A secure authentication framework based on blockchain and ensemble learning is proposed to address the problem that vehicle identity privacy data in Internet of Vehicles (IoV) is vulnerable to theft and tampering. First, a secure and efficient authentication method based on blockchain and Physical Unclonable Function (PUF) is implemented, which ensures the identity privacy of the vehicle when accessing IoV, and improves the problem of high resource overhead of the traditional IoV authentication scheme while guaranteeing security, and the computational overhead is about 2.424 ms at the first level of security framework. Secondly, an intrusion detection method based on Whale Optimization Algorithm (WOA) and Extreme Gradient Boosting (XGBoost) is proposed, and the detection model trained based on this method can effectively detect various attacks against IoV. As a security method at the second level of secure framework, the method outperforms related works in detecting malicious attacks with a detection accuracy of 98.41% for ToN-IoT and 99.99% for BoT-IoT.

中文翻译：

---

基于区块链和集成学习的车联网安全认证框架


针对车联网中车辆身份隐私数据易被窃取和篡改的问题，提出一种基于区块链和集成学习的安全认证框架。首先，实现了基于区块链和物理不可克隆功能（PUF）的安全高效的认证方法，保证了车辆接入车联网时的​​身份隐私，在保证安全的同时改善了传统车联网认证方案资源开销高的问题，第一级安全框架的计算开销约为2.424 ms。其次，提出了一种基于鲸鱼优化算法（WOA）和极限梯度提升（XGBoost）的入侵检测方法，基于该方法训练的检测模型能够有效检测针对车联网的各种攻击。作为安全框架第二级的安全方法，该方法在检测恶意攻击方面优于相关工作，对于ToN-IoT的检测准确率达到98.41%，对于BoT-IoT的检测准确率达到99.99%。