Privacy-Preserving Data Preprocessing for Fog Computing in 5G Network Security

In SG wireless networks, the highly growing concern of data privacy from end users drives security and privacy fundamental and strong requirements for information services. End users will always expect and constantly demand efficient and effective privacy-preserving based security services in SG communications. Those services should not only adjust the levels of security protection, but also optimize the entire secure data communication process from the users to an untrusted cloud, via multiple fog nodes. In this new paradigm, a list of options should be open at the side of fog nodes, so that they can adjust the desired level of enhanced security protection for users' data intelligently and dynamically. In this way, the load of computational overhead for enhanced security protection at the user side will be greatly reduced. The requested options of those security services will be provided based on learning the contributed attributes, and further be enforced by fog nodes, where the data will be subsequently adjusted in a suitable way. The idea of Quality of Protection (QoP) can be applied at the fog nodes in SG networks, so that fog nodes can supply different levels of security protection to different data protection demands. In this paper, we propose a privacy-preserving data preprocessing scheme for fog computing in SG network security. Specifically, this work is conducted in the perspective of QoP, aiming to preserve the security service option learned from attributes and to enable fog nodes to supply different levels of privacy protection services with different security demands from users.