Design And Implementation Of Automatic Test And Fault Diagnosis System For RF Indicators

With the development of test technology and artificial intelligence,automatic test technology is gradually getting rid of the professional customization to the modular and universal development.At the same time,diagnosis technology has also entered the stage of intelligent diagnosis technology,which is based on knowledge processing as the core,signal processing,modeling processing and knowledge processing.The joint research of automatic test and fault diagnosis has been one of the research focuses.In the wireless communication equipment research and development stage,the semi-automatic test program is used to test the radio frequency index,and the data analysis and test problem positioning are all completed by the test experts.The efficiency of the above work can not fully meet the needs of the current research tasksFor the problem of testing and fault location efficiency,the thesis designs and implements an automatic testing and intelligent fault diagnosis system for radio frequency index.First of all?the ZigBee and Sub-GHz radio frequency test requirements are analyzed in detail.An automatic test scheme based on virtual instrument technology is designed,and the Impeller environment of HUAWEI is used as a test program running management platform.Automatic test case design,the use of test suit in the form of test equipment control,command input,document read and write and other common content management,the system has good compatibility and portability.Then,according to the different test methods of 38 test items,such as peak power,spectrum template and receiving sensitivity,the realization process of the test program is described in detail,and the corresponding program flow chart are proposed.Next,based on the expert system and BP neural network,the application of expert system and BP neural network in the intelligent location of RF test has been discussed in the system fault diagnosis part.This thesis presents the basic architecture of an intelligent fault diagnosis system based on expert system and BP neural network as a supplement.The design of knowledge base,using knowledge representation method combined with object-oriented technology and production rules,and using MySQL database to build and maintain knowledge base.The reasoning mechanism uses the expert system based on knowledge as the core of the diagnosis reasoning.When the fault information can not be matched with the explicit knowledge base,the system will call the neural network based on the numerical calculation to reason.The interpretation mechanism is designed by using the preset text method,which can explain the diagnosis result and the reasoning process at the same time of the diagnosis result output.The system uses the WEB-based interactive interface,in order to ensure the system performance and stable operation under the premise of the realization of a more friendly way of human-computer interaction.The system has been put into operation in Huawei’s Network Hardware Lab(Nanjing),which significantly improves the efficiency of RF test.It also realizes the intelligent positioning of test problems and meets the design performance requirements.