Research On High-precision Indoor Positioning Algorithms Based On UWB In Different

With the improvement of living standard,people’s demand for location-based services in various scenarios is growing.Bei Dou,GPS and other satellite navigation systems cannot be positioned indoors due to the inability of satellite signals to penetrate walls,and traditional indoor positioning methods such as Bluetooth and Wi Fi are also difficult to meet people’s usage needs due to accuracy issues.Ultra-Wideband(UWB)has entered people’s view,which has many advantages such as strong multipath resistance,high temporal resolution and low power consumption,and can achieve centimeter-level positioning accuracy indoors.This paper studies and optimizes the high-precision indoor positioning algorithms in different scenarios from the actual indoor positioning requirements.Therefore,starting from the actual indoor positioning needs,this article studies and optimizes high-precision indoor positioning algorithms in different scenarios.Firstly,based on the definition and implementation process of UWB,this paper illustrates some characteristics and advantages of the signal,such as low transmitting power,fast signal transmission,strong anti-noise and high positioning accuracy.The principles of different algorithms for obtaining measurement information are compared,and the various sources of measurement errors are analyzed.Secondly,improvements were made to the Extended Kalman Filter localization algorithm based on phase difference of arrival.To address the problem of lag in prediction due to the absence of the tag motion state model in the traditional algorithm,an improved method of adaptive extended Kalman filtering is proposed that adaptively and dynamically adjusts the filtering gain K by the magnitude of the information contained in both the rate of change of phase difference and the rate of change of distance.The experimental results show that the improved algorithm effectively improves the positioning accuracy of the system.Then the time-of-flight based localization algorithm is optimized.An iterative weighted least squares positioning algorithm is proposed by weighting the measurement information of each base station by the measured distance.And proposed an estimation method for tag motion speed based on the measurement information of each base station.The estimated velocity is also used to refine the motion state model of the tag.Experiments are designed to verify the performance of the algorithm,which is able to estimate the velocity state well for tags in motion and improve the accuracy of localization.Finally,this paper proposes a layout structure for vehicle-mounted base stations.By comparing the signal quality of each base station in line-of-sight and non-line-of-sight environments in detail,a method for selecting base stations in line of sight environments based on signal strength was proposed.And the current measurement error was estimated using historical positioning results,and then the tracking and positioning of the label outside the vehicle was achieved through weighting.Due to the poor measurement quality of invehicle tags,it is impossible to locate them in the traditional way.A method of judging inside and outside the vehicle based on the measurement information and signal strength with the location of the vehicle is proposed,and experiments are designed to verify the feasibility of the algorithm.