ADS-B attack detection method based on improved DeepLabV3+

doi:10.11772/j.issn.1001-9081.2025081040

Journal of Computer Applications ›› 2026, Vol. 46 ›› Issue (9): 2867-2876.DOI: 10.11772/j.issn.1001-9081.2025081040

• Cyber security • Previous Articles

ADS-B attack detection method based on improved DeepLabV3+

Lixia XIE¹, Chaoyue CAO², Hongyu YANG¹^,²()

^1.College of Computer Science and Artificial Intelligence，Civil Aviation University of China，Tianjin 300300，China
^2.College of Safety Science and Engineering，Civil Aviation University of China，Tianjin 300300，China

Received:2025-09-09 Revised:2025-11-05 Accepted:2025-11-10 Online:2025-11-17 Published:2026-09-10
Contact: Hongyu YANG
About author:XIE Lixia， born in 1974， M. S.， professor. Her research interests include network information security.
CAO Chaoyue， born in 1998， M. S. candidate. His research interests include network information security， ADS-B security.
YANG Hongyu， born in 1969， Ph. D.， professor. His research interests include network and system security， software security detection， situational perception of network security.
Supported by:
Key Project of Civil Aviation Joint Fund of National Natural Science Foundation of China(U2433205)

基于改进DeepLabV3+的ADS-B攻击检测方法

谢丽霞¹, 曹超越², 杨宏宇¹^,²()

^1.中国民航大学计算机与人工智能学院，天津 300300
^2.中国民航大学安全科学与工程学院，天津 300300

通讯作者: 杨宏宇
作者简介:谢丽霞（1974—），女，重庆人，教授，硕士，CCF高级会员，主要研究方向：网络信息安全
曹超越（1998—），男，河北沧州人，硕士研究生，主要研究方向：网络信息安全、ADS-B安全
杨宏宇（1969—），男，吉林长春人，教授，博士生导师，博士，CCF高级会员，主要研究方向：网络与系统安全、软件安全检测、网络安全态势感知。
基金资助:
国家自然科学基金民航联合研究基金重点项目(U2433205)

Abstract

Abstract:

To address the problems that the communication messages of Automatic Dependent Surveillance-Broadcast （ADS-B） system lack encryption and identity authentication mechanisms and are vulnerable to spoofing attacks such as replay and trajectory forgery， but traditional detecting models cannot identify anomalies in the system， an ADS-B attack detection method based on an improved DeepLabV3+ model was proposed. First， the original tabular ADS-B data were reconstructed into grayscale images through grayscale mapping， thereby enabling the visualization of spatio-temporal features. Then， to overcome the weak feature extraction capability and blurred segmentation boundaries of DeepLabV3+ model when processing ADS-B image data， a Haar Wavelet Downsampling （HWD） module was used to replace the max-pooling operation in the ResNet101 backbone network， thereby preserving key features such as trajectory anomalies and local perturbations during dimensionality reduction effectively. Finally， a Simple parameter-free Attention Mechanism （SimAM） was introduced after the ADS-B low-level feature map to enhance the representation of local spatial structures， and a Fusion Convolutional Block Attention Module （F_CBAM） was incorporated after the Atrous Spatial Pyramid Pooling （ASPP） module to highlight critical semantic regions through the parallel interaction of channel and spatial attentions. Experimental results on an ADS-B dataset show that compared with the baseline DeepLabV3+ model， the proposed method achieves a 6.73 percentage points improvement in mean Intersection over Union （mIoU）. In complex attack scenarios， the proposed method has the F1-scores for Denial of Service （DoS） attacks and random noise reached 96.07% and 92.55%， respectively， which are significantly higher than those of mainstream models such as U-Net （U-shaped convolutional Network）， SegFormer （Segmentation Transformer）， and K-Net （Kernel Network）， providing a new approach for accurate detection of ADS-B attacks.

Key words: Automatic Dependent Surveillance-Broadcast (ADS-B), attack detection, semantic segmentation, Haar Wavelet Downsampling (HWD), attention mechanism

摘要：

针对广播式自动相关监视（ADS-B）系统通信报文缺乏加密与身份认证机制，易受到重放和轨迹伪造等欺骗性攻击，传统检测模型难以精确识别异常的问题，提出一种基于改进DeepLabV3+的ADS-B攻击检测方法。首先，通过构建灰度图将原本的ADS-B表格数据重构为灰度图，实现时空特征的图像化表达；其次，针对DeepLabV3+模型对ADS-B图像数据细节提取能力弱和分割图像边界模糊的问题，采用Haar小波下采样（HWD）模块替代ResNet101主干网络中的最大池化操作，在降维过程中有效保留轨迹异常与局部突变等关键特征；最后，在ADS-B低层特征图后引入无参注意力机制（SimAM），以强化局部空间结构表达，并在空洞空间金字塔池化（ASPP）模块后引入融合卷积块注意力模块（F_CBAM），通过通道与空间注意力并行交互突出关键语义区域。在ADS-B数据集上的实验结果表明，与基础DeepLabV3+模型相比，所提方法的平均交并比（mIoU）提高了6.73个百分点；在面对复杂攻击，尤其是拒绝服务（DoS）攻击和随机噪声攻击时，所提方法的F1分数分别达到96.07%与92.55%，显著优于U-Net（U-shaped convolutional Network）、SegFormer（Segmentation Transformer）和K-Net（Kernel Network）等主流模型，为ADS-B攻击的精准检测提供了新思路。

CLC Number:

TP393.8

Lixia XIE, Chaoyue CAO, Hongyu YANG. ADS-B attack detection method based on improved DeepLabV3+[J]. Journal of Computer Applications, 2026, 46(9): 2867-2876.

谢丽霞, 曹超越, 杨宏宇. 基于改进DeepLabV3+的ADS-B攻击检测方法[J]. 《计算机应用》唯一官方网站, 2026, 46(9): 2867-2876.

Figures/Tables 10

References 34

[1]	Khan H A， Khan H， Ghafoor S， et al. A survey on security of Automatic Dependent Surveillance-Broadcast （ADS-B） protocol： challenges， potential solutions and future directions ［J］. IEEE Communications Surveys and Tutorials， 2025， 27（5）： 3199-3226.
[2]	Wu Z， Shang T， Guo A. Security issues in Automatic Dependent Surveillance-Broadcast （ADS-B）： a survey ［J］. IEEE Access， 2020， 8： 122147-122167.
[3]	Strohmeier M， Lenders V， Martinovic I. On the security of the automatic dependent surveillance-broadcast protocol ［J］. IEEE Communications Surveys and Tutorials， 2015， 17（2）： 1066-1087.
[4]	杨宏宇，张恒鸣，朱友文，等. 面向LBS的ADS-B报文定位数据本地化差分隐私保护方法［J］. 网络与信息安全学报， 2025， 11（4）： 32-49.
	Yang Hongyu， Zhang Hengming， Zhu Youwen， et al. Differential privacy protection method for localization of ADS-B message positioning data oriented to LBS ［J］. Chinese Journal of Network and Information Security， 2025， 11（4）： 32-49.
[5]	Çevik N， Akleylek S. Sok of machine learning and deep learning based anomaly detection methods for automatic dependent surveillance-broadcast ［J］. IEEE Access， 2024， 12： 35643-35662.
[6]	Corrado S J， Puranik T G， Fischer O P， et al. A clustering-based quantitative analysis of the interdependent relationship between spatial and energy anomalies in ADS-B trajectory data ［J］. Transportation Research Part C： Emerging Technologies， 2021， 131： No.103331.
[7]	王布宏，罗鹏，阳勇，等. 空中交通管理系统网络安全研究综述与展望［J］. 电子与信息学报， 2025， 47（5）： 1230-1265.
	Wang Buhong， Luo Peng， Yang Yong， et al. A review and prospect of cybersecurity research on air traffic management systems ［J］. Journal of Electronics and Information Technology， 2025， 47（5）： 1230-1265.
[8]	Melvyn P， Couturier R， Ambert F， et al. Detecting ghost aircraft flooding in the surveillance of low-flying civil and military aircraft［C］// ICCCR 2024. Piscataway： IEEE， 2024： 328-332.
[9]	Zhang X， Li J. SR-GNN： spectral residuals with graph neural networks for anomaly detection in ADS-B data ［C］// FCS 2023， CCIS 1992. Singapore： Springer， 2024： 503-516.
[10]	Vajrobol V， Saxena G J， Singh S， et al. Enhancing aviation control security through ADS-B injection detection using ensemble meta-learning models with explainable AI ［J］. Alexandria Engineering Journal， 2025， 112： 63-73.
[11]	Habler E， Shabtai A. Using LSTM encoder-decoder algorithm for detecting anomalous ADS-B messages ［J］. Computers and Security， 2018， 78： 155-173.
[12]	Habler E， Shabtai A. Analyzing sequences of airspace states to detect anomalous traffic conditions ［J］. IEEE Transactions on Aerospace and Electronic Systems， 2022， 58（3）： 1843-1857.
[13]	Ji Y， Wang L， Wu W， et al. A method for LSTM-based trajectory modeling and abnormal trajectory detection ［J］. IEEE Access， 2020， 8： 104063-104073.
[14]	Yue M， Zheng H， Cui H， et al. GAN-LSTM-based ADS-B attack detection in the context of air traffic control ［J］. IEEE Internet of Things Journal， 2023， 10（14）： 12651-12665.
[15]	Joseph N S， Banerjee C， Pasiliao E， et al. FlightSense： a spoofer detection and aircraft identification system using raw ADS-B data［C］// Big Data 2020. Piscataway： IEEE， 2020： 3885-3894.
[16]	Heiselberg P， Sørensen K A， Heiselberg H. Aircraft detection and state estimation ［C］// IGARSS 2024. Piscataway： IEEE， 2024： 7693-7696.
[17]	Luo P， Wang B， Tian J. TTSAD： TCN-Transformer-SVDD model for Anomaly Detection in air traffic ADS-B data ［J］. Computers and Security， 2024， 141： No.103840.
[18]	Cai K， Di Z， Zhu Y， et al. An attention-based convolutional network framework for detection and localization of GNSS interference sources ［J］. IEEE Transactions on Aerospace and Electronic Systems， 2024， 60（3）： 2995-3011.
[19]	Chevrot A， Vernotte A， Legeard B. CAE： contextual auto-encoder for multivariate time-series anomaly detection in air transportation ［J］. Computers and Security， 2022， 116： No.102652.
[20]	Fried A， Last M. Facing airborne attacks on ADS-B data with autoencoders ［J］. Computers and Security， 2021， 109： No.102405.
[21]	Luo P， Wang B， Li T， et al. ADS-B anomaly data detection model based on VAE-SVDD ［J］. Computers and Security， 2021， 104： No.102213.
[22]	丁建立，张琪琪，王静，等. 基于Transformer-VAE的ADS-B异常检测方法［J］. 系统工程与电子技术， 2023， 45（11）： 3680-3689.
	Ding Jianli， Zhang Qiqi， Wang Jing， et al. ADS-B anomaly detection method based on Transformer-VAE［J］. Systems Engineering and Electronics， 2023， 45（11）： 3680-3689.
[23]	Chen L C， Zhu Y， Papandreou G， et al. Encoder-decoder with atrous separable convolution for semantic image segmentation［C］// ECCV 2018， LNCS 11211. Cham： Springer， 2018： 833-851.
[24]	Xu G， Liao W， Zhang X， et al. Haar wavelet downsampling： a simple but effective downsampling module for semantic segmentation ［J］. Pattern Recognition， 2023， 143： No.109819.
[25]	He K， Zhang X， Ren S， et al. Deep residual learning for image recognition ［C］// CVPR 2016. Piscataway： IEEE， 2016： 770-778.
[26]	Yang L， Zhang R Y， Li L， et al. SimAM： a simple， parameter-free attention module for convolutional neural networks ［J］. Proceedings of Machine Learning Research， 2021， 139： 11863-11874.
[27]	Woo S， Park J， Lee J Y， et al. CBAM： convolutional block attention module ［C］// ECCV 2018， LNCS 11211. Cham： Springer， 2018： 3-19.
[28]	Network OpenSky. OpenSky Network homepage ［EB/OL］. ［2025-04-10］..
[29]	Ronneberger O， Fischer P， Brox T. U-net： convolutional networks for biomedical image segmentation ［C］// MICCAI 2015， LNCS 9351. Cham： Springer， 2015： 234-241.
[30]	Xie E， Wang W， Yu Z， et al. SegFormer： simple and efficient design for semantic segmentation with Transformers ［C］// NeurIPS 2021. Red Hook： Curran Associates Inc.， 2021： 12077-12090.
[31]	Zhang W， Pang J， Chen K， et al. K-Net： towards unified image segmentation ［C］// NeurIPS 2021. Red Hook： Curran Associates Inc.， 2021： 10326-10338.
[32]	Poudel R P K， Liwicki S， Cipolla R. Fast-SCNN： fast semantic segmentation network［PP/OL］. arXiv （2019-02-12）［2025-04-10］..
[33]	Wang Q， Wu B， Zhu P， et al. ECA-Net： efficient channel attention for deep convolutional neural networks ［C］// CVPR 2020. Piscataway： IEEE， 2020： 11531-11539.
[34]	Guo M H， Liu Z N， Mu T J， et al. Beyond self-attention： external attention using two linear layers for visual tasks ［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence， 2023， 45（5）： 5436-5447.

方法	精确率	召回率	F1分数
LSTM-ED^［11］	86.61	91.64	88.04
VAE-SVDD^［21］	87.43	90.58	88.59
GAN-LSTM^［14］	86.88	88.84	87.25
TTSAD^［17］	87.85	90.46	88.43
DeepLabV3+^［23］	96.79	95.80	96.26
本文方法	98.79	98.21	98.48

方法	精确率	召回率	F1分数
LSTM-ED^［11］	86.61	91.64	88.04
VAE-SVDD^［21］	87.43	90.58	88.59
GAN-LSTM^［14］	86.88	88.84	87.25
TTSAD^［17］	87.85	90.46	88.43
DeepLabV3+^［23］	96.79	95.80	96.26
本文方法	98.79	98.21	98.48

攻击类型	评估指标	U-Net^［29］	SegFormer^［30］	K-Net^［31］	Fast-SCNN^［32］	DeepLabV3+^［23］	本文方法
高度偏移	精确率	93.82	84.00	92.39	53.64	93.26	96.24
	召回率	80.51	83.17	74.60	84.08	85.55	95.43
	F1分数	86.66	83.59	82.55	65.50	89.24	95.83
速度偏移	精确率	95.50	81.81	86.42	98.32	88.50	97.37
	召回率	85.42	82.68	91.56	50.75	98.46	99.70
	F1分数	90.18	82.24	88.92	66.94	93.21	98.52
重放攻击	精确率	43.77	78.72	51.17	58.30	85.23	82.80
	召回率	61.20	40.23	50.37	32.40	89.70	98.30
	F1分数	51.04	53.25	50.76	41.65	87.41	89.89
随机噪声	精确率	77.82	73.97	80.01	70.02	94.32	89.18
	召回率	85.68	85.11	82.09	79.95	76.94	96.19
	F1分数	81.56	79.15	81.03	74.65	84.75	92.55
航迹异常	精确率	98.83	91.48	96.06	97.89	93.48	93.73
	召回率	90.72	94.65	97.86	92.26	99.97	99.31
	F1分数	94.60	93.04	96.95	94.99	96.62	96.44
DoS攻击	精确率	47.15	33.52	38.74	29.11	90.01	95.96
	召回率	79.62	64.82	68.58	91.37	93.28	96.17
	F1分数	59.23	44.19	49.51	44.16	91.62	96.07
平均交并比		68.75	63.76	66.71	55.56	84.55	91.28

攻击类型	评估指标	U-Net^［29］	SegFormer^［30］	K-Net^［31］	Fast-SCNN^［32］	DeepLabV3+^［23］	本文方法
高度偏移	精确率	93.82	84.00	92.39	53.64	93.26	96.24
	召回率	80.51	83.17	74.60	84.08	85.55	95.43
	F1分数	86.66	83.59	82.55	65.50	89.24	95.83
速度偏移	精确率	95.50	81.81	86.42	98.32	88.50	97.37
	召回率	85.42	82.68	91.56	50.75	98.46	99.70
	F1分数	90.18	82.24	88.92	66.94	93.21	98.52
重放攻击	精确率	43.77	78.72	51.17	58.30	85.23	82.80
	召回率	61.20	40.23	50.37	32.40	89.70	98.30
	F1分数	51.04	53.25	50.76	41.65	87.41	89.89
随机噪声	精确率	77.82	73.97	80.01	70.02	94.32	89.18
	召回率	85.68	85.11	82.09	79.95	76.94	96.19
	F1分数	81.56	79.15	81.03	74.65	84.75	92.55
航迹异常	精确率	98.83	91.48	96.06	97.89	93.48	93.73
	召回率	90.72	94.65	97.86	92.26	99.97	99.31
	F1分数	94.60	93.04	96.95	94.99	96.62	96.44
DoS攻击	精确率	47.15	33.52	38.74	29.11	90.01	95.96
	召回率	79.62	64.82	68.58	91.37	93.28	96.17
	F1分数	59.23	44.19	49.51	44.16	91.62	96.07
平均交并比		68.75	63.76	66.71	55.56	84.55	91.28

方法	主干网络	mIoU	mPrecision	mRecall
DeepLabV3+	ResNet101	75.20	85.22	83.28
DeepLabV3+HWD	ResNet101	81.05	86.23	90.50

ADS-B attack detection method based on improved DeepLabV3+

基于改进DeepLabV3+的ADS-B攻击检测方法

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 10

References 34

Related Articles 15

Recommended Articles

Metrics

注意力机制	mIoU	mPrecision	mRecall
SimAM	83.22	87.55	91.80
EANet	82.85	86.67	90.78
ECA	81.95	86.35	90.13
CBAM	82.50	87.36	91.82
F_CBAM	83.75	88.15	92.32

ResNet101	SimAM	EANet	ECA	CBAM	F_ CBAM	HWD	mIoU	mPrecision
􀳫	􀳫		􀳫			􀳫	84.85	87.82
􀳫	􀳫			􀳫		􀳫	84.25	88.55
􀳫	􀳫				􀳫	􀳫	91.28	93.25
􀳫		􀳫	􀳫			􀳫	83.25	87.26
􀳫		􀳫		􀳫		􀳫	83.60	89.35
􀳫		􀳫			􀳫	􀳫	86.68	91.66

[1]	Yifan WANG, Hu HAN, Dong LI, Yating FAN, Lin LI. Heterogeneous enhancement and multi-source knowledge fusion for aspect-based sentiment analysis [J]. Journal of Computer Applications, 2026, 46(9): 2752-2760.
[2]	Jielong BAI, Chenyun FANG, Zhiwei QIAO. Sparse CT artifact suppresion Transformer network based on multi-scale attention adaptive fusion [J]. Journal of Computer Applications, 2026, 46(8): 2660-2667.
[3]	Guanghui LI, Licheng QU. Dynamic dilated convolution and hierarchical local-global attention model based on improved Lite-Mono architecture [J]. Journal of Computer Applications, 2026, 46(8): 2567-2576.
[4]	Yuanhang HUANG, Na RONG. Model-free photovoltaic hosting capacity assessment method using attention mechanism and deep reinforcement learning [J]. Journal of Computer Applications, 2026, 46(8): 2708-2715.
[5]	Xiaojin GUO, Xuyang SUI, Kenan ZHOU. Channel estimation of STAR-RIS-assisted hybrid-field communication system based on deep learning [J]. Journal of Computer Applications, 2026, 46(8): 2548-2554.
[6]	Ming LIU, Dongqi SHEN, Ziyang MENG. Point cloud registration network with dual-branch multi-level feature fusion [J]. Journal of Computer Applications, 2026, 46(8): 2584-2593.
[7]	Fengchun LIU, Xinying SHAO, Chunying ZHANG, Liya WANG, Jing REN. FCMdepth： monocular depth estimation framework with multi-scale feature optimization [J]. Journal of Computer Applications, 2026, 46(8): 2603-2611.
[8]	Wei LIU, Weigang LI, Zhiqiang TIAN. Representation learning method of hierarchical rotation-invariant geometric structure for point cloud classification and segmentation [J]. Journal of Computer Applications, 2026, 46(8): 2594-2602.
[9]	Jianqiang YAN, Beibei DONG, Boting QU, Chen PENG. Bi-directional diffusion dynamic graph convolutional network with multi-source information fusion and graph-level attention for traffic flow prediction [J]. Journal of Computer Applications, 2026, 46(7): 2327-2333.
[10]	Xiuli DU, Xing GAO, Xiaoyu ZHANG, Chengsheng PAN, Qijie ZOU. Video snapshot compressive imaging reconstruction method based on dense spatio-temporal deformable attention [J]. Journal of Computer Applications, 2026, 46(7): 2288-2296.
[11]	Guoyou ZHANG, Hongyu NIE, Lihu PAN, Rundong LEI. Point-MLPBLS： point cloud semantic segmentation network based on MLP cascaded broad learning system [J]. Journal of Computer Applications, 2026, 46(7): 2259-2266.
[12]	Wenling LIU, Yong LI, Jiahui LI, Houkang ZHANG. Unsupervised dehazing method for remote sensing images with non-uniform haze [J]. Journal of Computer Applications, 2026, 46(7): 2297-2306.
[13]	Yasong ZHANG, Bihui CONG, Shuang XU. Graph neural network node classification model incorporating clustering coefficients [J]. Journal of Computer Applications, 2026, 46(6): 1855-1862.
[14]	Jun QIN, Xintao JIAO, Biqing ZENG. Non-intrusive load monitoring method combining BiLSTM-Transformer and Kolmogorov-Arnold network [J]. Journal of Computer Applications, 2026, 46(6): 2026-2033.
[15]	Jinxiao ZHANG, Chenglong LI, Xinyan GAO, Ming ZHANG. 3D human pose estimation model based on temporal-spatial feature pyramid network and multi-hypothesis interaction mechanism [J]. Journal of Computer Applications, 2026, 46(6): 1965-1972.