An Improved Indoor UHF RFID Localization Method Based on Deviation Correction

In indoor localization environment, RFID signal propagation is complicated because of the reflection and diffraction and NLOS. The LANDMARC system and KNN algorithm cannot achieve satisfactory accuracy. By using the passive UHF RFID system, we propose an improved KNN method to rectify the target's coordinate position. First we use the traditional method to collect the reference tags RSSI and KNN algorithm to get the target position. Focused on the target's k nearest reference neighbor tags, we used the KNN algorithm k times to get each reference tag's calculated coordinate position. Compared with the known position, the average position deviation is used to get the target a more accuracy position. The simulation results dedicate that the improved KNN method increases localization accuracy and is suitable for the multi-targets indoor environment.