Research on Radar PRI Modulation Pattern Recognition Based on Recurrent Neural Network
Wei Qu, Gang Yao, Meng Lei · 2023
In this paper, the Recurrent Neural Network (RNN) model is introduced into the identification of radar emitters, which verifies the feasibility of RNN application to radar emitter identification, and the design process is simplified and the radar emitter identification method is easy to apply in engineering. Based on the radar emission source recognition model, the Pulse Repetition Interval(PRI) modulation pattern recognition research based on recurrent neural network was carried out. Firstly, aiming at the problems of gradient vanishing and gradient explosion in RNN, a PRI modulation pattern recognition method based on LSTM is proposed, and the recognition method process is introduced in detail. According to the characteristics of pulse train data, a data preprocessing method suitable for recurrent neural network is proposed. Finally, simulation verifies the effectiveness of the proposed method and tests the performance of the algorithm under different interference conditions.