Low Probability of Intercept Radar Signal Recognition by Staked Autoencoder and SVM
Muqing Zhang, Huali Wang, Kaijie Zhou, Peipei Cao · 2018
A novel low probability of intercept (LPI) radar signal recognition method based on stacked autoencoder combined with support vector machine (SVM) is proposed in this paper. The method firstly transforms the LPI radar signal to time-frequency (T-F) domain through Choi-Williams Distribution (CWD) to obtain the T-F images of signals. Then, a series of preprocessing methods are used to suppress the image noise and resize the images. Finally, the stacked autoencoder is used to extract features automatically, which are sent into SVM to complete signal recognition. Simulation results demonstrate that the method performs well in a low signal to noise ratio (SNR) condition and is suitable for the case of small sample.