On Number of Tags Estimation in RFID Systems

According to the latest version of the radio-frequency identification (RFID) standard, EPCglobal UHF Gen2, dynamic framed slotted ALOHA has been accepted and employed as the de facto collision resolution algorithm to share the channel usage when multiple tags respond to the reader's signal command simultaneously. However, the frame size in read cycle is usually far from optimal when the reader does not know the number of unidentified tags in its interrogation zone, and this will lead to several performance issues and technical limitations such as power consumption, longer reading time, and degraded system performance. In this paper, we show that the number of unidentified tags can be expressed as a function of the collision rate received by the reader when the reader collects tags' information, and then we use an extended Kalman filter-based tag number estimation algorithm to estimate the number of unidentified tags in an RFID system on run-time measurements. Simulation results show that our scheme provides much better accuracy than existing well-known approaches even in a fast-changing environment.