Radio frequency identification (RFID) technology brings tremendous advancement in Internet-of-Things, especially in supply chain and smart inventory management. Phase-based passive ultra high frequency RFID tag localization has attracted great interest, due to its insensitivity to the propagation environment and tagged object properties compared with the signal strength based method. In this paper, a phase-based maximum-likelihood tag positioning estimation is proposed. To mitigate the phase uncertainty, the likelihood function is reconstructed through trigonometric transformation. Weights are constructed to reduce the impact of unexpected interference and to augment the positioning performance. The experiment results show that the proposed algorithms realize fine-grained tag localization, which achieve centimeter-level lateral accuracy, and less than 15-centimeters vertical accuracy along the altitude of the racks.
Download