Attacks classification and data privacy protection in cloud-edge collaborative computing systems

Abstract

Collaborative computing systems used today are a significant problem, and the variety of attacks and data privacy protection play an important role. Accordingly, this research takes the latest technology, such as federated learning and cloud-edge collaborative computing systems. Among them, multi-national validation with attacks/without attacks architecture is mainly developed, and 'End-to-end privacy-preserving deep learning for attack classification method is used to classify each episode that occurs and done through End-to-end privacy-preserving deep understanding (E2EPPDL). This is the most essential core component of our research. We justified them with Time, Node Count, Routing count, and Data delivery ratio estimates.