An Efficient Methodology to Identify Missing Tags in Large-Scale RFID Systems

Abstract: Radio frequency identification (RFID) has been widely has broad applications. One such application is to use RFID to track inventory in warehouses and retail stores. In this application, timely identifying the missing items is an ongoing engineering problem. A feasible solution to this problem is to map each tag to a time slot and verify the presence of a tag by comparing the status of the predicted time slot and the actual time slot. However, existing works are time inefficient because they only verify tags one by one in singleton slots but ignore the collision slots mapped by multiple tags. To accelerate the identification process, we use bit tracking to verify tags in collision slots and design two protocols accordingly. We first propose the Sequential String based Missing Tag Identification (SSMTI) protocol, which converts all time slots to collision slots and enables tags in each slot to reply to a designed string simultaneously. By using bit tracking to decode the combined string, the reader can verify multiple tags together. To improve the performance of SSMTI when most tags are missing, we further propose the Interactive String based Missing Tag Identification (ISMTI) protocol. ISMTI improves the strategies of designing strings for each collided tag so that the reader can verify more tags using shorter strings than SSMTI.Besides, ISMTI can dynamically adjust the verification mechanism according to the proportion of missing tags to maintain time efficiency. We also provide theoretical analysis for proposed protocols to minimize execution time and evaluate their performance through extensive simulations. Compared with state-of-the-art solutions, the proposed SSMTI and ISMTI can reduce the time cost by as much as 39.74% and 68.87%.