Intra-pulse modulation radar signal recognition based on CLDN network

• Published in: IET radar, sonar & navigation Vol. 14; no. 6; pp. 803 - 810
• Main Authors: Wei, Shunjun, Qu, Qizhe, Su, Hao, Wang, Mou, Shi, Jun, Hao, Xiaojun
• Format: Journal Article
• Language: English
• Published: The Institution of Engineering and Technology 01.06.2020
• Subjects: autocorrelation domain; automatic modulation classification; civilian applications; CLDN network; convolutional neural nets; deep learning network; deep neural network; feature extraction; frequency domain; intra‐pulse modulation radar signal recognition; learning (artificial intelligence); LSTM network; military applications; modulation; pulse modulation; radar signal processing; radar signals; raw signal sequences; recognition methods; recognition performances; recurrent neural nets; Research Article; shallow CNN; shallow convolution neural network; short‐term memory network; signal classification; signal modulation types; signal‐to‐noise ratio; time domain; raw signal sequences; time domain; modulation; signal modulation types; CLDN network; pulse modulation; frequency domain; radar signal processing; shallow CNN; intra-pulse modulation radar signal recognition; autocorrelation domain; feature extraction; LSTM network; radar signals; shallow convolution neural network; learning (artificial intelligence); recognition methods; deep neural network; short-term memory network; recurrent neural nets; military applications; automatic modulation classification; signal classification; civilian applications; signal-to-noise ratio; convolutional neural nets; recognition performances; deep learning network
• ISSN: 1751-8784; 1751-8792
• DOI: 10.1049/iet-rsn.2019.0436

Automatic modulation classification of radar signals, which plays a significant role in both civilian and military applications, is researched in this study through a deep learning network. In this study, a novel network combined a shallow convolution neural network (CNN), long short-term memory (LSTM) network and deep neural network (DNN) is proposed to recognise six types of radar signals with different signal-to-noise ratio (SNR) levels from −14 to 20 dB. First, raw signal sequences in the time domain, frequency domain and autocorrelation domain are as input for a shallow CNN. Then the features extracted by CNN will be the input of LSTM network. Finally, DNNs will output the signal modulation types directly. The simulation results demonstrate that the accuracies in autocorrelation domain are all more than 90% at −6 dB and close to 100% when SNR > −2 dB. The recognition performances of the three domains are compared. Compared with other recognition methods, the proposed method has higher average accuracy and better performance under low SNR condition. The measured results show that the proposed method has achieved high accuracies of common four kinds of measured radar signals.