Research on Automatic Classification of Broadcast Signals Based on Deep Learning
Illegal broadcast signals refer to radio stations established without the approval of the national radio management department.Such signals can disrupt the normal radio communication order.Therefore,this paper proposes an illegal broadcast signal monitoring method based on time-frequency domain feature extraction and Back Propagation(BP)neural network.This method first uses the time-frequency domain analysis method to extract the characteristics of the signal,and then uses the BP model for classification.In the experiment,this article used MATLAB to generate a simulated broadcast signal data set and tested the method.The results show that the accuracy,recall rate and F1 value of this method all reach more than 99％,which proves the correctness of this method.
deep learningbroadcast signaltime-frequency domain feature extractionBack Propagation(BP)neural network
万方数据
维普
