Knowledge-based Cyber Physical Security at Smart Home: A Review
ACM Computing Surveys, Volume 57, Issue 3, Article 53, November 2024, pp. 1–36.
Smart home environments comprise heterogeneous devices, each with a different set of capabilities in terms of sensing, actuation, and communication. This diversity, combined with resource constraints and evolving contextual conditions, makes securing such environments a significant challenge. This research develops self-adaptive security techniques for heterogeneous smart home systems using edge computing rather than cloud-dependent services. The project employs the MAPE-K (Monitor-Analyse-Plan-Execute-Knowledge) methodology to dynamically configure smart spaces while addressing device diversity, resource limitations, and contextual changes. Device capabilities, environmental context, and security policies are captured in a knowledge base that drives the MAPE-K loop, enabling the system to plan and execute protective actions locally on edge nodes. By combining rule engines and AI planners with open-source frameworks such as Drools, OpenHAB, and OptaPlanner, the system adapts to new conditions while keeping sensitive data on the edge.
The research objectives include conducting a comprehensive literature review on knowledge-based techniques for smart home cyber-physical security, developing knowledge models that support the MAPE-K loop, and implementing and evaluating these techniques using open-source tools. The project also creates demonstrators that showcase knowledge-based self-adaptive security systems operating in realistic smart home configurations.
The project contributes towards building transparent, locally governed security mechanisms for the next generation of connected domestic environments. By grounding security decisions in explicit knowledge representations rather than opaque machine learning models, the approach supports auditability and user trust in automated home protection systems.