How RFID Attacks Are Expressed In OUtput Data

Attacks at the “RFID layer”, where tags and readers interact, appear in the output data used by the “strategic layer” to monitor or authorize entities. Previous research has used contextual information from the “real world layer” or “strategic layer” for attack detection purposes; this means that detection methods need to be customized to the application environment. In contrast, we introduce the concept of an “RFID Layer Context Model” to contextualize output data at the RFID layer. The improvement is that context at this layer is divorced from the application environment, and thus, attack detection does not need to be customized to the application. As the features we use are generic, this work takes the first steps towards automating the detection of attacks at the RFID layer.