Scholarlink

<p>With the emergence of quantum computing, traditional cryptographic methods face significant vulnerabilities, particularly in the Industrial Internet of Things (IIoT) environment, where cybersecurity, data privacy, and trust management are critical. In order to address the said problems concerned with post-quantum security, this paper proposes a quantum-secure framework for IIoT applications by integrating multi-agent reinforcement learning (MARL) for optimized blockchain consensus, blockchain-supported federated learning for privacy-preserving AI training, and post-quantum cryptography (PQC) for quantum-resistant security. The MARL-based consensus uses a deep Q-network with explainable AI to enhance transparency and trust. Federated learning employs dynamic participant selection and PQC (e.g., NTRUEncrypt, Kyber) to ensure privacy. Evaluations are carried out on Edge IIoT and Federated EMNIST datasets, which are publicly available datasets and are used in similar studies. The experimental results reveal that the proposed framework demonstrate improvements in scalability, security, and efficiency, hence, offering a robust solution for applications like smart cities and healthcare.</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 MARL-federated blockchain-based quantum secure framework for trust management in industrial internet of things

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年11月7日

查看全文
  • Gasim Alandjani

摘要

With the emergence of quantum computing, traditional cryptographic methods face significant vulnerabilities, particularly in the Industrial Internet of Things (IIoT) environment, where cybersecurity, data privacy, and trust management are critical. In order to address the said problems concerned with post-quantum security, this paper proposes a quantum-secure framework for IIoT applications by integrating multi-agent reinforcement learning (MARL) for optimized blockchain consensus, blockchain-supported federated learning for privacy-preserving AI training, and post-quantum cryptography (PQC) for quantum-resistant security. The MARL-based consensus uses a deep Q-network with explainable AI to enhance transparency and trust. Federated learning employs dynamic participant selection and PQC (e.g., NTRUEncrypt, Kyber) to ensure privacy. Evaluations are carried out on Edge IIoT and Federated EMNIST datasets, which are publicly available datasets and are used in similar studies. The experimental results reveal that the proposed framework demonstrate improvements in scalability, security, and efficiency, hence, offering a robust solution for applications like smart cities and healthcare.