Secured Blockchain With Adaptive Bi-GRU-Based Authentication and Optimal Key-Based Fully Homomorphic Encryption Framework in Digital Twin Environment

Abstract

The business and industry significantly utilize digital twin technology, so it gained lots of attention in recent years; data produced from actual resources are sent to a distant server in digital twin environments, which use digital twins in a virtual setting to run simulations. However, using digital twin technology in real life poses several difficulties. Finding a way to safely communicate the simulation's real-time data and data sharing is one of the most significant challenges. Sensitive information pertaining to data owners may be incorporated into the data produced by physical assets, and data leak to enemies might result in major privacy issues. To optimize the accessibility of digital twin data, data sharing with other data users must additionally be taken into account. To resolve these issues, an efficient framework is introduced in this research work with deep learning and encryption standards. Initially, a novel deep learning-based blockchain authentication scheme named Adaptive Bidirectional Gated Recurrent Unit (Ada-Bi-GRU) is implemented in the digital twin environment. The Updated Random Parameter-Aided Hippopotamus Optimization Algorithm (URP-HOA) tunes the attributes of the Ada-Bi-GRU. Hence, the security against different attacks is improved by the proposed Ada-Bi-GRU. Next, security issues in the blockchain authentication scheme are resolved by considering a novel privacy preservation technique in the digital twin environment. In this privacy preservation phase, the developed framework employed the Optimal Key-Based Fully Homomorphic Encryption (OK-FHE) mechanism. In this, the keys are generated optimally utilizing the enhanced HOA technique. Later, the effectiveness of the developed model is validated by comparison.