Aiming at the problem of radar radiation source signal recognition,an algorithm based on ConvNeXt model is proposed.Firstly,because the signals with different modulation modes have differ-ent characteristics in the time-frequency domain,the time-frequency conversion results of radar radia-tion source signal are regarded as images,and computer vision technology is used to recognize them.Secondly,the time-frequency images of radar signals of different modulation types are obtained by the Choi-Williams distribution(CWD)transformation,and the images are preprocessed.Thirdly,the time-frequency features are extracted and the radar signals are identified using ConvNeXt model,which solves the problem of low recognition accuracy under the condition of low SNR and limited samples.The experimental results show that ConvNext model has a stronger feature learning ability and im-proves the overall recognition rate of 16 kinds of signals effectively.The recognition accuracy of 6 types of signals with similar time-frequency characteristics(Frank,LFM,P1,P2,P3,P4)is higher.In addition,the algorithm is robust to small samples.
维普
万方数据
