Scholarlink

<p>As a new kind of intelligent vehicle, Connected and Automated vehicles (CAVs) can provide more convenient service by exchanging data with other vehicles and roadside units. Despite all kinds of foreseeable advantages, CAVs rely heavily on the sensor data and communicated information, and negative factors such as faults, errors, or network attacks may lead to serious consequences. To avoid the potential risks, a multi-scale wavelet Transformer-based anomaly detection model is proposed, performing wavelet decomposition and reconstruction on time-series data to learn raw data features at different scales for CAVs, as well as to capture the dependencies between the decomposed frequency components, and finally achieving data anomaly detection effectively. The experimental results show that compared with other advanced baseline methods, our new model can perform better in improving the accuracy and sensitivity of detection.</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 New Time-Series Anomaly Detection Model for Connected Autonomous Vehicles Based on Multi-scale Wavelet Transformer Network

  • 期刊论文

  • 发表日期：

    2025年8月1日

查看全文
  • Jian Yin,
  • Zhenjiang Zhang,
  • Jianjun Zeng,
  • Ziang Zhang

摘要

As a new kind of intelligent vehicle, Connected and Automated vehicles (CAVs) can provide more convenient service by exchanging data with other vehicles and roadside units. Despite all kinds of foreseeable advantages, CAVs rely heavily on the sensor data and communicated information, and negative factors such as faults, errors, or network attacks may lead to serious consequences. To avoid the potential risks, a multi-scale wavelet Transformer-based anomaly detection model is proposed, performing wavelet decomposition and reconstruction on time-series data to learn raw data features at different scales for CAVs, as well as to capture the dependencies between the decomposed frequency components, and finally achieving data anomaly detection effectively. The experimental results show that compared with other advanced baseline methods, our new model can perform better in improving the accuracy and sensitivity of detection.