On the Design of a Privacy Preserving Collaborative Platform for Cybersecurity

Abstract

Nowadays, cyber-attacks are targeting mobile devices, bank accounts, connected vehicles and cyber-physical systems. These attacks are becoming more complex and are raising safety problems when targeting physical environment. An efficient way to protect against these attacks is making several security actors collaborate in defining appropriate countermeasures. However, in practice, security actors refrain from collaborating to avoid sharing their proprietary security processes. These processes represent a critical knowledge as they reflect these actors brand images. In this work, we investigate the use of homomorphic encryption to define a privacy preserving framework for sharing processes between different cybersecurity actors and for providing confidential data analysis. We describe a high level design for a secure cloud platform managing encrypted data. The data analysis algorithms provided by the cloud platform are designed with our open source tool Cingulata, which enables designers to implement any data analysis function, compile it and run it on homomorphically encrypted data.