Scholarlink

<p>In the rapidly growing field of smart cities and intelligent transportation systems (ITS), Vehicle Ad-hoc Networks (VANETs) play a key role in enabling seamless interaction between vehicles, infrastructure, and people. To support this, stable and secure clustering methods are essential for ensuring reliable communication within VANETs. However, maintaining stable and secure connectivity in such dynamic networks remains a significant challenge. This paper presents a stable and secure clustering methodology, DNN-CCAS (Deep Neural Network–Canonical Correlation Analysis Scheme), designed to enhance both cluster stability and security in VANET environments. The proposed approach involves three key phases: cluster formation using an improved K-consonance technique, cluster head selection based on a linear metric, and secure data transmission validated by a deep learning model. Simulation experiments conducted using MATLAB, SUMO, and OMNeT++ indicate that the proposed approach attains an accuracy of 92.42%. Furthermore, it consistently surpasses existing methods with respect to clustering efficiency and communication security. These findings indicate the potential of DNN-CCAS to support robust and reliable vehicular communications in future smart transportation systems.</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

A Stable and Secure Clustering Methodology for Vehicles in VANET

  • 期刊论文

  • 发表日期：

    2025年10月9日

查看全文
  • Atul Barve,
  • Pushpinder Singh Patheja

摘要

In the rapidly growing field of smart cities and intelligent transportation systems (ITS), Vehicle Ad-hoc Networks (VANETs) play a key role in enabling seamless interaction between vehicles, infrastructure, and people. To support this, stable and secure clustering methods are essential for ensuring reliable communication within VANETs. However, maintaining stable and secure connectivity in such dynamic networks remains a significant challenge. This paper presents a stable and secure clustering methodology, DNN-CCAS (Deep Neural Network–Canonical Correlation Analysis Scheme), designed to enhance both cluster stability and security in VANET environments. The proposed approach involves three key phases: cluster formation using an improved K-consonance technique, cluster head selection based on a linear metric, and secure data transmission validated by a deep learning model. Simulation experiments conducted using MATLAB, SUMO, and OMNeT++ indicate that the proposed approach attains an accuracy of 92.42%. Furthermore, it consistently surpasses existing methods with respect to clustering efficiency and communication security. These findings indicate the potential of DNN-CCAS to support robust and reliable vehicular communications in future smart transportation systems.