Research On High-precision Fusion Positioning Algorithm For Ultra-wideband Communication Technology

With the innovation of intelligent Internet of Things technology,location acquisition has become more and more important in life,and human production and life are inseparable from accurate acquisition of spatial location.With the advent of the 5G era,higher requirements have been put forward for location services.In recent years,wireless positioning technologies have emerged one after another,including QR code positioning technology,Wireless Fidelity(WIFI),Ultra Wide Band(UWB),and Bluetooth,Radio Frequency Identification(RFID).Among them,UWB positioning technology has become the current mainstream positioning technology due to its high transmit power,good safety performance,and strong anti-interference ability.At present,some UWB products have been tested in related environments to provide location services for various devices.However,in complex indoor environments such as computer rooms and warehouses,the UWB signal will be greatly attenuated due to obstructions,resulting in large positioning deviations.,Affecting the final positioning result.In response to the above problems,this paper uses Micro-Electro-Mechanical Systems(MEMS)sensors to assist UWB to achieve positioning in complex environments,formulates a fusion positioning model,and proposes Kalman fusion Gauss Iteration(GUI)algorithm.Pedestrian motion data is collected separately through MEMS sensors and UWB for fusion,thereby solving the problem of positioning errors caused by UWB signal loss.The specific implementation steps of the fusion positioning algorithm used in this paper are as follows.Firstly,UWB and MEMS sensors are used to determine the pedestrian's initial position and state information,and the fusion algorithm is used to determine the pedestrian's heading angle in the initial state.Secondly,the nine-axis inertial sensor is used to determine the pedestrian's stride frequency according to the walking characteristics of the pedestrian,the peak information and threshold information output by the accelerometer,and the pedestrian step length estimation is determined according to the pedestrian step frequency and step length model.Finally,the fusion algorithm is used to iteratively filter the error of the step length estimation,and finally the pedestrian step length and heading information are estimated.In view of the above problems,this paper proposes the latest fusion positioning algorithm,and makes the prototype and host computer of the positioning system based on the fusion positioning model.Based on the STM32F103VET6 microprocessor,the pedestrian position data analysis and fusion algorithm are used to process the data,and the host computer interface is developed to realize the visual display of the position data waveform.The experimental data is obtained by recording the linear and curved trajectories of pedestrian movement.The verification shows that the fusion algorithm proposed in this paper is significantly better than the traditional fusion positioning algorithm,and the success rate of pedestrian position coordinate analysis is significantly improved.