A Few-Shots OFDM Target Augmented Identification Method Based on Transfer Learning

doi:10.16183/j.cnki.jsjtu.2022.041

Abstract

Abstract:

Under the few-shots condition caused by non-cooperative scenes, robust extraction of communication emitter features and accurate identification of targets are the difficulties and hotspots of current research. Aimed at the problem of emitter identification under the few-shots condition of orthogonal frequency division multiplexing (OFDM) signals, this paper proposes a non-cooperative target identification method based on phase/time domain flipping data augmentation and source domain instance-based transfer learning. The data set is expanded by different domain flipping data augmentation methods, and the improved residual network is applied to achieve the purpose of promoting the identification rate of the OFDM emitter. Then, transfer learning is introduced to strengthen the generalization ability of the identification model. The experimental results show that the data augmentation method can significantly improve the OFDM emitter identification rate under the few-shots condition. Furthermore, the transfer learning method accelerates the convergence speed, slightly increases the recognition rate, and improves robustness of the model.

Key words: orthogonal frequency division multiplexing (OFDM), few-shots identification, data augmentation, transfer learning, deep learning, target identification

CLC Number:

TN929.5
TP18

TANG Zeyu, ZOU Xiaohu, LI Pengfei, ZHANG Wei, YU Jiaqi, ZHAO Yaodong. A Few-Shots OFDM Target Augmented Identification Method Based on Transfer Learning[J]. Journal of Shanghai Jiao Tong University, 2022, 56(12): 1666-1674.

Figures/Tables 13

Fig.1

Fig.2

Fig.3

Fig.4

Fig.5

Fig.6

Tab.1

Fig.7

Tab.2

Fig.8

Tab.3

Fig.9

Tab.4

References 24

[1]	徐晋凯, 谢钧. 辐射源识别技术发展研究[J]. 信息技术与网络安全, 2021, 40(10): 1-7.
	XU Jinkai, XIE Jun. Research on the development of specific emitter identification[J]. Information Technology and Network Security, 2021, 40(10): 1-7.
[2]	张谦, 王吉, 唐泽宇, 等. 基于深度学习的通信电台个体识别技术[J]. 电子信息对抗技术, 2021, 36(2): 36-40.
	ZHANG Qian, WANG Ji, TANG Zeyu, et al. Individual identification technique of communication transmitters based on deep learning[J]. Electronic Information Warfare Technology, 2021, 36(2): 36-40.
[3]	陈悦, 雷迎科, 李昕, 等. 基于IQ图特征的通信辐射源个体识别[J]. 信号处理, 2021, 37(1): 120-125.
	CHEN Yue, LEI Yingke, LI Xin, et al. Specific emitter identification of communication radiation source based on the characteristics IQ graph features[J]. Journal of Signal Processing, 2021, 37(1): 120-125.
[4]	黄少驰, 朱晓蕾. 一种基于射频指纹的通信个体识别方法[J]. 航天电子对抗, 2018, 34(1): 31-34.
	HUANG Shaochi, ZHU Xiaolei. A communication individual recognition method based on radio frequency fingerprint[J]. Aerospace Electronic Warfare, 2018, 34(1): 31-34.
[5]	O’SHEA T J, ROY T, CLANCY T C. Over-the-air deep learning based radio signal classification[J]. IEEE Journal of Selected Topics in Signal Processing, 2018, 12(1): 168-179. doi: 10.1109/JSTSP.2018.2797022 URL
[6]	李鹏, 阮晓钢, 朱晓庆, 等. 基于深度强化学习的区域化视觉导航方法[J]. 上海交通大学学报, 2021, 55(5): 575-585.
	LI Peng, RUAN Xiaogang, ZHU Xiaoqing, et al. A regionalization vision navigation method based on deep reinforcement learning[J]. Journal of Shanghai Jiao Tong University, 2021, 55(5): 575-585.
[7]	HE K M, ZHANG X Y, REN S Q, et al. Deep residual learning for image recognition[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas, NV, USA: IEEE, 2016: 770-778.
[8]	陈浩, 杨俊安, 刘辉. 基于深度残差适配网络的通信辐射源个体识别[J]. 系统工程与电子技术, 2021, 43(3): 603-609.
	CHEN Hao, YANG Junan, LIU Hui. Communication transmitter individual identification based on deep residual adaptation network[J]. Systems Engineering and Electronics, 2021, 43(3): 603-609.
[9]	PAN Y W, YANG S H, PENG H, et al. Specific emitter identification based on deep residual networks[J]. IEEE Access, 2019, 7: 54425-54434. doi: 10.1109/ACCESS.2019.2913759
[10]	曲凌志, 杨俊安, 刘辉, 等. 一种基于复数残差网络的通信辐射源个体识别方法[J]. 信号处理, 2021, 37(1): 95-103.
	QU Lingzhi, YANG Junan, LIU Hui, et al. A method of personal identification of communication radiation source based on complex-valued residual network[J]. Journal of Signal Processing, 2021, 37(1): 95-103.
[11]	HE T, ZHANG Z, ZHANG H, et al. Bag of tricks for image classification with convolutional neural networks[C]//2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Long Beach, CA, USA: IEEE, 2019: 558-567.
[12]	李辰, 李建勋. 卷积神经网络的正交性特征提取方法及其应用[J]. 上海交通大学学报, 2021, 55(10): 1320-1329.
	LI Chen, LI Jianxun. Orthogonal features extraction method and its application in convolution neural network[J]. Journal of Shanghai Jiao Tong University, 2021, 55(10): 1320-1329.
[13]	WANG Y Q, YAO Q M, KWOK J T, et al. Generalizing from a few examples[J]. ACM Computing Surveys, 2021, 53(3): 1-34.
[14]	方章闻, 张金艺, 李科, 等. 小样本条件下的通信辐射源半监督特征提取[J]. 系统工程与电子技术, 2020, 42(10): 2381-2389.
	FANG Zhangwen, ZHANG Jinyi, LI Ke, et al. Semi-supervised feature extraction of communication emitter under small sample condition[J]. Systems Engineering and Electronics, 2020, 42(10): 2381-2389.
[15]	ZHOU H J, BAI J, WANG Y R, et al. Few-shot electromagnetic signal classification: A data union augmentation method[J/OL]. (2021-09-15)[2022-01-05]. https://doi.org/10.1016/j.cja.2021.07.014. doi: 10.1016/j.cja.2021.07.014
[16]	ZHANG W, JIANG L, DING C Z, et al. Small sample identification for specific emitter based on adversarial embedded networks[C/OL]. (2021-09-30)[2022-01-06]. https://doi.org/10.1007/978-3-030-87358-5_18. doi: 10.1007/978-3-030-87358-5_18
[17]	何新林, 戚宗锋, 李建勋. 基于隐变量后验生成对抗网络的不平衡学习[J]. 上海交通大学学报, 2021, 55(5): 557-565.
	HE Xinlin, QI Zongfeng, LI Jianxun. Unbalanced learning of generative adversarial network based on latent posterior[J]. Journal of Shanghai Jiao Tong University, 2021, 55(5): 557-565.
[18]	LIU Y H, XU H, QI Z S, et al. Specific emitter identification against unreliable features interference based on time-series classification network structure[J]. IEEE Access, 2020, 8: 200194-200208. doi: 10.1109/ACCESS.2020.3035813 URL
[19]	ZHOU H J, JIAO L C, ZHENG S L, et al. Generative adversarial network-based electromagnetic signal classification: A semi-supervised learning framework[J]. China Communications, 2020, 17(10): 157-169.
[20]	RIBANI R, MARENGONI M. A survey of transfer learning for convolutional neural networks[C]//2019 32nd SIBGRAPI Conference on Graphics, Patterns and Images Tutorials. Rio de Janeiro, Brazil: IEEE, 2019: 47-57.
[21]	WEISS K, KHOSHGOFTAAR T M, WANG D D. A survey of transfer learning[J]. Journal of Big Data, 2016, 3(1): 1-40. doi: 10.1186/s40537-015-0036-x URL
[22]	FENG Y T, CHENG Y J, WANG G L, et al. Radar emitter identification under transfer learning and online learning[J]. Information, 2019, 11(1): 15. doi: 10.3390/info11010015 URL
[23]	KUZDEBA S, ROBINSON J, CARMACK J. Transfer learning with radio frequency signals[C]//2021 IEEE 18th Annual Consumer Communications & Networking Conference. Las Vegas, NV, USA: IEEE, 2021: 1-9.
[24]	LIU H, HU P J, LIU Z. Communication specific emitter identification based on transfer learning[J]. Journal of Physics: Conference Series, 2020, 1626(1): 012027. doi: 10.1088/1742-6596/1626/1/012027 URL

数据输入方式	平均识别准确率 ± 标准差
图像	32.23±2.78
实部	51.39±3.25
实部和虚部组合	51.57±3.34
幅值	66.16±3.45
幅值和相位组合	67.28±2.80

卷积核结构	平均识别准确率±标准差
7×7原卷积核	67.28±2.90
1×7调整后卷积核	69.83±1.48

数据增强方法	平均识别准确率±标准差
无数据增强	69.83±1.48
子载波调序	77.30±2.24
添加高斯噪声	77.69±2.38
相位域翻转	78.48±2.14
时域翻转	79.35±1.26

迁移学习作用对比	平均识别准确率±标准差
无迁移学习	79.35±1.26
有迁移学习(源域为5部小米手机)	77.76±2.06
有迁移学习(源域为5部vivo手机)	76.60±1.94