A Privacy-Level Model of User-Centric Cyber-Physical Systems

In an interconnected cyber-world, Cyber-Physical Systems (CPSs) appear to play an increasingly important role in smart ecosystems. A variety of resource-constrained thin clients, such as sensors, RFIDs, actuators and smart devices, are included in the list of CPS. These devices can be used in a number of medical, vehicular, aviation, military and smart cities applications. A plethora of sensitive data is transmitted in insecure wireless or wired environments whilst adversaries are eager to eavesdrop, modify or destroy sensed data invading the privacy of user-centric CPSs. This work presents an overview and analysis of the most effective attacks, privacy challenges and mitigation techniques for preserving the privacy of users and their interconnected devices. In order to preserve privacy, a privacy-level model is proposed in which users have the capability of assigning different privacy levels based on the variety and severity of privacy challenges and devices’ capabilities. Finally, we evaluate the performance of specific CPSs at different privacy-levels in terms of time and consumed energy in an experimental test-bed that we have developed.