Scholarlink

<p>The Internet of Vehicles (IoV) has become a fundamental component of intelligent transportation systems, enabling advanced services such as traffic management, collision avoidance, and driver assistance. However, the high level of openness and connectivity in IoV makes it highly vulnerable to Distributed Denial of Service (DDoS) attacks. Such attacks can severely disrupt vehicular communications and compromise user safety. Meanwhile, existing intrusion detection systems (IDSs) suffer from limitations in adapting to IoV’s dynamic topology and resource-constrained environment. To address these challenges, this paper proposes a novel ensemble learning–based IDS framework for accurate detection of multiple DDoS attack types. The framework integrates two optimized convolutional neural network (CNN) models to enhance detection performance. In addition, a dynamic deployment algorithm is designed to determine the optimal placement of the IDS among fog servers, unmanned aerial vehicles (UAVs), and roadside units (RSUs) based on real-time network conditions. To further improve resource efficiency, a game-theoretic strategy is employed to minimize energy consumption while maintaining high detection accuracy. Extensive experiments conducted on three benchmark datasets—VDoS-LRS, CICDDoS2019, and VDDD—demonstrate that the proposed IDS achieves over 99% detection accuracy and significantly outperforms existing methods in terms of precision and computational efficiency. These results confirm the effectiveness of combining ensemble deep learning with game-theoretic strategy for securing IoV against DDoS attacks.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Enhancing security in iov: an ensemble learning approach for DDoS detection

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年3月19日

查看全文
  • Seyed Amin Hosseini Seno,
  • Zahra Janfada,
  • Somayeh Soltani,
  • Amirhossein Mohajerzadeh

摘要

The Internet of Vehicles (IoV) has become a fundamental component of intelligent transportation systems, enabling advanced services such as traffic management, collision avoidance, and driver assistance. However, the high level of openness and connectivity in IoV makes it highly vulnerable to Distributed Denial of Service (DDoS) attacks. Such attacks can severely disrupt vehicular communications and compromise user safety. Meanwhile, existing intrusion detection systems (IDSs) suffer from limitations in adapting to IoV’s dynamic topology and resource-constrained environment. To address these challenges, this paper proposes a novel ensemble learning–based IDS framework for accurate detection of multiple DDoS attack types. The framework integrates two optimized convolutional neural network (CNN) models to enhance detection performance. In addition, a dynamic deployment algorithm is designed to determine the optimal placement of the IDS among fog servers, unmanned aerial vehicles (UAVs), and roadside units (RSUs) based on real-time network conditions. To further improve resource efficiency, a game-theoretic strategy is employed to minimize energy consumption while maintaining high detection accuracy. Extensive experiments conducted on three benchmark datasets—VDoS-LRS, CICDDoS2019, and VDDD—demonstrate that the proposed IDS achieves over 99% detection accuracy and significantly outperforms existing methods in terms of precision and computational efficiency. These results confirm the effectiveness of combining ensemble deep learning with game-theoretic strategy for securing IoV against DDoS attacks.