Hybrid Intrusion Detection System for RPL IoT Networks Using Machine Learning and Deep Learning

The Internet of Things (IoT) is transforming everyday objects. However, its devices’ limited memory, processing power, and network capabilities make them susceptible to security breaches. The Routing Protocol for Low-Power and Lossy Networks (RPL) is a promising IoT protocol but faces significant security challenges. Existing research often focuses on individual attacks, utilizing various mitigation strategies, including machine learning and deep learning for detection. This paper proposes an Intrusion Detection System (IDS) using the ROUT-4-2023 dataset, which encompasses Black Hole, Flooding, DODAG Version Number, and Decreased Rank attacks. The study utilizes statistical information graphs to investigate network traffic features encompassing all four attacks. Additionally, it experiments with various machine learning models and deep learning architectures for comparative analysis, focusing on confusion matrix outcomes and computational efficiency. Results indicate that the Random Forest classifier achieves 99% accuracy, while Transformers reach 97% F1-Score with a training time of only 16.8 minutes over five epochs.