基于双路射频指纹卷积神经网络与特征融合的雷达辐射源个体识别

肖易寒; 王博煜; 于祥祯; 蒋伊琳

doi:10.11999/JEIT231236

基于双路射频指纹卷积神经网络与特征融合的雷达辐射源个体识别

doi: 10.11999/JEIT231236 cstr: 32379.14.JEIT231236

1.
哈尔滨工程大学先进船舶通信与信息技术工业和信息化部重点实验室哈尔滨 150000
2.
上海无线电设备研究所上海 201100

详细信息

作者简介:
肖易寒：女，博士，副教授，研究方向为信号处理、机器学习等

王博煜：男，硕士生，研究方向为特定辐射源识别

于祥祯：男，博士，副研究员，研究方向为雷达信号处理等

蒋伊琳：男，博士，副教授，研究方向为深度学习、信号处理等

通讯作者:
蒋伊琳　jiangyilin@hrbeu.edu.cn

中图分类号: TN957.51; TN911.3
计量
- 文章访问数: 395
- HTML全文浏览量: 148
- PDF下载量: 48
- 被引次数: 0
出版历程
- 收稿日期: 2023-11-07
- 修回日期: 2024-04-15
- 网络出版日期: 2024-04-25
- 刊出日期: 2024-08-30

Radar Emitter Identification Based on Dual Radio Frequency Fingerprint Convolutional Neural Network and Feature Fusion

1.
Key Laboratory of Advanced Marine Communication and Information Technology, Harbin Engineering University, Harbin 150000, China
2.
Shanghai Radio Equipment Research Institute, Shanghai 201100, China

摘要

摘要: 为实现雷达辐射源个体识别不受信号参数、调制方式的影响，该文提出基于双路射频指纹卷积神经网络(Dual RFF-CNN2)和特征融合的雷达辐射源个体识别方法。首先从接收的射频信号中提取原始I/Q(Raw-I/Q)信号；其次分别对Raw-I/Q两路信号进行轴向积分双谱(AIB)和围线积分双谱(SIB)降维以构建双谱积分矩阵；最后将Raw-I/Q信号及双谱积分矩阵共同送入Dual RFF-CNN2网络并进行特征融合以实现雷达辐射源个体识别。实验结果表明，该方法具有较高的识别准确率，提取的“指纹特征”具备稳定性、鲁棒性。
- 雷达辐射源个体识别 /
- 双路射频指纹卷积神经网络 /
- 特征融合 /
- 指纹特征 /
- 原始I/Q信号
Abstract: In order to achieve identification of radar emitter unaffected by signal parameters and modulation methods, a method based on Dual Radio Frequency Fingerprint Convolutional Neural Network (Dual RFF-CNN2) and feature fusion is proposed in this paper. Firstly, Raw-In-phase/Quadrature (Raw-I/Q) signals are extracted from the received radio frequency signals. Secondly, Axially Integral Bispectrum (AIB) and Square Integral Bispectrum (SIB) dimensionality reduction are performed separately on Raw-I/Q signals to construct the bispectrum integration matrix. Finally, both the Raw-I/Q signals and the bispectrum integration matrix are fed into the Dual RFF-CNN2 network for feature fusion to achieve identification of radar emitter. Experimental results demonstrate that this method achieves high identification accuracy, and the extracted "fingerprint features" exhibit stability and robustness.

HTML全文

图 1 基于Dual RFF-CNN2网络的辐射源个体识别示意图

下载: 全尺寸图片幻灯片

图 2 指纹特征可视化热力图

下载: 全尺寸图片幻灯片

图 3 不同抽取倍率下识别准确率变化曲线

下载: 全尺寸图片幻灯片

图 4 鲁棒性实验结果

下载: 全尺寸图片幻灯片

图 5 损失与识别准确率迭代曲线

下载: 全尺寸图片幻灯片

图 6 5部辐射源发射信号的暂稳态特征图

下载: 全尺寸图片幻灯片

图 7 损失与识别准确率迭代曲线

下载: 全尺寸图片幻灯片

图 8 不同算法识别正确率随信噪比变化曲线对比图

下载: 全尺寸图片幻灯片

表 1 信号参数设置

调制方式	参数	取值
LFM	中心频率(GHz)	0.5, 1.0, 2.0
	带宽(MHz)	10, 20, 30
	脉宽(${\text{μs}}$)	10
	调频斜率	+/–
BPSK	中心频率(GHz)	0.5, 1.0, 2.0
	巴克码序列长度	13
	脉宽(${\text{μs}}$)	10

下载: 导出CSV

表 2 不同算法的性能对比

网络模型	平均识别准确率(%)	模型大小 (kB)	浮点运算量(M)
DualRFF-CNN2	95.8	40.37	10.65
RFF-CNN2+积分矩阵	93.4	20.18	2.41
RFF-CNN2+Raw-I/Q矩阵	90.3	20.18	8.24
IQCNet+Raw-I/Q矩阵	89.5	25.69	12.06
ORACLE+Raw-I/Q矩阵	92.3	133.67	0.58
ResNet1D+稳态特征	87.1	8.86	0.79
RFF-CNN2+原始信号	82.1	20.18	8.24
CNN+1.5维谱	66.5	10.94	1.36
LSTM+暂态特征	78.8	468.61	18.88
DRN+Raw-I/Q	90.9	17.25	2.76

下载: 导出CSV

参考文献(22)

[1]	ZHAO Shiqiang, ZENG Deguo, WANG Wenhai, et al. Mutation grey wolf elite PSO balanced XGBoost for radar emitter individual identification based on measured signals[J]. Measurement, 2020, 159(5): 107777. doi: 10.1016/j.measurement.2020.107777.
[2]	ELDEMERDASH Y A, DOBRE O A, ÜRETEN O, et al. Identification of cellular networks for intelligent radio measurements[J]. IEEE Transactions on Instrumentation and Measurement, 2017, 66(8): 2204–2211. doi: 10.1109/tim.2017.2687539.
[3]	MERCHANT K, REVAY S, STANTCHEV G, et al. Deep learning for RF device fingerprinting in cognitive communication networks[J]. IEEE Journal of Selected Topics in Signal Processing, 2018, 12(1): 160–167. doi: 10.1109/JSTSP.2018.2796446.
[4]	陈蒙, 邢小鹏, 陈世文, 等. 基于贝塞尔曲线的雷达信号脉内无意调相特征提取及个体识别[J]. 信息工程大学学报, 2022, 23(1): 9–17. doi: 10.3969/j.issn.1671-0673.2022.01.002. CHEN Meng, XING Xiaopeng, CHEN Shiwen, et al. Unintentional phase modulation on pulse feature extraction and radar specific emitter identification based on Bezier curve[J]. Journal of Information Engineering University, 2022, 23(1): 9–17. doi: 10.3969/j.issn.1671-0673.2022.01.002.
[5]	秦鑫, 黄洁, 王建涛, 等. 基于无意调相特性的雷达辐射源个体识别[J]. 通信学报, 2020, 41(5): 104–111. doi: 10.11959/j.issn.1000-436x.2020084. QIN Xin, HUANG Jie, WANG Jiantao, et al. Radar emitter identification based on unintentional phase modulation on pulse characteristic[J]. Journal on Communications, 2020, 41(5): 104–111. doi: 10.11959/j.issn.1000-436x.2020084.
[6]	RU Xiaohu, YE Haohuan, LIU Zheng, et al. An experimental study on secondary radar transponder UMOP characteristics[C]. 2016 European Radar Conference, London, UK, 2016: 314–317.
[7]	LUO Zhenyu, CAO Yunhe, YEO T S, et al. Few-Shot radar jamming recognition network via time-frequency self-attention and global knowledge distillation[J]. IEEE Transactions on Geoscience and Remote Sensing, 2023, 61: 5105612. doi: 10.1109/TGRS.2023.3280322.
[8]	李宝平, 魏坡. 基于CWD谱图和改进CNN的无线电调制分类[J]. 电子测量技术, 2023, 46(5): 50–56. doi: 10.19651/j.cnki.emt.2210805. LI Baoping and WEI Po. Radio modulation classification based on CWD spectrogram and improved CNN[J]. Electronic Measurement Technology, 2023, 46(5): 50–56. doi: 10.19651/j.cnki.emt.2210805.
[9]	ZHEN Pan, ZHANG Bangning, CHEN Zhibo, et al. Spectrum sensing method based on wavelet transform and residual network[J]. IEEE Wireless Communications Letters, 2022, 11(12): 2517–2521. doi: 10.1109/LWC.2022.3207296.
[10]	LI Jianfeng, HUANG Dingkun, YAN Xiaopeng, et al. Low SNR FM signal preprocessing method based on low-order cyclic statistics and WVD distribution[C]. The 2nd International Conference on Robotics, Artificial Intelligence and Intelligent Control, Mianyang, China, 2023: 258–262. doi: 10.1109/RAIIC59453.2023.10280797.
[11]	SATIJA U, TRIVEDI N, BISWAL G, et al. Specific emitter identification based on variational mode decomposition and spectral features in single hop and relaying scenarios[J]. IEEE Transactions on Information Forensics and Security, 2019, 14(3): 581–591. doi: 10.1109/TIFS.2018.2855665.
[12]	肖易寒, 李栋年, 于祥祯, 等. 基于参数优化VMD和LightGBM的雷达辐射源个体识别[J]. 航空兵器, 2022, 29(2): 93–100. doi: 10.12132/ISSN.1673-5048.2021.0073. XIAO Yihan, LI Dongnian, YU Xiangzhen, et al. Radar emitter individual identification based on parameter optimization VMD and LightGBM[J]. Aero Weaponry, 2022, 29(2): 93–100. doi: 10.12132/ISSN.1673-5048.2021.0073.
[13]	陈翔, 汪连栋, 许雄, 等. 基于Raw I/Q和深度学习的射频指纹识别方法综述[J]. 雷达学报, 2023, 12(1): 214–234. doi: 10.12000/JR22140. CHEN Xiang, WANG Liandong, XU Xiong, et al. A review of radio frequency fingerprinting methods based on Raw I/Q and deep learning[J]. Journal of Radars, 2023, 12(1): 214–234. doi: 10.12000/JR22140.
[14]	TU Ya, LIN Yun, ZHA Haoran, et al. Large-scale real-world radio signal recognition with deep learning[J]. Chinese Journal of Aeronautics, 2022, 35(9): 35–48. doi: 10.1016/j.cja.2021.08.016.
[15]	SANKHE K, BELGIOVINE M, ZHOU Fan, et al. ORACLE: Optimized radio clAssification through convolutional neuraL nEtworks[C]. The IEEE Conference on Computer Communications, Pairs, France, 2019: 370–378. doi: 10.1109/INFOCOM.2019.8737463.
[16]	YU Jiabao, HU Aiqun, LI Guyue, et al. A robust RF fingerprinting approach using multisampling convolutional neural network[J]. IEEE Internet of Things Journal, 2019, 6(4): 6786–6799. doi: 10.1109/JIOT.2019.2911347.
[17]	WAN Tao, JI Hao, XIONG Wanan, et al. Deep learning-based specific emitter identification using integral bispectrum and the slice of ambiguity function[J]. Signal, Image and Video Processing, 2022, 16(7): 2009–2017. doi: 10.1007/s11760-022-02162-x.
[18]	ELMAGHBUB A and HAMDAOUI B. Leveraging hardware-impaired out-of-band information through deep neural networks for robust wireless device classification[J]. arXiv preprint arXiv: 2004.11126, 2020.
[19]	王亮, 肖易寒. Transformer网络在雷达辐射源识别中的应用[J]. 应用科技, 2021, 48(5): 81–85,104. doi: 10.11991/yykj.202101008. WANG Liang and XIAO Yihan. Application of transformer network in radar emitter recognition[J]. Applied Science and Technology, 2021, 48(5): 81–85,104. doi: 10.11991/yykj.202101008.
[20]	崔天舒, 赵文杰, 黄永辉, 等. 基于射频指纹的测控地面站身份识别方法[J]. 航天电子对抗, 2021, 37(3): 6–9,23. doi: 10.16328/j.htdz8511.2021.03.002. CUI Tianshu, ZHAO Wenjie, HUANG Yonghui, et al. Radio frequency fingerprint-based TT&C ground station identification method[J]. Aerospace Electronic Warfare, 2021, 37(3): 6–9,23. doi: 10.16328/j.htdz8511.2021.03.002.
[21]	SUN Liting, WANG Xiang, YANG Afeng, et al. Radio frequency fingerprint extraction based on multi-dimension approximate entropy[J]. IEEE Signal Processing Letters, 2020, 27: 471–475. doi: 10.1109/LSP.2020.2978333.
[22]	翁琳天然, 彭进霖, 何元, 等. 基于深度残差网络的ADS-B信号辐射源个体识别[J]. 航空兵器, 2021, 28(4): 24–29. doi: 10.12132/ISSN.1673-5048.2020.0095. WENG Lintianran, PENG Jinlin, HE Yuan, et al. Specific emitter identification of ADS-B signal based on deep residual network[J]. Aero Weaponry, 2021, 28(4): 24–29. doi: 10.12132/ISSN.1673-5048.2020.0095.