Research On Ultra-Wideband 3D Positioning System And Algorithm

Ultra-wideband(UWB)technology has the characteristics of high temporal resolution,strong resistance to multipath effect,and low system complexity.It is the wireless positioning technology with the highest accuracy.In this thesis,an UWB 3D positioning system is designed to precisely locate targets in the air using four portable and movable ground base stations.First,for the hardware part,the base station for positioning is designed and fabricated in this thesis.The ground base station is an omnidirectional high gain UWB 3D antenna array working in X-band.The ground base station is composed of four dipole antenna arrays with center frequency of 10 GHz,bandwidth of 500 MHz and the gain in the working band greater than 10 d B.The ground base station is composed of four dipole antenna arrays.The center frequency of the dipole antenna array is 10 GHz,the bandwidth is 500 MHz,and the gain in the working band is greater than 10 d B.The airborne base station is controlled by an electronic switch,and only one of the four dipole antenna arrays is turned on during the positioning process.If the signal strength of the ground base station received by the airborne base station is below the threshold at this time,the microcontroller controls the switch of the airborne base station to use the four antenna arrays in sequence,selecting the direction with the highest signal strength and turning on the antenna array in that direction.The hardware part of the design ensures that the airborne base station achieves a wider positioning range with low power consumption during operation.For the algorithm part,based on the hardware design of the 3D positioning system,this thesis analyzes the commonly used rang-based positioning algorithms and finds that the target height cannot be located using the least squares method because of the small height difference between the four ground base stations.Therefore,this thesis replaces the solution idea and uses machine learning iterative algorithm for localization,and proposes a hybrid algorithm which combines the TOA algorithm with the LM algorithm,The hybrid algorithm can accurately locate the height of the target by iteration without providing accurate initial values,and overcomes the disadvantage that the accuracy of the calculation result of the iterative algorithm depends on the initial value.The algorithm has an accuracy of 40 m for 3D positioning in a 4km range.Because the hybrid algorithm has inaccurate localization results at some positions,the Kalman filter is used in this thesis to reduce the fluctuation of localization results.And the localization accuracy is improved by four times after adjusting the parameters in the filtering algorithm.This thesis uses chip DW1000 to build a model for scaling experiments,and the experimental results and simulation data have good agreement.It is finally concluded that the 3D positioning system built in the area of 200m× 200 m,using only four ground base stations with a height difference of 1m,can achieve a 3D positioning accuracy of 5m in the range of 2km,and about 10 m in the range of 2km-4km.Finally,for the software part of the system,this thesis introduces the functions of the positioning program,including the base station time-delay calibration,the ground base station automatically locates its own coordinates and other functions.The program can achieve the positioning functions by dividing the time slots for single and multiple targets.In addition,the control rules of multi-target delay positioning and the timing design of the targets in the positioning process are given.