Browsing by Author "Wafa Bouras"

Now showing 1 - 1 of 1
  • Thumbnail Image
    ItemOpen Access
    IoMT-Enabled 5G for Patient Identification A Resilient Blockchain-Based Federated Learning Framework for Optimized and Secure Participant Selection
    (University of M'Sila, 2025) Wafa Bouras
    The Internet of Medical Things (IoMT) has revolutionized modern healthcare by enabling continuous monitoring and real-time data exchange among medical devices. However, the heterogeneity of data sources, limited computational resources, and increasing security threats pose significant challenges to the deployment of intelligent and privacy-preserving solutions. This thesis proposes an enhanced framework that integrates Federated Learning (FL) with lightweight blockchain consensus mechanisms to address key issues in participant selection and system robustness. A comparative study of existing participant selection methods is presented, followed by the design of a refined probabilistic model that balances optimization and privacy. Furthermore, we introduce a blockchain-assisted role assignment mechanism to improve transparency and trust among distributed participants. The proposed framework, BlockGuard-RD, is evaluated against various threat scenarios such as data poisoning, impersonation, and denial-of-service (DoS) attacks. Experimental results demonstrate the framework’s ability to enhance model accuracy, improve resource efficiency, and maintain high security standards within IoMT environments. Ultimately, this work contributes a robust and adaptive solution for secure, scalable, and privacy-aware machine learning in medical cyber-physical systems