Radar Emitter Identification Based on Improved Convolutional Neural Network
Kun Li · 2019
Aiming at the problem that the traditional pulse description word is difficult to effectively identify complex radar emitters, a method of extracting time-frequency characteristics of radar emitters and using improved convolutional neural network for recognition is proposed. Firstly, two-dimensional time-frequency images are obtained by time-frequency analysis of radar signals. Then, the convolution layer of convolutional neural network is used to extract time-frequency features, the full connection layer is used to integrate features, and softmax is used for classification and recognition. Five kinds of common radar signals are identified by simulation experiments, which proves the validity of this method.