Geometrical approach to multi-phase RFID filtering in dense environments

Abstract

In the upcoming ubiquitous computing environment, almost every object that needs to be identified will be attached to an radio frequency identification (RFID) tag. Omnipresent RFID tags lead to the concern that huge volumes of RFID data could easily overwhelm related networks and systems. However, for most applications, only a small subset of ID data is meaningful. It is critical to filter out the unnecessary ID data based on application requirements. A multi-phase RFID filtering technique that basically consists of pre-filtering and post-filtering phases is presented. To the best of our knowledge, most RFID filtering works have focused only on single-phase post-filtering. Our multi-phase filtering technique harmonises pre-filtering and post-filtering in order to raise the RFID system filtering efficiency. The main challenge for this type of multi-phase filtering technique is to derive air-protocol filters for pre-filtering from read filters for post-filtering that are guaranteed to cover the union of all read filters. The authors introduce geometrical algorithms for deriving air-protocol filters from read filters and verify their performance via simulations using analytical time models. The results show that our multi-phase filtering technique reduces the overall filtering time by almost 50% compared to the single-phase post-filtering technique.