Research On Electromagnetic Interference Signal Identification Technology Based On Neural Network

In recent years,with the rapid development of radio communication technology,the electromagnetic environment of communication network is becoming more and more complex,and various electromagnetic interference phenomena occur frequently,especially in the field of military aviation and civil aviation,electromagnetic environment monitoring and anti-interference appear to be particularly important.The identification of radio signals is the first problem to be solved in aviation network anti-jamming.In view of the complexity and diversity of the current electromagnetic environment and the emergence of a variety of troublesome interference signals,this thesis proposes a method for identifying the external and internal electromagnetic interference signals of the engine room.The main research contents are as follows:(1)Aiming at different radio interference signals outside the cabin,this thesis proposes an identification algorithm based on Long Short-Term Memory(LSTM)neural network.This algorithm firstly analyzes the characteristics of emi signals,constructs a two-level cascade LSTM neural network according to the timing characteristics of emi signals,and constructs a classification model by using the method of alternating connection between the full connection layer and the LSTM layer.Then,the algorithm model is optimized by using the relevant model optimization strategy,and finally various interference signal types are identified efficiently.The results show that when the SNR is greater than 5dB,the model's average recognition accuracy of various modulation signals is over 95%.(2)As for the electromagnetic radiation signals existing in the cabin,a Convolutional Neural Networks(CNN)based algorithm is proposed to identify the electromagnetic radiation signals that are combined with the Software Defined Radio(SDR).The algorithm firstly through the SDR to collect different mechanical and electrical equipment to work the small power of radiation signals,and then using the correlation noise reduction algorithm data sets of the radiation signal noise reduction processing,in order to solve the electromagnetic radiation signal sample purity is not high,classification,characteristics is not obvious,the final structures,CNN network structure,the use of different parts of the network function layered extraction radiation signal characteristics,and realize the high precision of recognition.The results show that the accuracy of the algorithm is about 92%.