Efficient RFID Data Cleaning Method

Abstract

RFID (Radio Frequency Identification) technology transfers data between movable tagged objects and readers without line of sight, and the captured data tends to be noisy. The inherent unreliability makes the data unreliable to application. Nowadays, the main solution is to use sliding window, but it is difficult to decide the window size, especially when the tag moves frequently or with high false positive. To solve the mentioned problems, SWKF (Sliding Window based on Kalman Filter Pre-processing) is proposed. It preprocesses the RFID data to make the read rate close to the real one, detects and filters the mobile tags. Then, the preprocessed data is smoothed to further improve accuracy. At the same time, the mid-window slide point reduces the output. Through the combination of Kalman Filter and sliding window, SWKF provides accurate RFID data to application. DOI: http://dx.doi.org/10.11591/telkomnika.v11i3.2154