Efficient Physical-Layer Unknown Tag Identification in Large-scale RFID Systems

Feng Zhu, Bin Xiao, Jia Liu, Li Jun Chen

Research output: Journal article publicationJournal articleAcademic researchpeer-review

22 Citations (Scopus)


Radio frequency identification (RFID) is an automatic identification technology that brings a revolutionary change to quickly identify tagged objects from the collected tag IDs. Considering the misplaced and newly added tags, fast identifying such unknown tags is of paramount importance, especially in large-scale RFID systems. Existing solutions can either identify all unknown tags with low time-efficiency, or identify most unknown tags quickly by sacrificing the identification accuracy. Unlike existing work, this paper proposes a protocol that utilizes physical layer (PHY) information to identify the intact unknown tag set with high efficiency. We exploit the physical signals in collision slots to separate unknown tags from known tags, a new technique to speed up the ID collection. Such new technique was verified in an RFID prototype system using the USRP-based reader and WISP tags. We also evaluated our protocol to show the efficiency of leveraging PHY signals to successfully get all unknown tag IDs without wasted known tag ID transmission. Simulation results show that our protocols outperform prior unknown tag identification protocols. For example, given 1000 unknown tags and 10 000 known tags, our best protocol has 56.8% less time to the state-of-the-art protocol when collecting all unknown tag IDs.

Original languageEnglish
Article number7736075
Pages (from-to)283-295
Number of pages13
JournalIEEE Transactions on Communications
Issue number1
Publication statusPublished - 1 Jan 2017


  • Physical layer
  • RFID system
  • time efficiency
  • unknown tag identification

ASJC Scopus subject areas

  • Electrical and Electronic Engineering


Dive into the research topics of 'Efficient Physical-Layer Unknown Tag Identification in Large-scale RFID Systems'. Together they form a unique fingerprint.

Cite this