Research And Design Of UWB Indoor Positioning System

The rapid development of wireless sensor network and Internet technology has increased the demand of people for indoor positioning,in shopping malls,warehouses,supermarkets,prisons and other places,the precise positioning of people or things can improve the efficiency of management greatly,in the outdoor environment,GPS can provide high-precision positioning,but in the indoor environment,the wall and the other obstacles make it hard for GPS use of its advantages.Luckily,UWB signal has the advantages of resisting anti-multipath effect,good anti-attenuation performance and low power consumption.Therefore,it is of great significance to develop an indoor positioning system based on UWB.The main contents of this paper are as follows:Firstly,a symmetrical bilateral bidirectional ranging method which can eliminate the influence of time dissynchronization and temperature drift is studied.The UWB Poll packet is first broadcast by the pending positioning tag.the base stations within System receive the Poll and after a period of delay send Resp package,label received Resp and send Final packet a period of time later,when the base station receives the Final package from label,a full range is completed,The actual measurement result shows that the ranging error of this method is about 4.7cm in the line-of-sight environment,and about 16.4cm in the non-line-of-sight environment with human body shielding.Secondly,the chan-weighted centroid algorithm based on Kalman Filter is used to locate the tag.Firstly,The algorithm takes Kalman Filter to process TDOA distance difference vector to eliminate big orientation deviation caused by larger ranging error,then the value of the generation after filtering into the chan's localization algorithm to solve the tag estimated coordinates,in the end,combining the advantages of improved weighted centroid algorithm and the result of forecast positioning solution,get the more accurate and stable coordinates,The MATLAB is adopted to simulate the algorithm,the simulation results show that the positioning error is only 0.07 m.Finally,combining with the research achievement of distance measuring and positioning,a UWB indoor positioning system was developed,the hardware part of system uses STM32 carrys DWM1000 communication module,The upper machine was developed using the python language based on the Py Qt library,and meanwhile,the localization algorithm proposed in this paper was transplanted on the upper machine,after completing distance measurement,range information will be sent to the upper machine,the positioning coordinates of the tag will be displayed on the interface,The measured results show that the average positioning error of the system is only8.4cm in the line-of-sight environment and 9.2cm in the non-line-of-sight environment.Simulation and spatial actual positioning data show that the system designed by handing the distance difference vector with Kalman Filter and modifying the estimated coordinates with the improved weighted centroid algorithm can basically meet the requirements of indoor positioning and provide some ideas for the research and development of indoor positioning system.