Exploring Tag Distribution in Multi-Reader RFID Systems
Feng Zhu,Bin Xiao,Jia Liu,Bin Wang,Qingfeng Pan,Li-Jun Chen
DOI: https://doi.org/10.1109/tmc.2016.2591950
IF: 6.075
2017-01-01
IEEE Transactions on Mobile Computing
Abstract:Radio Frequency Identification (RFID) brings a revolutionary change in a range of applications by automatically monitoring and tracking products. With the proliferation of RFID-enabled applications, multiple readers are needed for ensuring the full coverage of numerous RFID tags. In this paper, we focus on the tag distribution problem in multi-reader RFID systems. The problem is to fast identify the tag set beneath each reader, which is a fundamental premise of efficient product inventory and management. Only with such tag set information can we localize specific tags in a reader and expedite the tag query information collection. As an RFID system usually contains a large number of tags and multiple readers, the traditional solution to identify tags by individual readers is highly time inefficient. We propose an Inference-Based protocol (IB) that identifies the tag distribution based on information inference rules and the aggregated physical signals to improve operational efficiency. In our protocol, three kinds of inference rules based on internal information reported by a single reader, external information shared by multiple readers, and history information retained by the system are fully exploited to infer tag distribution. With these rules, all readers can cooperatively work together and quickly obtain the tag distribution in the system. We also build a prototype RFID system using the USRP-based reader and WISP programmable tags, and then implement the IB protocol. The experimental results and extended simulations show that IB outperforms the state-of-the-art protocols.