A Study On Indoor Localization Algorithm Of Passive UHF RFID Based On AOA And PDOA

With the rapid development of information technology,people are paying more and more attention to the localization.Global Positioning System(GPS)can provide us accurate location in outdoor environment.However,there are many obstacles in indoor environment,and GPS cannot provide us accurate localization information because of the signal attenuation.Ultra High Frequency Radio Frequency Identification(UHF RFID)is getting more and more consideration to be the optimal selection for indoor positioning systems because it is easy to deployed,low cost and has high read-write speed.Due to the complex signal propagations caused by walls,clapboard,ceiling and so on,the UHF RFID signals may suffer from reflections,refractions and diffractions.The accuracy of indoor localization is affected by the phenomenon of non-line of sight(NLOS)and multipath arrivals seriously.To improve the accuracy of indoor localization,in this paper,we combine AOA with PDOA methods to implement an indoor localization method based on passive UHF RFID.We use the array antenna to distinguish multipath signals and choose the two strongest paths according to received signal strength(RSS)to perform localization.The angles of the two paths are obtained through the phase difference of the received signals at different array elements,and the distances of the two paths are estimated through PDOA measurement.After obtaining the angles and distances,we establish virtual stations to convert NLOS paths into LOS paths.The possible positions of tag are calculated through virtual stations,angle and distance information which are derived from the two signal paths.Then the weighted least squares combined with residual weighted(WLS-RW)algorithm is proposed to calculate real position of the tag.In order to obtain higher positioning accuracy,we also make a further study on the characteristics of the AOA algorithms and PDOA algorithm.We compare the accuracy of MODE(Method of Direction Estimation)algorithm and MUSIC(Multiple Signal Classification)algorithm with space smoothing technology.We also study the influence on the PDOA accuracy when a signal meet an obstacle and it is reflected,because there is phase change when a signal is reflected.At last,we establish the simulation environment in MATLAB and simulate the spread of the wireless signal through Ray-tracing.Simulation results demonstrate that,in a 10m*10m room,with multipath and NLOS signal environment,the position error of our proposed method is less than 0.5m with 80% probability.At the same time,our method has higher precision than traditional algorithms.