Blockchain-Integrated Framework for Data Security: An Application Based on IoT Data and Deep Learning

Abstract

The rapid development of the IoT (Internet of Things) ecosystem leads to the creation of big data environments that require real-time analysis. In this comprehensive data ecosystem, anomaly detection and data security emerge as critical requirements. This paper presents a comprehensive approach that integrates a deep learning model developed for anomaly detection in IoT network traffic and a blockchain-based data storage structure designed to ensure data integrity. In the research, network traffic data of a sample device from the N-BaIoT dataset is used. The developed deep learning model was able to classify attack and normal traffic patterns with high accuracy. Data security is ensured with Fernet encryption algorithm, while data integrity is protected using blockchain technology. Experimental results show that the proposed system achieves significant performance metrics in terms of both anomaly detection accuracy and data security verification. The proposed framework contributes to the development of more secure and reliable IoT systems by providing an innovative solution to anomaly detection and data security challenges in IoT environments.