Securing Cyber-Physical Spaces with Hybrid Analytics: Vision and Reference Architecture

Abstract

Considering the massive increase in the number of crimes in the last decade, as well as the outlook toward smarter cities and more sustainable urban living, the emerging cyber-physical space (CPS) obtained by the interaction of such physical spaces with the cyber elements around them (e.g., think of Internet-of-Things devices or hyperconnected mobility), plays a key role in the protection of urban social living, e.g., social events or daily routines. For example, the hyperconnectedness of a CPS to many networks can lead to potential vulnerability. This vision paper aims to outline a vision and reference architecture where CPS protection is center-stage and where CPS models as well as so-called hybrid analytics work jointly to help the Law Enforcement Agents (LEAs), e.g., in event monitoring and early detection of criticalities. As a part of validating said reference architecture, we implement a case study in the scope of VISOR, a Dutch government project aimed at improving CPS protection using hybrid analytics. We conduct a field experiment in the Paaspop social event and festival grounds to test and select the most appropriate device configuration. There we experiment with a CPS protection pipeline featuring several components reflected in the reference architecture, e.g., the KGen middleware, a prototype tool to anonymize structured big data using genetic algorithms, and SENSEI, a framework for dark web marketplace analytics. We conclude that hybrid analytics offer a considerable ground for more sustainable CPS.