Scholarlink

<p>The Industrial Internet of Things (IIoT) introduces complex security challenges in key management due to its distributed architecture and resource-constrained nodes. This paper presents a novel blockchain-enabled key management framework designed to ensure decentralized, tamper-resistant, and scalable security across the IIoT network. The system combines Elliptic Curve Cryptography for easy key creation with smart contracts on a private Ethereum blockchain to allow safe key sharing and management throughout its lifecycle. We also present a way to verify the identity of nodes that takes trust into account, using a Graph Convolutional Network improved with the Ranger optimizer to make it work better and more reliably—something that hasn’t been done before in IIoT security systems. Tests using the X-IIoTID dataset, which includes different types of IIoT traffic during normal and attack situations like spoofing and injection, show that the system reaches 98.9% accuracy in authentication while keeping response times low and processing speeds high. The framework effectively mitigates Sybil and replay attacks, providing a scalable and secure solution that improves both trust and performance in distributed IIoT environments, reinforcing secure industrial data exchange.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A blockchain-enabled encrypted neural network framework for trust-aware key management and node authentication in Industrial Internet of Things

  • 期刊论文

  • 发表日期：

    2025年6月21日

查看全文
  • Abhishek Dwivedi,
  • Ratish Agarwal,
  • Mohammad Yahya,
  • Noha Alduaiji,
  • Piyush Kumar Shukla

摘要

The Industrial Internet of Things (IIoT) introduces complex security challenges in key management due to its distributed architecture and resource-constrained nodes. This paper presents a novel blockchain-enabled key management framework designed to ensure decentralized, tamper-resistant, and scalable security across the IIoT network. The system combines Elliptic Curve Cryptography for easy key creation with smart contracts on a private Ethereum blockchain to allow safe key sharing and management throughout its lifecycle. We also present a way to verify the identity of nodes that takes trust into account, using a Graph Convolutional Network improved with the Ranger optimizer to make it work better and more reliably—something that hasn’t been done before in IIoT security systems. Tests using the X-IIoTID dataset, which includes different types of IIoT traffic during normal and attack situations like spoofing and injection, show that the system reaches 98.9% accuracy in authentication while keeping response times low and processing speeds high. The framework effectively mitigates Sybil and replay attacks, providing a scalable and secure solution that improves both trust and performance in distributed IIoT environments, reinforcing secure industrial data exchange.