Scholarlink

<p>Vehicular Ad hoc Network (VANET) highly relies on communication between vehicles that eventually leads to Intelligent Transportation Systems (ITS). The malicious nodes send false messages, thus disrupts the traffic flow and potential accidents. The dynamic characteristics of VANET make the data routing process more difficult, thus prioritizing high throughput, low latency and higher reliability is significant in vehicular communication. To improve network security, a novel authentication-based routing protocol is proposed in VANET using machine learning approaches. The proposed framework constitutes two major concerns authentication and optimal routing, thus ensuring efficient and secure data paths by minimizing malicious attacks. The first objective is to implement the authentication mechanism of vehicle nodes to enhance network security by preventing the manipulation of data from malicious nodes. The node authentication is done by using the Hybrid Adaptive Networks (HAN) model, where the Conditional Random Field (CRF) and Ridge classifier are adopted. Here, the parameter optimization is done using the Updated Random Attribute-based Circle-Inspired Optimization (URA-CIO) for improving the performance of the HAN model. This HAN-based node authentication verifies the identity of vehicles and RSUs participating in the VANET. After the node authentication is done, the second objective, optimal routing is executed. In order to perform effective routing, the optimal path is preferred based on the recommended URA-CIO approach. Thus, the optimal routing is made based on multi-objective constraints such as shortest path, residual energy, average end-to-end delay, path availability, throughput, and packet delivery ratio. The accuracy of the proposed URA-CIO-HAN-based node authentication and optimal routing task in VANET has achieved 92% accuracy, and also the conventional models, such as Bi-linear mapping, KNN, ELM, and DMN attained 84.73%, 85.61%, 88.42%, and 89.82% accuracy rate. Thus, the developed authentication-based optimal routing scheme allows the VANET to dynamically adjust to diverse conditions including security threats and network congestion and the optimal performance of the method is assessed using multiple metrics.</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

An Efficient Hybrid Adaptive Network-based Node Authentication and Optimal Routing Mechanism for Enhancing Security in VANET Environment

  • 期刊论文

  • 发表日期：

    2026年3月14日

查看全文
  • Nageswara Rao Putta,
  • Yeetha Ravi Kumar,
  • P. Nirupama,
  • M. Vikram,
  • Gudivada Lokesh

摘要

Vehicular Ad hoc Network (VANET) highly relies on communication between vehicles that eventually leads to Intelligent Transportation Systems (ITS). The malicious nodes send false messages, thus disrupts the traffic flow and potential accidents. The dynamic characteristics of VANET make the data routing process more difficult, thus prioritizing high throughput, low latency and higher reliability is significant in vehicular communication. To improve network security, a novel authentication-based routing protocol is proposed in VANET using machine learning approaches. The proposed framework constitutes two major concerns authentication and optimal routing, thus ensuring efficient and secure data paths by minimizing malicious attacks. The first objective is to implement the authentication mechanism of vehicle nodes to enhance network security by preventing the manipulation of data from malicious nodes. The node authentication is done by using the Hybrid Adaptive Networks (HAN) model, where the Conditional Random Field (CRF) and Ridge classifier are adopted. Here, the parameter optimization is done using the Updated Random Attribute-based Circle-Inspired Optimization (URA-CIO) for improving the performance of the HAN model. This HAN-based node authentication verifies the identity of vehicles and RSUs participating in the VANET. After the node authentication is done, the second objective, optimal routing is executed. In order to perform effective routing, the optimal path is preferred based on the recommended URA-CIO approach. Thus, the optimal routing is made based on multi-objective constraints such as shortest path, residual energy, average end-to-end delay, path availability, throughput, and packet delivery ratio. The accuracy of the proposed URA-CIO-HAN-based node authentication and optimal routing task in VANET has achieved 92% accuracy, and also the conventional models, such as Bi-linear mapping, KNN, ELM, and DMN attained 84.73%, 85.61%, 88.42%, and 89.82% accuracy rate. Thus, the developed authentication-based optimal routing scheme allows the VANET to dynamically adjust to diverse conditions including security threats and network congestion and the optimal performance of the method is assessed using multiple metrics.