基于改进卷积神经网络和射频指纹的无人机检测与识别

doi:10.11772/j.issn.1001-9081.2023030299

《计算机应用》唯一官方网站 ›› 2024, Vol. 44 ›› Issue (3): 876-882.DOI: 10.11772/j.issn.1001-9081.2023030299

基于改进卷积神经网络和射频指纹的无人机检测与识别

周景贤¹, 李希娜²()

^1.中国民航大学信息安全测评中心天津 300300
^2.中国民航大学计算机科学与技术学院，天津 300300

收稿日期:2023-03-23 修回日期:2023-05-30 接受日期:2023-06-02 发布日期:2023-06-20 出版日期:2024-03-10
通讯作者: 李希娜
作者简介:周景贤（1981—），男，河南信阳人，副研究员，博士，主要研究方向：安全认证协议、数据隐私保护、物联网安全架构；
基金资助:
中央高校基础研究基金资助项目(3122018C036);民航安全能力项目(PESA219074)

UAV detection and recognition based on improved convolutional neural network and radio frequency fingerprint

Jingxian ZHOU¹, Xina LI²()

^1.Information Security Evaluation Center，Civil Aviation University of China，Tianjin 300300，China
^2.School of Computer Science and Technology，Civil Aviation University of China，Tianjin 300300，China

Received:2023-03-23 Revised:2023-05-30 Accepted:2023-06-02 Online:2023-06-20 Published:2024-03-10
Contact: Xina LI
About author:ZHOU Jingxian， born in 1981， Ph. D.， associate research fellow. His research interests include security authentication protocol， data privacy protection， security architecture for internet of things.
Supported by:
Fundamental Research Funds for Central Universities(3122018C036);Project of Civil Aviation Safety Capacity(PESA219074)

摘要/Abstract

摘要：

针对无人机（UAV）在图像识别时易受环境干扰，而传统信号识别难以准确提取特征且实时性较差的问题，提出一种基于改进卷积神经网络（CNN）和射频（RF）指纹的无人机检测识别方法。首先，使用通用软件无线电外设（USRP）捕获环境中的无线电信号，经过多分辨率分析获取偏差值，检测是否为无人机射频信号；其次，将检测到的无人机射频信号经过小波变换和主成分分析（PCA）处理，获得射频信号频谱，作为神经网络的输入；最后，构建轻量级残差神经网络（LRCNN），输入射频频谱进行网络训练，进行无人机的分类识别。实验结果表明，所提方法可以有效检测并识别无人机信号，平均识别精度可达84%；在信噪比（SNR）大于20 dB时，LRCNN的识别精度达到了88%，相较于支持向量机（SVM）、原始OracleCNN分别提高31和7个百分点，在识别精度和鲁棒性方面比这两种方法均有所提升。

关键词: 无人机安全, 射频指纹, 小波变换, 注意力残差网络, 卷积神经网络

Abstract:

In order to solve the problems that the UAV （Unmanned Aerial Vehicle） is vulnerable to environmental interference in image recognition， and the traditional signal recognition is difficult to accurately extract features and has poor real-time performance， a UAV detection and recognition method based on improved CNN （Convolutional Neural Network） and RF （Radio Frequency） fingerprint was proposed. Firstly， a USRP （Universal Software Radio Peripheral） was used for capturing radio signals in an environment， a deviation value was obtained through multi-resolution analysis， to detect whether the radio signal was an unmanned aerial vehicle radio frequency signal or not. Secondly， the detected unmanned aerial vehicle radio frequency signal was subjected to wavelet transformation and PCA （Principal Component Analysis） to obtain a radio frequency signal spectrum which was used as an input of a neural network. Finally， a LRCNN （Lightweight Residual Convolutional Neural Network） was constructed， and the RF spectrum was input to train the network for UAV classification and recognition. Experimental results show that LRCNN can effectively detect and recognize UAV signals， and the average recognition accuracy reaches 84%. When the SNR （Signal-to-Noise Ratio） is greater than 20 dB， the recognition accuracy of LRCNN reaches 88%， which is 31 and 7 percentage points higher than those of SVM （Support Vector Machine） and the original OracleCNN， respectively. Compared with these two methods， LRCNN has improved recognition accuracy and robustness.

Key words: UAV (Unmanned Aerial Vehicle) security, radio frequency fingerprint, wavelet transform, attention residual network, Convolutional Neural Network (CNN)

中图分类号:

TP391

周景贤, 李希娜. 基于改进卷积神经网络和射频指纹的无人机检测与识别[J]. 计算机应用, 2024, 44(3): 876-882.

Jingxian ZHOU, Xina LI. UAV detection and recognition based on improved convolutional neural network and radio frequency fingerprint[J]. Journal of Computer Applications, 2024, 44(3): 876-882.

图/表 14

图1 无人机RF信号识别流程

Fig. 1 Recognition flow of UAV RF signal

图2 无人机射频指纹检测识别算法框架

Fig. 2 Algorithm framework for UAV radio frequency fingerprint detection and recognition

图3 二维小波变换过程

Fig. 3 Process of two-dimensional wavelet transform

图4 二维小波变换及主成分分析处理后信号

Fig. 4 Signals before and after two-dimensional wavelet transform and PCA

图5 残差块的结构

Fig. 5 Structure of ResBlock

图6 LRCNN结构

Fig. 6 Structure of LRCNN

表1 LRCNN参数

Tab. 1 Parameters of LRCNN

参数	值
卷积层1	卷积核 $50 × 2 × 7$
ReLU	Max（0，x）
残差块卷积层1	卷积核 $50 × 2 × 7$
残差块卷积层2	卷积核 $50 × 2 × 7$
全局平均池化层	$2 × 2$
全连接层1（FC1）	神经元80
全连接层2（FC2）	神经元4
dropout率	50%
I2正则化参数	λ=0.000 1

表1 LRCNN参数

Tab. 1 Parameters of LRCNN

参数	值
卷积层1	卷积核 $50 × 2 × 7$
ReLU	Max（0，x）
残差块卷积层1	卷积核 $50 × 2 × 7$
残差块卷积层2	卷积核 $50 × 2 × 7$
全局平均池化层	$2 × 2$
全连接层1（FC1）	神经元80
全连接层2（FC2）	神经元4
dropout率	50%
I2正则化参数	λ=0.000 1

表2 Adam随机梯度下降算法参数

Tab. 2 Parameters of Adam stochastic gradient descent algorithm

名称	含义	取值
Learning_rate	学习速率、学习步长	0.000 1
Beta1	一阶矩估计的指数衰减因子	0.990 0
Beta2	二阶矩估计的指数衰减因子	0.990 0
Epsilon	避免除0参数（≥0）	0.001 0
Use_locking	为true时，更新操作使用锁	—

表3 不同batch_size对应的训练精度与时间

Tab. 3 Training time and accuracy with different batch_size

batch_size	训练时间/min	训练精度/%
2	35	80.6
4	20	84.2
8	12	67.8

图7 识别精度随训练次数变化

Fig. 7 Recognition accuracy varying with number of training times

图8 LRCNN消融实验结果

Fig. 8 Ablation experiment results of LRCNN

表4 无人机数量对识别精度的影响

Tab. 4 Influence of number of UAVs on recognition accuracy

无人机数	不同算法的识别精度/%
无人机数	LRCNN	OracleCNN	SVM
2	88.2	84.1	63.3
4	86.2	81.3	60.6
8	85.2	78.7	57.3
16	83.9	75.9	52.4

图9 训练次数对识别精度的影响

Fig. 9 Influence of training times on recognition accuracy

图10 SNR对识别精度的影响

Fig. 10 Influence of SNR on recognition accuracy

参考文献 22

1	WANG R， LI Z， TANG J， et al. RF fingerprint identification of commercial UAV in outdoor environment ［C］// Proceedings of the 2022 International Conference on Computing， Communication， Perception and Quantum Technology. Piscataway： IEEE， 2022： 367-371. 10.1109/ccpqt56151.2022.00070
2	XU X， ZENG Y， GUAN Y L， et al. Overcoming endurance issue： UAV-enabled communications with proactive caching ［J］. IEEE Journal on Selected Areas in Communications， 2018， 36（6）： 1231-1244. 10.1109/jsac.2018.2844979
3	李超群，王金明.基于短时傅里叶变换的无人机射频指纹分类识别［J］.通信技术，2022，55（9）：1202-1207. 10.3969/j.issn.1002-0802.2022.09.014
	LI C Q， WANG J M. Classification and identification of UAV RF fingerprints based on short time Fourier transform ［J］. Communication Technology， 2022， 55（9）： 1202-1207. 10.3969/j.issn.1002-0802.2022.09.014
4	J-P HUTTNER， FRIEDRICH M. Current challenges in mission planning systems for UAVs： a systematic review ［C］// Proceedings of the 2023 Integrated Communication， Navigation and Surveillance Conference. Piscataway： IEEE， 2023： 1-7. 10.1109/icns58246.2023.10124299
5	SU Y， ZHOU J， GUO Z. A trust-based security scheme for 5G UAV communication systems ［C］// Proceedings of the 2020 IEEE International Conference on Dependable， Autonomic and Secure Computing， International Conference on Pervasive Intelligence and Computing， International Conference on Cloud and Big Data Computing， International Conference Conference on Cyber Science and Technology Congress. Piscataway： IEEE， 2020： 371-374. 10.1109/dasc-picom-cbdcom-cyberscitech49142.2020.00072
6	PIRKER D， FISCHER T， LESJAK C， et al. Global and secured UAV authentication system based on hardware-security ［C］// Proceedings of the 8th IEEE International Conference on Mobile Cloud Computing， Services， and Engineering. Piscataway： IEEE， 2020： 84-89. 10.1109/mobilecloud48802.2020.00020
7	URETEN O， SERINKEN N. Wireless security through RF fingerprinting ［J］. Canadian Journal of Electrical and Computer Engineering， 2007， 32（1）： 27-33. 10.1109/cjece.2007.364330
8	SANKHE K， BELGIOVINE M， ZHOU F， et al. ORACLE： optimized radio classification through convolutional neural networks［C］// Proceedings of the 2019 IEEE Conference on Computer Communications. Piscataway： IEEE， 2019： 370-378. 10.1109/infocom.2019.8737463
9	SONG L， GAO Z， HUANG J， et al. A lightweight radio frequency fingerprint extraction scheme for device identification ［C］// Proceedings of the 2023 IEEE Wireless Communications and Networking Conference. Piscataway： IEEE， 2023： 1-5. 10.1109/wcnc55385.2023.10118789
10	YANG Y， YAN T. RF fingerprint recognition method based on DBN-SVM ［C］// Proceedings of the IEEE 10th International Conference on Information， Communication and Networks. Piscataway： IEEE， 2022： 572-577. 10.1109/icicn56848.2022.10006500
11	LI Y， CHEN L， CHEN J， et al. A low complexity feature extraction for the RF fingerprinting process ［C］// Proceedings of the 2018 IEEE Conference on Communications and Network Security. Piscataway： IEEE， 2018： 1-2. 10.1109/cns.2018.8433156
12	KENNEDY I O， SCANLON P， BUDDHIKOT M M. Passive steady state RF fingerprinting： a cognitive technique for scalable deployment of co-channel femto cell underlays ［C］// Proceedings of the 3rd IEEE Symposium on New Frontiers in Dynamic Spectrum Access Networks. Piscataway： IEEE， 2008： 1-12. 10.1109/dyspan.2008.46
13	O’SHEA J， CORGAN J， CLANCY T C. Unsupervised representation learning of structured radio communication signals ［C］// Proceedings of the 1st International Workshop on Sensing， Processing and Learning for Intelligent Machines. Piscataway： IEEE， 2016： 1-5. 10.1109/splim.2016.7528397
14	FENG J， TANG X， ZHANG B， et al. A lightweight deep learning RF fingerprint recognition method ［C］// Proceedings of the 4th International Conference on Communications， Information System and Computer Engineering. Piscataway： IEEE， 2022： 452-457. 10.1109/cisce55963.2022.9851177
15	LI C， WANG J， WANG W， et al. RF-based on feature fusion and convolutional neural network classification of UAVs ［C］// Proceedings of the IEEE 8th International Conference on Computer and Communications. Piscataway： IEEE， 2022： 1899-1904. 10.1109/iccc56324.2022.10065895
16	XU J， ZHAO H. Detection to primary user based on radio frequency fingerprint in cognitive radio ［C］// Proceedings of the 2nd International Congress on Image and Signal Processing. Piscataway： IEEE， 2009： 1-5. 10.1109/cisp.2009.5302025
17	NGUYEN P， TRUONG H， RAVINDRANATHAN M， et al. Matthan： drone presence detection by identifying physical signatures in the drone’s RF communication ［C］// Proceedings of the 15th Annual International Conference on Mobile Systems， Applications， and Services. New York： ACM， 2017： 211-224. 10.1145/3081333.3081354
18	EZUMA M， ERDEN F， KUMAR ANJINAPPA C， et al. Detection and classification of UAVs using RF fingerprints in the presence of Wi-Fi and Bluetooth interference ［J］. IEEE Open Journal of the Communications Society， 2022， 1： 60-76.
19	NIE W， HAN Z-C， ZHOU M， et al. UAV detection and identification based on WiFi signal and RF fingerprint ［J］. IEEE Sensors Journal， 2021， 21（12）： 13540-13550. 10.1109/jsen.2021.3068444
20	李雨珊，谢非佚，陈松林，等.适合终端的射频指纹信号特征提取及识别［J］.通信技术，2018，51（1）：63-66. 10.3969/j.issn.1002-0802.2018.01.012
	LI Y S， XIE F Y， CHEN S L， et al. Feature extraction and recognition of radio frequency fingerprint signal suitable for terminal ［J］. Communication Technology， 2018， 51（1）： 63-66. 10.3969/j.issn.1002-0802.2018.01.012
21	HE K， ZHANG X， REN S， et al. Deep residual learning for image recognition ［C］// Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition. Piscataway： IEEE， 2016： 770-778. 10.1109/cvpr.2016.90
22	SHU Y， XU Y. End-to-end captcha recognition using deep CNN-RNN network ［C］// Proceedings of the IEEE 3rd Advanced Information Management， Communicates， Electronic and Automation Control Conference. Piscataway： IEEE， 2019： 54-58. 10.1109/imcec46724.2019.8983895

[1]	秦璟, 秦志光, 李发礼, 彭悦恒. 基于概率稀疏自注意力神经网络的重性抑郁疾患诊断[J]. 《计算机应用》唯一官方网站, 2024, 44(9): 2970-2974.
[2]	李云, 王富铕, 井佩光, 王粟, 肖澳. 基于不确定度感知的帧关联短视频事件检测方法[J]. 《计算机应用》唯一官方网站, 2024, 44(9): 2903-2910.
[3]	赵宇博, 张丽萍, 闫盛, 侯敏, 高茂. 基于改进分段卷积神经网络和知识蒸馏的学科知识实体间关系抽取[J]. 《计算机应用》唯一官方网站, 2024, 44(8): 2421-2429.
[4]	陈虹, 齐兵, 金海波, 武聪, 张立昂. 融合1D-CNN与BiGRU的类不平衡流量异常检测[J]. 《计算机应用》唯一官方网站, 2024, 44(8): 2493-2499.
[5]	张春雪, 仇丽青, 孙承爱, 荆彩霞. 基于两阶段动态兴趣识别的购买行为预测模型[J]. 《计算机应用》唯一官方网站, 2024, 44(8): 2365-2371.
[6]	高阳峄, 雷涛, 杜晓刚, 李岁永, 王营博, 闵重丹. 基于像素距离图和四维动态卷积网络的密集人群计数与定位方法[J]. 《计算机应用》唯一官方网站, 2024, 44(7): 2233-2242.
[7]	王东炜, 刘柏辰, 韩志, 王艳美, 唐延东. 基于低秩分解和向量量化的深度网络压缩方法[J]. 《计算机应用》唯一官方网站, 2024, 44(7): 1987-1994.
[8]	黄梦源, 常侃, 凌铭阳, 韦新杰, 覃团发. 基于层间引导的低光照图像渐进增强算法[J]. 《计算机应用》唯一官方网站, 2024, 44(6): 1911-1919.
[9]	李健京, 李贯峰, 秦飞舟, 李卫军. 基于不确定知识图谱嵌入的多关系近似推理模型[J]. 《计算机应用》唯一官方网站, 2024, 44(6): 1751-1759.
[10]	沈君凤, 周星辰, 汤灿. 基于改进的提示学习方法的双通道情感分析模型[J]. 《计算机应用》唯一官方网站, 2024, 44(6): 1796-1806.
[11]	李牧, 骆宇, 柯熙政. 基于调频连续波雷达的人体生命体征检测算法[J]. 《计算机应用》唯一官方网站, 2024, 44(6): 1978-1986.
[12]	姚迅, 秦忠正, 杨捷. 生成式标签对抗的文本分类模型[J]. 《计算机应用》唯一官方网站, 2024, 44(6): 1781-1785.
[13]	孙敏, 成倩, 丁希宁. 基于CBAM-CGRU-SVM的Android恶意软件检测方法[J]. 《计算机应用》唯一官方网站, 2024, 44(5): 1539-1545.
[14]	高文烁, 陈晓云. 基于节点结构的点云分类网络[J]. 《计算机应用》唯一官方网站, 2024, 44(5): 1471-1478.
[15]	席治远, 唐超, 童安炀, 王文剑. 基于双路时空网络的驾驶员行为识别[J]. 《计算机应用》唯一官方网站, 2024, 44(5): 1511-1519.

基于改进卷积神经网络和射频指纹的无人机检测与识别

UAV detection and recognition based on improved convolutional neural network and radio frequency fingerprint

RichHTML

PDF

可视化

摘要/Abstract

引用本文

使用本文

图/表 14

参考文献 22

相关文章 15

编辑推荐

Metrics