Scholarlink

<p>With the growth of the Internet of Things (IoT), new challenges emerge to secure systems while maintaining privacy and ensuring the user of the system is able to understand the threat. The proposed system, FLBCXAI, integrates federated learning, blockchain, and explainable AI (XAI) to address these challenges. The system functions in three layers; first is the unsupervised anomaly detection using a CNN-BiLSTM-Attention model, the second is secure aggregation of model updates using blockchain verified federated learning, and the third is explanation of the threat using attention and SHAP and the entire model. FLBCXAI aims to increase resilience of the system while maintaining privacy and transparency of sensitive secure information. The system is deployed in a healthcare environment primarily to address the urgent response systems through the distributed edge, fog, and cloud layers. The framework offers a foundational model in sustaining systems through private federated learning, blockchain consensus, and explainable decision making in critical healthcare systems.</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

An integrated framework of blockchain and federated learning with explainable AI for enhanced security of IoT healthcare systems

  • 期刊论文

  • 发表日期：

    2025年12月2日

查看全文
  • Kedar Nath Singh,
  • Anuj Kumar Singh

摘要

With the growth of the Internet of Things (IoT), new challenges emerge to secure systems while maintaining privacy and ensuring the user of the system is able to understand the threat. The proposed system, FLBCXAI, integrates federated learning, blockchain, and explainable AI (XAI) to address these challenges. The system functions in three layers; first is the unsupervised anomaly detection using a CNN-BiLSTM-Attention model, the second is secure aggregation of model updates using blockchain verified federated learning, and the third is explanation of the threat using attention and SHAP and the entire model. FLBCXAI aims to increase resilience of the system while maintaining privacy and transparency of sensitive secure information. The system is deployed in a healthcare environment primarily to address the urgent response systems through the distributed edge, fog, and cloud layers. The framework offers a foundational model in sustaining systems through private federated learning, blockchain consensus, and explainable decision making in critical healthcare systems.