TY - GEN
T1 - Radio LFM Fuze Interference Identification Based on Convolutional Neural Network and Attention Mechanism
AU - Xian, Zhaoxia
AU - Bai, Zhiquan
AU - Yang, Jikai
AU - Zhao, Jinqiu
AU - Wang, Caifeng
AU - Hao, Xinhong
N1 - Publisher Copyright:
© 2023 IEEE.
PY - 2023
Y1 - 2023
N2 - In modern battlefields, the fuze is exposed to a complex electromagnetic environment, and suffers severe interference. To identify the typical interferences and improve the reliability of fuze, this paper designs a radio linear frequency modulation (LFM) fuze interference identification network based on convolutional neural network (CNN) and attention mechanism (AM). Firstly, the received signal at the fuze is transformed from the time domain to the time-frequency domain by the short-time Fourier transform (STFT) to obtain the time-frequency images as the network input. Considering the powerful feature extraction capability of CNN, we take it to obtain the semantic features of the time-frequency images and perform the interference recognition, where AM is further employed to focus on more important channel information and improve the recognition accuracy. Simulations show that the recognition accuracy of the network can be over 98% at a jamming-to-signal ratio (JSR) of -16.0dB with better robustness and low complexity.
AB - In modern battlefields, the fuze is exposed to a complex electromagnetic environment, and suffers severe interference. To identify the typical interferences and improve the reliability of fuze, this paper designs a radio linear frequency modulation (LFM) fuze interference identification network based on convolutional neural network (CNN) and attention mechanism (AM). Firstly, the received signal at the fuze is transformed from the time domain to the time-frequency domain by the short-time Fourier transform (STFT) to obtain the time-frequency images as the network input. Considering the powerful feature extraction capability of CNN, we take it to obtain the semantic features of the time-frequency images and perform the interference recognition, where AM is further employed to focus on more important channel information and improve the recognition accuracy. Simulations show that the recognition accuracy of the network can be over 98% at a jamming-to-signal ratio (JSR) of -16.0dB with better robustness and low complexity.
KW - attention mechanism
KW - convolutional neural network
KW - fuze interference signal identification
KW - time-frequency analysis
UR - http://www.scopus.com/inward/record.url?scp=85186085111&partnerID=8YFLogxK
U2 - 10.1109/ICCT59356.2023.10419606
DO - 10.1109/ICCT59356.2023.10419606
M3 - Conference contribution
AN - SCOPUS:85186085111
T3 - International Conference on Communication Technology Proceedings, ICCT
SP - 1410
EP - 1414
BT - 2023 IEEE 23rd International Conference on Communication Technology
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 23rd IEEE International Conference on Communication Technology, ICCT 2023
Y2 - 20 October 2023 through 22 October 2023
ER -