diagnosis in paper, combined with advanced ideas ofartificial intelligence and intelligent diagnostic method, improves the original analysis methodand enhances the work efficiency gives full play to the advantage of the expert system andprovides an effective method and theory for the traffic failure analysis in the mobilecommunication field.In the traffic fault diagnosis expert system design and implementation, the reasoningmethod combined with case-based reasoning and rule-based reasoning was adopted, whileusing their respective advantages and making up for their deficiencies. In the process of thecase-based reasoning, the object-oriented approach was used to express instance, the KNNalgorithm was used to strike instance similarity and the hierarchical analysis method was usedto calculate weights.At the same time, according to the particularity of the traffic failure analysis, thealgorithm in this paper was made better appropriately based on the KNN algorithm and thesimilarity frequency dependence was added. In the process of the rule-based reasoning, the traditional production rules was applied to implement the rules store. The depth-first searchstratagem was used to complete the rule matching and search; and the method according tothe time sequence was used to resolve the conflict. As for the hybrid reasoning process, thetwo methods CBR which is prioritized and RBR which is complementary were combinedand supplemented to provide a theoretical basis for applied research of traffic failure analysisin the mobile communication field.Finally, the software development platform combined with VS2008+SQL Server2008+Devexpress was used to achieve the development of the traffic failure analysis system.Besides, the experimental effect after using the system was compared with the effect beforethe trial result, it shows that the system has the capability of traffic fault diagnosis, andimproves the efficiency of the network optimization, while it was designed flexibly to make iteasier to late extension and clocking with other applications platform.
Keywords/Search Tags:Traffic Fault Diagnosis, case-based reasoning, rule-basedreasoning, KNN, Similarity
PDF Full Text Request
Related items