Energy-efficient distributed federated learning offloading and scheduling healthcare system in blockchain based networks

Abstract

Many disease detection and prevention applications in digital healthcare systems are widely used but often focus only on prediction and classification, ignoring processing performance and data privacy issues. The study investigates the Energy-Efficient Distributed Federated Learning Offloading and Scheduling Healthcare Systems in Blockchain-Based Networks problem for healthcare applications. In order to solve the problem, the study presents the Energy-Efficient Distributed Federated Learning Offloading and Scheduling (EDFOS) system in blockchain based networks. EDFOS consisted of different schemes such as energy efficient offloading and scheduling and meet the quality of services (QoS) of applications during performing in the system. Simulation results show that EDFOS reduces power consumption by 39%, training and testing time by 29%, and resource leakage and deadlines by 36% compared to existing healthcare systems. The EDFOS platform is an effective solution for addressing the issues of power consumption and data privacy in healthcare applications.