Development of an intelligent intrusion detection system for IoT networks using deep learning

Abstract

The Internet of Things (IoT) enables transformative applications; however, it faces significant security and privacy challenges. Due to the impracticality of securing individual devices, network-level security is preferred. However, attack diversity, device heterogeneity, and traditional security limitations necessitate advanced data analysis. Researchers increasingly use deep learning, which excels in handling large-scale data, to develop robust intrusion detection systems. This study investigates the security challenges of IoT, reviews existing intrusion detection systems, and explores the application of deep learning for intrusion detection in IoT networks. A total of 21 neural network models, along with an ensemble learning model, were designed and trained using the BoT-IoT dataset. Performance was evaluated using standard classification metrics including accuracy, precision, recall, and F1-score. The ensemble learning model demonstrated the highest performance, achieving 99.985% accuracy, 99.28% precision, 99.80% recall, and 99.54% F1-score. Compared to previous studies, the proposed model improves accuracy, recall, and F1-score by 0.005%, 0.55%, and 0.84%, respectively. These results highlight the effectiveness of deep learning-based approaches in enhancing IoT network security.