NLOS Identification And Mitigation In Ultra-wideband Positioning

In recent years,the vigorous development of key technologies such as the Internet of Things(Io T)has made location-based services widely used in various fields,accelerating the breakthrough and development of positioning technology.At present,some relatively mature positioning technologies have played an indispensable role in various fields,and the demand for high-precision,low-cost,and universal indoor positioning applications in various industries is increasing.Ultra-Wide Band(UWB)technology stands out in the field of indoor positioning with its unique advantages.However,in complex indoor environments,UWB signal propagation is easily blocked by obstacles,resulting in non-line-of-sight(NLOS)propagation,resulting in NLOS error,which seriously affects the positioning accuracy.Therefore,effectively identifying and correcting the NLOS error has important research significance for the improvement of UWB positioning accuracy.This thesis studies the NLOS recognition and NLOS error correction problems in UWB positioning.The main work is as follows:First of all,this thesis draws on the advantages of the Ada Boost algorithm,which is easy to combine multiple signal features and has a high detection rate.Combined with the NLOS feature parameters,a NLOS recognition method based on Ada Boost is proposed.According to the signal difference caused by Line of Sight(LOS)propagation and NLOS propagation,five NLOS feature parameters are selected to construct the corresponding weak learner,and then several weak learners composed of a single feature are jointly constructed to construct a strong learner with high recognition performance to achieve effective recognition of NLOS propagation signals.Based on the MATLAB experimental platform,the algorithm is simulated and tested under different feature sets and different NLOS ratios.The results show that the proposed method has a high NLOS recognition rate and relatively stable recognition performance.Secondly,in order to solve the problem that the existing UWB positioning method is affected by the NLOS error,which leads to the decrease of the positioning accuracy,an improved unscented Kalman filter algorithm based on the measurement value correction and M-estimation is proposed on the basis of the NLOS identification.According to the positive deviation of the NLOS measurement value,the original observation value is compared and corrected,and a nonlinear regression model Mestimation method for adaptively adjusting the noise covariance matrix is proposed.The M-estimation method incorporates the standard unscented Kalman filter to alleviate the influence of the localization accuracy drop caused by NLOS outliers in moving target localization,thereby improving the localization accuracy in the indoor NLOS environment.Finally,the improved method proposed in this thesis is implemented based on the P440 platform.The test results in different NLOS ratio environments show that the recognition rate and stability of the NLOS recognition method proposed in this thesis are due to the traditional method,the improved NLOS error correction and the location estimation algorithm has superior performance,the localization accuracy is higher than the traditional filtering-based localization algorithm,and has good robustness,which can effectively reduce the impact of NLOS.