The Internet of Vehicles (IoV) is an innovative area of interest in modern mobility that is rapidly evolving while facing complex challenges. Traditional IoV networks are susceptible to intrusion threats, which can lead to data leakage and seizure of vehicle control by attackers, thereby endangering vehicle users’ privacy and personal safety. An Attribute and User Trust Score-based Zero Trust Access Control Model (AU-ZTAC) is proposed, combining the zero-trust and attribute-based access control models to meet network protection requirements while achieving fine-grained dynamic access control and incorporating trust evaluation in the access control process to better reflect users’ intent. Experimental results demonstrate the effectiveness and feasibility of trust assessment through the proposed model. A comparison with the classical schemes illustrates that AU-ZTAC allows for more flexible and fine-grained access control in complex access control environments while improving IoV security.

With the rapid development of smart technologies, the Internet of Vehicles (IoV) is revolutionizing transportation and mobility. However, the complexity and interconnectedness of IoV systems lead to a growing number of security incidents caused by vulnerabilities. Current vulnerability classification algorithms often struggle to address the low occurrence frequency and incomplete information associated with IoV vulnerabilities, resulting in decreased precision and recall rates of classifiers. To address these challenges, an effective vulnerability classification algorithm (KG-KNN), is proposed, designed to handle imbalanced sample data. KG-KNN integrates the vulnerability information of IoV and the association relationship between features by constructing a feature knowledge graph to form a complete knowledge system. It adds the correlation relationship between features to the similarity calculation, calculates vulnerability similarity from multiple dimensions, and improves the prediction performance of the classifier. The experimental results show that compared to the k-NearestNeighbor (KNN), Support Vector Machine (SVM), Deep Nueral Network (DNN) and TFI-DNN classification algorithms, KG-KNN can effectively deal with imbalanced sample data and has different degrees of improvement in precision, recall, and the F1 score.

The advancement of Web3.0 technology has brought about an urgent need for ensuring the safety and reliability of the software systems. Program analysis, a crucial aspect of software security research needs a unified solution for various cross-language program. Moreover, the previous studies regard the necessity of capturing structured features from the ASTs, commonly holding a conception that plain text and compiled binary instructions were challenging to represent and identify. This paper proposes a method of program instruction file embedding directly into natural language. we train large language models (LLMs) with 20 billion to identify defect and non-defect of WebAssembly. Experiment results demonstrate that program instruction analysis surpasses traditional techniques, achieving state-of-the-art accuracy exceeding 98.1 percent. Our study also suggests a practical approach of plain text embedding using a 7.65 billion parameters language model. Interestingly, misformatted source cdoes are readable to humans but un-compilable, and the accuracy remains above 98.63 percent. This paper not only introduces novel instruction and plain text embedding approach for future program security analysis, but also provides new insights for subsequent research about the three program analysis forms of plain text, ASTs, and instructions.