Design and Analysis of Ethereum Blockchain Enabled IoT Based Model for Secure Data Transmission

Abstract

Ensuring the security and privacy of sensitive health data in Internet of Things (IoT)-based healthcare systems (HCS) is a critical challenge. This paper proposes a robust security framework by integrating blockchain mechanisms and deep learning (DL) approaches to enhance security and data privacy. The proposed framework leverages the Ethereum blockchain with zero knowledge proof (ZKP) to ensure data integrity and confidentiality, while the interplanetary file system (IPFS) provides secure and efficient data storage. Additionally, a novel At-GAN-BiLSTM model is introduced for intrusion detection by combining the attention mechanism, generative adversarial networks (GAN) and bidirectional long short-term memory (Bi-LSTM) to improve detection accuracy and also help to enhance model robustness. The proposed model is evaluated by two different benchmark datasets, namely CICIDS-2018 (D1) and ToN-IoT (D2), achieving accuracies of 99.9% and 99.1%, respectively. Comparative investigation shows that the proposed approach reduces false alarm rates (FAR) and performs better than current models in identifying impersonation, insider, and man-in-the-middle (MITM) attacks. By integrating blockchain and DL, the proposed framework significantly enhances intrusion detection, data security, and overall system resilience, addressing key vulnerabilities in IoT-based healthcare security.