Scholarlink

<p>The growing significance of vehicular data collection and monitoring demands robust mechanisms to ensure data security, enable real-time performance tracking, and extract valuable insights. This paper proposes a blockchain-based framework integrating secure transmission, immutable storage, and advanced analysis of vehicle Controller Area Network (CAN) data. The framework acquires CAN data from vehicles, records it through blockchain transactions to guarantee immutability and transparency, and visualizes insights via an interactive dashboard for vehicle owners and researchers. Advanced machine learning models—Random Forest and Decision Trees for predictive analytics, alongside k-means clustering for uncovering hidden data patterns—are employed for in-depth analysis. Evaluation results demonstrate a blockchain transaction efficiency rate of 98% and predictive model accuracies exceeding 92%, validating the framework’s capability to enhance security and optimize vehicle performance. These findings highlight the framework’s potential to advance intelligent transportation systems and pave the way for future research on secure and intelligent vehicular networks.</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

Blockchain based machine learning approach for secure and efficient vehicular data monitoring and analysis

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年11月7日

查看全文
  • R. Hanumantharaju,
  • K. N. Shreenath,
  • B. J. Sowmya,
  • S. Supreeth,
  • G. Shruthi,
  • S. Rohith

摘要

The growing significance of vehicular data collection and monitoring demands robust mechanisms to ensure data security, enable real-time performance tracking, and extract valuable insights. This paper proposes a blockchain-based framework integrating secure transmission, immutable storage, and advanced analysis of vehicle Controller Area Network (CAN) data. The framework acquires CAN data from vehicles, records it through blockchain transactions to guarantee immutability and transparency, and visualizes insights via an interactive dashboard for vehicle owners and researchers. Advanced machine learning models—Random Forest and Decision Trees for predictive analytics, alongside k-means clustering for uncovering hidden data patterns—are employed for in-depth analysis. Evaluation results demonstrate a blockchain transaction efficiency rate of 98% and predictive model accuracies exceeding 92%, validating the framework’s capability to enhance security and optimize vehicle performance. These findings highlight the framework’s potential to advance intelligent transportation systems and pave the way for future research on secure and intelligent vehicular networks.