| Summary: | Ensuring robust security in the Internet of Things (IoT) landscape is of paramount importance. This research article presents a novel approach to enhance IoT security by leveraging collaborative threat intelligence and integrating blockchain technology with machine learning (ML) models. The iOS application acts as a central control centre, facilitating the reporting and sharing of detected threats. The shared threat data is securely stored on a blockchain network, enabling ML models to access and learn from a diverse range of threat scenarios. The research focuses on implementing Random Forest, Decision Tree classifier, Ensemble, LSTM, and CNN models on the IoT23 dataset within the context of a Collaborative Threat Intelligence Framework for IoT Security. Through an iterative process, the models’ accuracy is improved by reducing false negatives through the collaborative threat intelligence system. The article investigates the implementation details, privacy considerations, and the seamless integration of ML-based techniques for continuous model improvement. Experimental evaluations on the IoT23 dataset demonstrate the effectiveness of the proposed system in enhancing IoT security and mitigating potential threats. The research contributes to the advancement of collaborative threat intelligence and blockchain technology in the context of IoT security, paving the way for more secure and reliable IoT deployments.
|
|---|