Research And Design Of Wireless Communications Equipment Fault Diagnosis Expert System

The fault diagnosis for wireless communication equipments is of great significance to the normal operation of communication system,the social production and life,as well as the national security.Today,with the in-depth development of the theory and technology of fault diagnosis,the artificial intelligence methods are being used in diagnosis because of its advantages.Expert system,neural network methods,and the neural network expert system,which have aroused widespread attention, are being improved gradually in research and practice.In this paper,based on the development of the integrated test system for wireless communication equipments,the fault diagnosis is focused and discussed.After presentation of the basic principles of fault diagnosis,the technology about expert system and neural network is studied Respectively.Especially BP neural network,as well as its designing methods,are discussed and analysed.According to the theoretical study of expert system and the fault characteristics of wireless communication equipments, a rule-fuzzy-based fault diagnosis expert system is implemented,which mainly covers:fuzzy resoning,the system architecture,the design of database and knowledge base,the reasoning machine,and Knowledge Base Maintenance and Management.And as the core parts in the design,the knowledge base and the reasoning machine are detailed in the text.Although some successful systems have been developed based on the rule and fuzzy resoning,there are inborn shortages in knowledge acquisition and association and so on which bring limitation to the development of fault diagnosis.While neural network can resolve these problems.Therefore,the method which combines traditional expert system with neural network becomes one of the main research directions of intelligent fault diagnosis.After the implement of the rule-based expert system, there is a further study of the neural network expert system.Through comparing the different points between expert system and neural network,how to combine the two and what characteristics of the combined system will be,are discussed and studied.According to practical needs,a structure of the neural network expert system for wireless communication equipment fault diagnosis is proposed,and its functions and implementing methods are presented as well.In the combined system,the knowledge base,reasoning,and expositing are still the most crucial parts.Finally,based on a transmitter and its fault diagnosis,a BP network is designed and applied.And then the network is simulated in the Matlab neural network toolbox.And the results shows that the neural network can identify the fault mode effectively.After that,some viewpoints about how to set the parameters for BP network in the follow-up development,such as the hidden nodes and the learning functions,are presented.