Abstract：In order to solve the problem that the correct recognition rate of Low Probability of Intercept (LPI) radar signal is low and the feature extraction is difficult, an automatic classification and recognition system based on Choi-Williams Distribution (CWD) and stacked Sparse Auto-Encoder (sSAE) is proposed. The system starts from the time-frequency image which reflects the essential characteristics of the signal. Firstly, the CWD is performed on the LPI radar signal to obtain the two-dimensional time-frequency image. Then, the obtained time-frequency original image is preprocessed and the preprocessed image is sent into the multilayer SAE for off-line training. Finally, the feature automatically extracted from the SAE is sent to the softmax classifier, to achieve on-line classification and identification of the radar signal. Simulation results show that the classification system achieves overall correct recognition rate of 96.4% at SNR of for the eight LPI radar signals (LFM, BPSK, Costas, Frank and T1~T4), which is better than the method of manually designing the extract signal characteristics under low SNR conditions.
WANG Xing, ZHOU Yipeng, ZHOU Dongqing, et al. Research on low probability of intercept radar signal recognition using deep belief network and bispectra diagonal slice[J]. Journal of Electronics & Information Technology, 2016, 38(11): 2972-2976. doi: 10.11999/JEIT160031.
[4]
LUNDEN J and KOIVUNEN V. Automatic radar waveform recognition[J]. IEEE Journal of Selected Topics in Signal Processing, 2007, 1(1): 124-136. doi: 10.1109/JSTSP.2007. 897055.
[5]
ZHANG Ming, LIU Lutao, DIAO Ming, et al. LPI radar waveform recognition based on time-frequency distribution[J]. Sensors, 2016, 16(10): 1682-1706. doi: 10.3390/s16101682.
[6]
HINTON G, OSINDERO S, TEH Y W, et al. A fast learning algorithm for deep belief nets[J]. Neural Computation, 2006, 18(7): 1527-1554. doi: 10.1162/neco.2006.18.7.1527.
[7]
BENGIO Y, LAMBIN P, POPOVICI D, et al. Greedy layer-wise training of deep networks[C]. Advances in Neural Information Processing Systems, Hyatt Regency Vancouver, 2007: 153-160.
[8]
LECUN Y, BENGIO Y, HINTON G, et al. Deep learning[J]. Nature, 2015, 521(7553): 436-444. doi: 10.1038/nature14539.
[9]
NG A. Sparse autoencoder[J]. CS294A Lecture Notes, 2011, 72(2011): 1-19. doi: 10.1371/journal.pone.0006098.
[10]
TAO Chao, PAN Hongbo, LI Yansheng, et al. Unsupervised spectral-spatial feature learning with stacked sparse autoencoder for hyperspectral imagery classification[J]. IEEE Geoscience and Remote Sensing Letters, 2015, 12(12): 2438-2442. doi: 10.1109/LGRS.2015.2482520.
[11]
ZHANG Lu, MA Wenping, ZHANG Dan, et al. Stacked sparse autoencoder in PolSAR data classification using local spatial information[J]. IEEE Geoscience and Remote Sensing Letters, 2016, 13(9): 1359-1363. doi: 10.1109/LGRS.2016. 2586109.
[12]
SUN Wenjun, SHAO Siyu, ZHAO Rui, et al. A sparse auto- encoder-based deep neural network approach for induction motor faults classification[J]. Measurement, 2016, 89: 171-178. doi: 10.1016/j.measurement.2016.04007.
[13]
FENG Zhipeng, LIANG Ming, CHU Fulei, et al. Recent advances in time-frequency analysis methods for machinery fault diagnosis: A review with application examples[J]. Mechanical Systems and Signal Processing, 2013, 38(1): 165-205. doi: 10.1016/j.ymssp.2013.01017.
