Fusion of RSS and Phase Shift Using the Kalman Filter for RFID Tracking

Radio frequency identification (RFID)-based indoor tracking system has been used for the localization and tracking of forklift and automated guided vehicles in warehouses. Most of the existing tracking methods rely on either received signal strength (RSS) or phase shift alone. This paper presents a new RFID tracking method that fuses the RSS and phase shift together to predict the instant position of a mobile target. The fingerprints database is first constructed by collecting the RSS values at reference points. After data normalization, the fingerprints are fed into the extreme learning machine for the rough position estimation. The radial speed of the RFID tag can be calculated by measuring the phase difference between two consecutive time steps. Then, the linear least squares method is exploited to fit the instant speed of the mobile tag. On the assumption that the trajectory of the mobile tag could be approximated by a series of line segments, the Kalman filter is utilized afterward to enhance the position estimation by combing the rough location estimation and the instant velocity. The performance of our method is studied in simulations and experimentally verified in our lab. The results demonstrate that our method is feasible and achieves higher accuracy than traditional tracking methods.