Scholarlink

<p>IoT devices are more and more used in real life. Especially, many companies and households are also gradually using IoT devices in large numbers to realize the goals of smart homes, smart schools, offices, and so on. However, the security problems of these IoT systems and devices are not focused on so much, leading to many security holes in the data system, creating opportunities for hackers to attack the system with many kinds of attacks. In this paper, we study and build a deep learning model to classify and detect the IoT devices that are not in the system’s pre-registered white list, the detection of strange IoT devices intruding into the IoT network system. Our proposed method helps to prevent the risk of system failure based on detecting strange and unauthorized IoT devices trying to join our IoT network. We implement and use a publicly available dataset to evaluate our proposed method. According to the experimental results, our method achieves very good results when we detect ten IoT devices with up to 99.78% accuracy.</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

Unauthorized iot devices detection based on network traffic using augmentation fusion classification method

  • 期刊论文

  • 发表日期：

    2025年10月22日

查看全文
  • Duong Van Dan,
  • Ta Minh Thanh

摘要

IoT devices are more and more used in real life. Especially, many companies and households are also gradually using IoT devices in large numbers to realize the goals of smart homes, smart schools, offices, and so on. However, the security problems of these IoT systems and devices are not focused on so much, leading to many security holes in the data system, creating opportunities for hackers to attack the system with many kinds of attacks. In this paper, we study and build a deep learning model to classify and detect the IoT devices that are not in the system’s pre-registered white list, the detection of strange IoT devices intruding into the IoT network system. Our proposed method helps to prevent the risk of system failure based on detecting strange and unauthorized IoT devices trying to join our IoT network. We implement and use a publicly available dataset to evaluate our proposed method. According to the experimental results, our method achieves very good results when we detect ten IoT devices with up to 99.78% accuracy.