تشخیص حملات در شبکه‌های خودرویی خودران با استفاده از یادگیری عمیق

نوع مقاله : مقاله پژوهشی

نویسندگان

استادیار گروه مهندسی کامپیوتر، دانشکده فنی و مهندسی، دانشگاه شهرکرد، شهرکرد، ایران

10.22034/abmir.2025.22510.1083

چکیده

وسایل نقلیه مدرن، ازجمله وسایل نقلیه خودران و وسایل نقلیه متصل، به‌طور فزاینده‌ای به محیط خارج از خود متصل می‌شوند و از این طریق عملکردها و خدمات مختلفی را ارائه می‌کنند. افزایش اتصال‌پذیری باعث افزایش حملات اینترنتی به وسایل نقلیه خودران گردیده است و درنتیجه، باعث آسیب‌پذیری این وسایل در برابر تهدیدات سایبری شده است. به‌دلیل ضعف و یا عدم وجود رویه‌های احراز هویت و رمزگذاری در شبکه‌های خودرو، استفاده از سیستم‌های تشخیص نفوذ یکی از روش‌های ضروری برای محافظت از سیستم خودروهای مدرن در برابر حملات سایبری است. در این مقاله، یک سیستم تشخیص نفوذ مبتنی بر یادگیری عمیق با استفاده از تشخیص تصاویر برای سیستم‌های وسایل نقلیه پیشنهادشده است. همچنین، از تکنیک تبدیل بردار ویژگی‌ها به تصاویر برای بهینه‌سازی تشخیص استفاده شده‌است. سیستم تشخیص نفوذ پیشنهادی با استفاده از تکنیک یادگیری گروهی مبتنی بر میانگین بهینه‌شده‌است. در آزمایش‌ها، روش پیشنهادی بیش از 99.25 درصد نرخ تشخیص و به همین مقدار معیار F1 را در دو مجموعه داده امنیتی استاندارد شامل مجموعه داده‌های Car-Hacking و مجموعه داده CICIDS2017 نشان داده است. همچنین، زمان اجرای روش بر روی تجهیزات اینترنت اشیاء اندازه‌گیری شده‌است که نشان‌دهنده قابلیت اجرای روش پیشنهادی است.

کلیدواژه‌ها

موضوعات


عنوان مقاله [English]

Network Attacks Detection in Autonomous Vehicles using Deep Nerul Network

نویسندگان [English]

  • Abbas Horri
  • Leila Samimi-Dehkordi
Assistant Professor, Department of Computer Engineering, Faculty of Engineering, Shahrekord University, Shahrekord Iran
چکیده [English]

Modern vehicles, including autonomous vehicles and connected vehicles, are increasingly connected to their external environment, thereby providing various functions and services. Increasing connectivity has increased cyber attacks on self-driving vehicles and, as a result, has made these devices vulnerable to cyber threats. Due to the weakness or absence of authentication and encryption procedures in car networks, the use of intrusion detection systems is one of the necessary methods to protect the modern car system against cyber attacks. In this paper, an intrusion detection system based on deep learning using image recognition for vehicle systems is proposed. Also, the technique of converting feature vectors into images has been used to optimize detection. The proposed intrusion detection system is optimized using average-based ensemble learning technique. In experiments, the proposed method has shown more than 99.25% detection rate and the same amount of F1 criterion in two standard security datasets including Car-Hacking dataset and CICIDS2017 dataset. Therefore, the proposed method is effective for detecting cyber attacks in vehicular networks. Also, the execution time of the method has been measured on the Internet of Things equipment, which shows the feasibility of the proposed method.

کلیدواژه‌ها [English]

  • Self-driving
  • Intrusion detection
  • Deep Learning
  • IoT
[1]     S. T. Mehedi, A. Anwar, Z. Rahman, and K. Ahmed, “Deep transfer learning based intrusion detection system for electric vehicular networks,” Sensors, vol. 21, no. 14, pp. 1–23, 2021, doi: 10.3390/s21144736.
[2]     Ramezanzadeh, M. Barzegar, and H. Motameni, “Automatic Security Assessment of Petri Nets-Based Threat Routes,” Electronic and Cyber Defense, vol. 9, no. 4, pp. 87–98, 2022. [In Persian].
[3]     L. Yang, A. Moubayed, I. Hamieh, and A. Shami, “Tree-based intelligent intrusion detection system in internet of vehicles,” 2019 IEEE Glob. Commun. Conf. GLOBECOM 2019 - Proc., no. October, 2019, doi: 10.1109/GLOBECOM38437.2019.9013892.
[4]     L. Yang, A. Moubayed, and A. Shami, “MTH-IDS: A Multitiered Hybrid Intrusion Detection System for Internet of Vehicles,” IEEE Internet Things J., vol. 9, no. 1, pp. 616–632, 2022, doi: 10.1109/JIOT.2021.3084796.
[5]     Rosay, F. Carlier, and P. Leroux, “Feed-forward neural network for Network Intrusion Detection,” IEEE Veh. Technol. Conf., vol. 2020-May, 2020, doi: 10.1109/VTC2020-Spring48590.2020.9129472.
[6]     H. M. Song, J. Woo, and H. K. Kim, “In-vehicle network intrusion detection using deep convolutional neural network,” Veh. Commun., vol. 21, p. 100198, 2020, doi: 10.1016/j.vehcom.2019.100198.
[7]     L. Yang, D. M. Manias, and A. Shami, “PWPAE: An Ensemble Framework for Concept Drift Adaptation in IoT Data Streams,” 2021 IEEE Glob. Commun. Conf. GLOBECOM 2021 - Proc., no. September, 2021, doi: 10.1109/GLOBECOM46510.2021.9685338.
[8]     E. Seo, H. M. Song, and H. K. Kim, “GIDS: GAN based Intrusion Detection System for In-Vehicle Network,” 2018 16th Annu. Conf. Privacy, Secur. Trust. PST 2018, pp. 0–5, 2018, doi: 10.1109/PST.2018.8514157.
[9]     Sharafaldin, A. H. Lashkari, and A. A. Ghorbani, “Toward generating a new intrusion detection dataset and intrusion traffic characterization,” ICISSP 2018 - Proc. 4th Int. Conf. Inf. Syst. Secur. Priv., vol. 2018-Janua, no. Cic, pp. 108–116, 2018, doi: 10.5220/0006639801080116.
[10] F. Aloraini, A. Javed, and O. Rana, "Adversarial attacks on intrusion detection systems in in-vehicle networks of connected and autonomous vehicles," Sensors, vol. 24, no. 12, p. 3848, 2024. doi: 10.3390/s24123848.
[11] S. Firasta, Y. R. Srivastava and V. Rao, "Cognitive Detection of Anomalies in Autonomous In-Vehicle Network Communication," 2024 International Conference on Advancements in Power, Communication and Intelligent Systems (APCI), KANNUR, India, 2024, pp. 1-6, doi: 10.1109/APCI61480.2024.10617212.
[12] M. D. Hossain, H. Inoue, H. Ochiai, D. Fall, and Y. Kadobayashi, “LSTM-based intrusion detection system for in-vehicle can bus communications,” IEEE Access, vol. 8, pp. 185489–185502, 2020, doi: 10.1109/ACCESS.2020.3029307.
[13] Y. Zhu et al., “Converting tabular data into images for deep learning with convolutional neural networks,” Sci. Rep., vol. 11, no. 1, pp. 1–12, 2021 doi: 10.1038/s41598-021-90923-y.
[14] Rahali, A. H. Lashkari, G. Kaur, L. Taheri, F. Gagnon, and F. Massicotte, “DIDroid: Android malware classification and characterization using deep image learning,” ACM Int. Conf. Proceeding Ser., pp. 70–82, 2020, doi: 10.1145/3442520.3442522.
[15] T. H. De Huang and H. Y. Kao, “R2-D2: ColoR-inspired Convolutional NeuRal Network (CNN)-based AndroiD Malware Detections,” in Proceedings - 2018 IEEE International Conference on Big Data, Big Data 2018, Jan. 2019, pp. 2633–2642, doi: 10.1109/BigData.2018.8622324.
[16] Bruna and S. Mallat, “Invariant scattering convolution networks,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 35, no. 8, pp. 1872–1886, 2013, doi: 10.1109/TPAMI.2012.230.
[17] Q. Sun and B. Pfahringer, “Bagging ensemble selection,” Lect. Notes Comput. Sci. (including Subser. Lect. Notes Artif. Intell. Lect. Notes Bioinformatics), vol. 7106 LNAI, pp. 251–260, 2011, doi: 10.1007/978-3-642-25832-9_26.
[18] Tanha, Y. Abdi, N. Samadi, N. Razzaghi, and M. Asadpour, “Boosting methods for multi-class imbalanced data classification: an experimental review,” J. Big Data, vol. 7, no. 1, 2020, doi: 10.1186/s40537-020-00349-y.
[19] R. E. Schapire, "The boosting approach to machine learning: an overview," in Nonlinear Estimation and Classification, D. D. Denison, M. H. Hansen, C. C. Holmes, B. Mallick, and B. Yu, Eds. New York, NY: Springer, 2003, vol. 171, pp. 1-14. doi: 10.1007/978-0-387-21579-2_9.
[20] Moubayed, A. Shami, P. Heidari, A. Larabi, and R. Brunner, “Edge-Enabled V2X Service Placement for Intelligent Transportation Systems,” IEEE Trans. Mob. Comput., vol. 20, no. 4, pp. 1380–1392, 2021, doi: 10.1109/TMC.2020.2965929.