Trust-Driven Blockchain-Enabled Deep Learning Architecture for Secure and Scalable IIoT Networks: Implications for Industrial Adoption and Market Differentiation

Abstract

Nowadays, the Industrial Internet of Things (IIoT) has transformed the fields of smart manufacturing and industrial automation by enabling continuous data exchange between interconnected devices. However, the trustworthiness, reliability, security, and scalability of data transmission across distributed IIoT networks remain key issues due to vulnerabilities in traditional centralized architectures. The present study proposes a novel Stimulated and Secure Blockchain-Based Learning System (SSBLS) to facilitate robust, decentralized, and verifiable data transfer within IIoT ecosystems while enhancing trust and data integrity. The proposed framework integrates a hybrid blockchain infrastructure employing Elliptic Curve Cryptography (ECC) for lightweight encryption and a dual-consensus mechanism based on Delegated Proof of Stake (DPoS) and Practical Byzantine Fault Tolerance (PBFT) for achieving consensus with minimal latency and energy consumption in resource-constrained IIoT environments. To detect anomalous data patterns and ensure real-time threat mitigation, a Convolutional Long Short-Term Memory (CNN-LSTM) model is deployed, enabling the learning of spatial and temporal dependencies from streaming IIoT sensor data. Furthermore, to enhance model convergence and accuracy under resource constraints, the Butterfly Optimization Algorithm (BOA) is applied for hyperparameter tuning, thereby improving computational efficiency and detection reliability. The system was evaluated using a publicly available industrial IIoT dataset simulating sensor-driven manufacturing environments. Quantitative results indicate that the proposed SSBLS framework achieves a Mean Absolute Error (MAE) of 0.041, Root Mean Square Error (RMSE) of 0.067, and R² score of 0.983, demonstrating high prediction fidelity and system robustness. Additionally, the blockchain mechanism ensures zero data tampering incidents. It achieves a throughput improvement of 23.5% as compared to traditional client-server architectures, thus validating its potential for trusted and scalable IIoT deployments. Furthermore, the trust-centric architecture not only positions IIoT vendors in compliance with industry standards but also distinguishes them in an increasingly security-conscious market.