On Fast and Reliable Missing Event Detection Protocol for Multitagged RFID Systems

Abstract

With the rapid development of radio-frequency identification (RFID) technology, the ever-increasing research effort has been dedicated to devising various RFID-enabled services. The missing event detection, the functionality of detecting missing objects, is one of the most important services in many Internet-of-Things applications such as inventory management. Prior detection protocols only work in single-tagged RFID systems and would waste much time on repeated checks on one object in the emerging multitagged systems where each object is attached by multiple tags, leaving efficient detection in the new scenario unaddressed. To bridge the gap, this article is devoted to detecting missing multitagged objects. The key technicality is to build a filter from a subset of tags instead of whole in prior works to avoid repeated detections of one object and reduce detection time. Specifically, we first provide a basic solution based on the Bloom filter which can specify only tags in the chosen subset to participate in the final detection. To further improve time efficiency, we propose an advanced protocol that exploits tag ID knowledge and sparsity of slots mapped by only tags in the chosen subset to build a more compact compressive filter. Moreover, a composite vector is used to efficiently coordinate tags to report its presence. We conduct theoretical analysis on optimum protocol parameters and extensive simulations to verify the feasibility of the protocols. The results show that the advanced protocol achieves more than $2\times $ performance gain in terms of time efficiency over the Bloom filter-based basic protocol.