Scholarlink

The Internet of Things (IoT) enriches daily life with connectivity but adds major cybersecurity concerns, including Distributed Denial of Service (DDoS) attacks endangering data security and privacy. In this research, this work introduces a new approach for DDoS attack detection in IoT networks based on the combination of the Learning Vector Quantization (LVQ) neural network and the Gray Wolf Optimization (GWO) algorithm. The NSL-KDD dataset was preprocessed, and GWO was used to optimize the most important parameters of the LVQ network, including the size of the hidden layer. Effective searching and utilization of the GWO algorithm provided efficient parameter tuning and avoidance of local optima. The proposed model demonstrated a remarkable accuracy of 96.3% in detecting DDoS attacks, demonstrating its capability to improve IoT system cybersecurity. The results demonstrate the effectiveness of the method and its usability for reducing IoT-based cyberattacks.

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

Detection of DDoS Attacks in IOT Using Advanced Threat Detection Algorithms

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Karrar Ali Hussein,
  • Yasir Ali Khalid,
  • Abbas Abd Alhussein Haddad

摘要

The Internet of Things (IoT) enriches daily life with connectivity but adds major cybersecurity concerns, including Distributed Denial of Service (DDoS) attacks endangering data security and privacy. In this research, this work introduces a new approach for DDoS attack detection in IoT networks based on the combination of the Learning Vector Quantization (LVQ) neural network and the Gray Wolf Optimization (GWO) algorithm. The NSL-KDD dataset was preprocessed, and GWO was used to optimize the most important parameters of the LVQ network, including the size of the hidden layer. Effective searching and utilization of the GWO algorithm provided efficient parameter tuning and avoidance of local optima. The proposed model demonstrated a remarkable accuracy of 96.3% in detecting DDoS attacks, demonstrating its capability to improve IoT system cybersecurity. The results demonstrate the effectiveness of the method and its usability for reducing IoT-based cyberattacks.