Research On Multipath Error Elimination Algorithm Based On UWB Positioning Technology

UWB positioning technology is widely used in today’s society where the network is deployed on a large scale.In the indoor environment,when UWB positioning technology is used for Location Based Service(LBS),due to the influence of multi-path interference and Non-Light of Sight(NLOS),the observation contains a large error,causes a low positioning accuracy.In order to reduce the influence of multipath effect on the positioning accuracy,this paper proposes a Time Difference of Arrival(TDo A)positioning scheme based on the UWB inverse positioning framework.In order to achieve high-precision real-time positioning function,this paper focuse on three aspects: the elimination of environmental noise,the construction and optimization of the multipath map model,and the design real-time visualization software.The main research results are as follows:1)Aiming at the problem that the traditional TDo A positioning solution algorithm cannot satisfy the low complexity and high precision at the same time,a weighted least squares-Taylor series expansion algorithm with joint analytical solution and iterative solution is proposed.The analytical solution can be quickly obtained as the initial label coordinates;the Taylor algorithm can use TDo A observations and initial label coordinates to iterate to obtain high-precision label coordinates.Compared with the commonly used Chan-Taylor positioning algorithm,the joint algorithm effectively reduces the algorithm complexity and improves the positioning solution speed.2)Aiming at the problems of environmental noise and outliers in TDo A observations,an adaptive threshold function wavelet denoising method is proposed.By introducing adaptive parameters,the oscillation of the hard threshold function and the distortion of the soft threshold function are effectively weakened.so that the reconstructed signal can satisfy both high continuity and high similarity.The optimal solution of wavelet basis function and decomposition level is analyzed by simulation experiments,and the feasibility of the method is verified by TDo A observation data.The experimental results show that compared with the single WLS-Taylor algorithm,the positioning accuracy of the combined AWD method is improved by 8.7%,and the convergence time of the positioning solution is effectively reduced,the average number of iterations is reduced by 57.1%.3)Aiming at the problem that the multipath map method will inevitably overfit or underfit the spatial multipath error,a trend surface analysis based multipath map method(TMPM)is proposed.The fitting coefficients,effective modeling time,static positioning accuracy,and dynamic trajectory tracking are analyzed using TDo A observations.The experimental results show that the average multipath error correction rate of TMPM method is 80.2%,and the average positioning accuracy is 16.0cm,which is 22.7% higher than that of MPM,and the local antiinterference and overall stability of dynamic trajectory tracking are effectively improved.4)According to the indoor inverse positioning framework and the positioning method mentioned above,this paper designs a real-time positioning visualization system based on the C/S framework and the Swing GUI toolkit.The system can collect data,utilize the processing module,the solution module and the display module,and display the positioning results on the user interface in real time.The experimental results show that in the static test scenario,the positioning accuracy of the system can be maintained at about 20cm;in the dynamic test scenario,the response time of the system is 0.3s,and the system has good stability.