A Privacy Preserving Federated Learning BasedIoT Framework Using Cloud Computing

Abstract

This abstract explores the transformative impact of IoT on modern life, empha- sizing the integration of Federated Learning (FL), Edge Computing, and Secure Offloading in AI applications. The rapid evolution of IoT has revolutionized com- mercial operations and consumer interactions, driven by advanced sensing and computational capabilities in mobile devices. However, concerns over data privacy and limited computational resources hinder the deployment of compute-intensive applications. FL emerges as a distributed AI paradigm, ensuring privacy and saving network resources. Edge computing optimizes service delivery, reducing latency and energy consumption, supported by intelligent offloading algorithms and blockchain technology for secure and efficient edge services. Challenges like slow learning speeds persist but are addressed through ongoing advancements in neural networks.The proposed framework is compared with the benchmark mod- els and it was observed that the proposed framework suppress the benchmark models.