Smart Contract-Enhanced Residual GRU with Merkle-Damgard Cryptography for IoT Attack Detection

Abstract

As IoT continues to expand, the security of connected devices remains a critical concern, particularly in the face of DDoS attacks. This study introduces a novel approach that leverages blockchain technology through smart contracts integrated with an advanced attack detection mechanism. Central to this approach is the Enhanced Residual Gated Recurrent Unit (ERGRU) architecture, designed to effectively identify and mitigate DDoS attacks within IoT networks. The Adaptive Coati Optimization Algorithm (ACOA) was used to adjust the hyperparameters of the ERGRU model, such as the learning rate and the number of GRU neurons, to further improve detection accuracy. In addition, the proposed framework uses a one-way compression function to generate secure hashes for input data, utilizing the Merkle-Damgård cryptography technique to ensure data integrity and confidentiality. The proposed solution was tested through a rigorous process using a DDoS dataset. Performance was assessed by focusing on metrics such as processing time, data integrity rate, and confidentially rate. The results demonstrate the effectiveness of the proposed smart contract-based framework in providing a durable and efficient protection mechanism against DDoS attacks in IoT environments.