SHFL: K-Anonymity-Based Secure Hierarchical Federated Learning Framework for Smart Healthcare Systems

Dynamic and smart Internet of Things (IoT) infrastructures allow the development healthcare systems, which are equipped with mobile health embedded sensors to enable a broad range applications. These IoT applications provide access clients’ information. However, rapid increase in number devices social networks has generated concerns regarding secure sharing client’s location. In this regard, federated learning (FL) is an emerging paradigm decentralized machine that guarantees training shared global model without compromising data privacy client. To end, we propose K-anonymity-based hierarchical (SHFL) framework for systems. proposed FL approach, centralized server communicates hierarchically multiple directly indirectly connected devices. particular, SHFL formulates clusters location-based services achieve distributed FL. addition, utilizes K-anonymity method hide location cluster Finally, evaluated performance by configuring different architectures datasets. The experiments validated provides adequate generalization network scalability accurate systems privacy.