ZETA: ZEro-Trust Attack Framework with Split Learning for Autonomous Vehicles in 6G Networks

被引:5
作者
Khowaja, Sunder Ali [1 ]
Khuwaja, Parus [2 ]
Dev, Kapal [3 ,4 ]
Singh, Keshav [5 ]
Nkenyereye, Lewis [6 ]
Kilper, Dan [7 ,8 ]
机构
[1] Digital & Data Technol Univ Dublin, Fac Comp, Dublin, Ireland
[2] Univ Sindh, Inst Business Adm, Jamshoro, Pakistan
[3] Munster Technol Univ, CONNECT Ctr, Cork, Ireland
[4] Munster Technol Univ, Dept Comp Sci, Cork, Ireland
[5] Natl Sun Yat Sen Univ, Inst Commun Engn ICE, Kaohsiung, Taiwan
[6] Sejong Univ, Dept Comp & Informat Secur, Seoul, South Korea
[7] Trinity Coll Dublin, CONNECT Ctr, Dublin, Ireland
[8] Trinity Coll Dublin, Sch Engn, Dublin, Ireland
来源
2024 IEEE WIRELESS COMMUNICATIONS AND NETWORKING CONFERENCE, WCNC 2024 | 2024年
关键词
Zero-Trust Attack; Autonomous vehicles; Split Learning; 6G network; Immersive applications;
D O I
10.1109/WCNC57260.2024.10571158
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
In past, due to data and model security concerns, modern communication systems mainly focus on the use of edge computing devices for enabling immersive applications and services. Federated learning is one of the preferred solutions but it stresses the computation capability of the edge devices for immersive applications. Much research is now focusing on split learning as an alternative due to its ability of performing joint training with limited computing resources. However, split learning is also vulnerable to data reconstruction, feature space hijacking, and model inversion attacks, which are quite common concerning immersive applications such as Metaverse. In this regard, we propose a ZEro-Trust Attack (ZETA) framework for data reconstruction and model inversion attacks for autonomous vehicles opting for split learning strategies. We propose the joint training of client, server, and shadow models for both the reconstruction and main task to fool existing methods. Our experimental results demonstrate that the proposed method is capable of reconstructing client's data with an error of 0.0032. This study is proposed as a basis to design more sophisticated defense mechanisms for autonomous vehicles to protect user services in 5G/6G networks.
引用
收藏
页数:6
相关论文
共 15 条
[1]   SplitGuard: Detecting and Mitigating Training-Hijacking Attacks in Split Learning [J].
Erdogan, Ege ;
Kupcu, Alptekin ;
Cicek, A. Ercument .
PROCEEDINGS OF THE 21ST WORKSHOP ON PRIVACY IN THE ELECTRONIC SOCIETY, WPES 2022, 2022, :125-137
[2]  
Fu J., 2023, NETW DISTR SYST SEC
[3]  
Khowaja S.A., 2022, IEEE T NETWORK SCI E
[4]  
Khowaja S.A., 2023, IEEE GLOBECOM
[5]   SPIN: Simulated Poisoning and Inversion Network for Federated Learning-Based 6G Vehicular Networks [J].
Khowaja, Sunder Ali ;
Khuwaja, Parus ;
Dev, Kapal ;
Antonopoulos, Angelos .
ICC 2023-IEEE INTERNATIONAL CONFERENCE ON COMMUNICATIONS, 2023, :6205-6210
[6]  
Krizhevsky A., 2009, Technical report
[7]  
Li Y., 2023, ARXIV
[8]  
Li Y., 2023, P IEEE CVF C COMP VI, P1904
[9]   Unleashing the Tiger: Inference Attacks on Split Learning [J].
Pasquini, Dario ;
Ateniese, Giuseppe ;
Bernaschi, Massimo .
CCS '21: PROCEEDINGS OF THE 2021 ACM SIGSAC CONFERENCE ON COMPUTER AND COMMUNICATIONS SECURITY, 2021, :2113-2129
[10]   Man vs. computer: Benchmarking machine learning algorithms for traffic sign recognition [J].
Stallkamp, J. ;
Schlipsing, M. ;
Salmen, J. ;
Igel, C. .
NEURAL NETWORKS, 2012, 32 :323-332