A knowledge-based approach for detecting misuses in RFID systems

In this chapter we describe a methodology for applying intrusion detection techniques to identify RFID tag cloning which is primarily focused on the integration of the principles of ontology modeling and inferential reasoning in the intrusion detection paradigm. This methodology stands on three key points: i) the definition of an ontology model formalizing expected and actual profiles, each of them being based on location and tracking information about RFID tagged objects; ii) the application of an inferential engine that, exploiting inference patterns proper to the logic underlying the ontology formalism, checks the consistency between expected and actual profiles of tagged objects; iii) the detection and identification of anomalous conditions in presence of inconsistencies. We followed this methodology for implementing a Misuse Detection System (MDS) that operates at middleware layer and exhibits an active and event-dependent behavior in response to new tracking information in order to detect tag cloning.