Study On The RFID Indoor Locating Technology Based On The Parameter Estimation

With the rapid development of manufacturing automation, the location of movingtargets in the manufacturing plants has an increasing of the complex environment in theworkshop, the effect of traditional indoor positioning technology on infrared, ultrasoundand other technology is usually more susceptible to be interfered by multipath andrandom noise of the environment. With low-cost, non-line-of-sight, non-contact, theRFID technology becomes an excellent candidate for manufacturing plant location. Inthis paper, a depth study has been done in both the indoor positioning technology basedon RFID and its applications in the manufacturing site.On the basis of the Received Signal Strength Indication (RSSI) propagation mode,the distance-power model is put forward according to analysis of the energy loss throughthe propagation path of UHF RFID system. The new model of Indoor location algorithmbased on parameter estimation model on RFID technology is put forward creativelythrough mathematical derivation, and the indoor location algorithm on the new model isgained, which has more scientific value compared with the traditional algorithm.An implementation is proposed according to the RFID localization algorithm basedon parameter estimation. The algorithm is gained from a large number of RSSI value andlocation information labels of reference tags. The solution of achieving positioning comesfrom statistical algorithm model, and the validity of RFID localization algorithm isverified by simulation with the software of MATLAB.A more valuable platform has been built to verify the accuracy and stability of thenew algorithm based on parameter estimation and the traditional algorithm calledLANDMARC. The experimental results show that the most accurate positioning is up to0.35m in the parameter estimation algorithm, which has an increase of15percentcompared with LANDMARC. The reliability and stability about LANDMRC are greatlyaffected by value with random changes under the same experimental conditions, but thelocalization algorithm based parameter estimation performs very well, which behavesbetter to adapt to the complex experimental environment.