Scholarlink

<p>The broad adoption of the Internet of Things (IoT) in our daily lives requires focusing on their security. However, traditional security mechanisms are ineffective due to the constrained nature of IoT devices in terms of memory, processing power, and power consumption. Intrusion detection systems are critical to ensure the security of IoT environments against cyber threats. This study introduces a hybrid metaheuristic and deep learning approach to enhance anomaly-based intrusion detection in IoT environments. The Butterfly Optimization Algorithm is used for feature selection, effectively reducing data dimensionality while retaining essential information for classification. The selected features are then utilized to train a long Short-Term Memory network. This method improves the detection accuracy of known and unknown attacks, providing a robust solution for intrusion detection in IoT systems. The proposed model is evaluated using the IoTID20 and TON_IoT datasets, achieving 98.87 and 99.32% accuracy, respectively, and it increased Intrusion Detection System effectiveness against IoT cybersecurity attacks.</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

HyMD2I: Hybrid Metaheuristic-Deep Learning Approach for Intrusion Detection in IoT

  • 期刊论文

  • 发表日期：

    2025年6月20日

查看全文
  • Oumeima Boubertakh,
  • Ali Sahnoun,
  • Abdelhafid Zitouni,
  • Saad Harous

摘要

The broad adoption of the Internet of Things (IoT) in our daily lives requires focusing on their security. However, traditional security mechanisms are ineffective due to the constrained nature of IoT devices in terms of memory, processing power, and power consumption. Intrusion detection systems are critical to ensure the security of IoT environments against cyber threats. This study introduces a hybrid metaheuristic and deep learning approach to enhance anomaly-based intrusion detection in IoT environments. The Butterfly Optimization Algorithm is used for feature selection, effectively reducing data dimensionality while retaining essential information for classification. The selected features are then utilized to train a long Short-Term Memory network. This method improves the detection accuracy of known and unknown attacks, providing a robust solution for intrusion detection in IoT systems. The proposed model is evaluated using the IoTID20 and TON_IoT datasets, achieving 98.87 and 99.32% accuracy, respectively, and it increased Intrusion Detection System effectiveness against IoT cybersecurity attacks.