Research on signal modulation identification method based on residual neural network
DOI:
https://doi.org/10.54097/00qdkms1Keywords:
Deep learning, Modulation recognition, Residual network, LightweightAbstract
Aiming at the problems of large number of parameters, high complexity and low recognition accuracy of the current modulation recognition model based on high-performance neural network, a lightweight signal modulation recognition algorithm based on residual neural network (ResNet) is proposed. Firstly, the ResNet network model is built, which contains two residual units and three fully connected layers. Secondly, the pruning method of network slimming is introduced to compress the neural network model without affecting the recognition accuracy. Finally, the model was trained using modulated signal samples for the RaidoML2016.10a international standard dataset. The experimental results show that the recognition accuracy of the pruning ResNet modulation recognition algorithm can reach 95% under the condition of 12dB signal-to-noise ratio, and the number of model parameters is only 5.7×104.
References
J.W. ZHANG. Research on digital signal modulation recognition technique based on deep learning [D]. University of Electronic Science and Technology, 2022.
Z. YUAN. Research on image enhancement and edge detection technology based on deep convolutional network[D]. Zhejiang University,2020.O'Shea Timothy J, Corgan Johnathan, Clancy T Charles. Convolutional Radio Modulation Recognition Networks[J]. ENGINEERING APPLICATIONS OF NEURAL NETWORKS, 2016, 629: 213-226.
Timothy James O’Shea, Tamoghna Roy, T. Charles Clancy. Over-the-Air Deep Learning Based Radio Signal Classification[J]. Selected Topics in Signal Processing, IEEE Journal of, 2018, 12(1): 168-179.
HU SHISHENG, PEI YIYANG, LIANG P P, et al. Deep neural network for robust modulation classification underrun certain noise conditions[J]. IEEE Transactions on Ve-hicular Technology, 2020, 69(1):564-577.
ZHANG SICHENG, LIN YUN, KANG JIAN, et al. Modulation signal recognition model based on lightweight deep learning network[J]. Journal of Terahertz. Science and Electronic Information 'Technology,2021, 19(1): 54-59. (in Chinese).
Kumar Satish, Mahapatra Rajarshi, Singh Anurag. Automatic Modulation Recognition: An FPGA Implementation[J]. IEEE COMMUNICATIONS LETTERS, 2022, 26(9):2062-2066.
Wei X. Research on automatic modulation classification technology based on neural network architecture search [D]. Jiangsu: Nanjing University of Posts and Telecommunications, 2022.
Zhang M. Efficient neural network design for radar signal recognition [D]. Nanjing University of Posts and Telecommunications, 2022.
LI SING-BAI,YANG Jie, HUANG ZHITAO, WANG Xiang. Deep learning based modulation recognition algorithm for communication signals[J]. Space Electronics Technology, 2019, 16(01):49-54+74.
Ma WANLI. Research on signal modulation mode recognition based on deep learning [D]. Beijing: Academy of Military Science,2020.
Wang XUEYAN. Similar flower identification and three-dimensional reconstruction under deep learning [D]. Beijing: Beijing Forestry University,2021.
Kaiming He, Xiang yu Zhang, Shaoqing Ren, Jian Sun. Deep Residual Learning for Image Recognition[J]. Computer Science, 2015, 1508-2806.
Z. Liu. Research on modulation mode recognition algorithm based on deep learning [D]. Hubei: Central South University for Nationalities,2021.
Szegedy C,Vanhoucke V, Ioffe S, et al. Rethinking the Inception Architecture for Computer Vision[J]. CoRR,2015, 1512.00567.
Liu Zhuang, Li Jianguo, Shen Zhiqiang. Learning Efficient Convolutional Networks through Network Slimming[A]. 2017 IEEE International Conference on Computer Vision (ICCV) [C], 2017.
SHAO MINLAN, ZHOU HONGJIAN, ZHANG Haoran. Modulated signal recognition algorithm based on convolutional neural network [J]. Laboratory Research and Exploration, 2021, 40(08):28-31+41.
Downloads
Published
Issue
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
How to Cite
