A Collaborative IDS for IoT Environments Using Machine Learning and Blockchain
Loading...
Date
2025
Journal Title
Journal ISSN
Volume Title
Publisher
Faculty of Sciences
Abstract
The rapid expansion of the Internet of Things (IoT) has revolutionized industrial environments, offering unprecedented connectivity and automation, yet it also introduces significant security challenges due to the heterogeneous nature and resource constraints of IoT devices. This research presents a decentralized, edge-centric Intrusion Detection System (IDS) tailored for Industry 4.0 settings, integrating machine learning-based anomaly detection with lightweight blockchain technology to ensure secure, real-time monitoring and alert propagation. The system comprises layered components, including strong devices for processing and detection, edge gateways for coordination and blockchain interaction, and weak devices with minimal computational capabilities. Various attack scenarios are simulated using the CICIDS2017 dataset, with machine learning algorithms such as Random Forest and XGBoost employed for classification, alongside hyperparameter optimization using the Grey Wolf Optimizer (GWO). Alerts are securely logged and shared via a custom blockchain implemented using SQLite, maintaining data integrity and traceability. Experimental results demonstrate high detection accuracy and efficient response coordination, validating the system’s scalability, lightweight design, and effectiveness in securing IoT infrastructures within smart industrial environments.