Font Size: a A A

Integration Based On Fuzzy Neural Network And Two-tuple Linguistic Communication Station Evaluation Method

Posted on:2019-11-14Degree:MasterType:Thesis
Country:ChinaCandidate:Z Z ZhangFull Text:PDF
GTID:2416330572957950Subject:Electronic and communication engineering
Abstract/Summary:PDF Full Text Request
Wireless communication station is the basic unit of military communication,and its effectiveness directly determines the degree of information support in combat operations.At present,there are many researches on wireless communication performance,but there are some problems.Some evaluation methods are simple linear superposition of influencing factors,some are too dependent on expert experience,subjective influence is greater,and some are inaccurate transformation between qualitative and quantitative problems.Based on the practical application of the unit,this paper presents a communication station capacity evaluation system based on fuzzy neural network.Fuzzy neural network is the combination of fuzzy control theory and artificial neural network,which makes the system have good robustness.The system designed in this paper includes preprocessing module,self-learning module and evaluation module.Data pre-processing module is to clean and organize the data,distinguish subjective factors and objective factors,and input data into different modules.The objective factors include terrain,weather,electromagnetic environment,power,etc.The data are normalized and input into the self-learning module.The self-learning module determines whether self-learning needs to be triggered or not.If not,input the evaluation module according to the results of fuzzy neural network.Subjective factors,including the quality of personnel,equipment,and so on,because of the subjectivity and randomness of the evaluation,the value is converted into binary semantics when it passes through the pre-processing module.The evaluation module is integrated with the evaluation results of objective factors judged by the self-learning module by binary semantics weighted integration as the final evaluation results.The system is based on the practical application of the unit designed,in the consideration of indicators as far as possible to achieve the unity of efficiency and accuracy,the choice of learning samples is based on the actual data accumulated by the establishment of telecommunications stations in previous years.Through the simulation test,basically consistent with the test sample,although there are some discrete points,but through the actual analysis,the error is within the acceptable range,the evaluation of communication stations has a strong guiding significance.
Keywords/Search Tags:Communication station, neural network, Two-tuple linguistic, Communication capability evaluation
PDF Full Text Request
Related items