| With the progress of the times,people’s demand for indoor positioning is also getting higher and higher,especially in crowded commercial areas,shopping centers,museums,hospitals,and other public spaces.Accurate positioning technology can provide users with more convenient services,such as map navigation,search,advertising recommendations,and so on.With the demand for high-precision positioning,many indoor positioning technologies have been rapidly developed.Ultra wide band(UWB)technology has received widespread attention due to its unique signal characteristics,and occupies an important position in many radio based indoor positioning technologies.UWB signals can achieve high positioning accuracy within the range of Line of Sight(LOS),but their positioning accuracy will also decrease as the measurement error increases.However,achieving high-precision positioning under Non Line of Sight(NLOS)conditions remains a challenge.Therefore,Thesis improves the UWB based localization algorithm for LOS conditions to achieve higher accuracy localization.For NLOS conditions,a fingerprint technology localization method is proposed,and traditional fingerprint localization algorithms are optimized.The main work of the paper is as follows:1.Through the simulation of the LOS environment,this article deeply discusses the principle of the TDOA algorithm based on UWB,and combines it with the traditional Chan algorithm and Taylor algorithm to verify the performance of the two algorithms under different error conditions.The results show that when the ranging error is large,the positioning accuracy of the TDOA algorithm will decrease.Therefore,Thesis proposes an improved Constrained Total Least Squares(CTLS)localization algorithm based on the alternating direction multiplier method.Through simulation experiments,it is found that even in the case of large measurement errors,it can achieve good localization results.2.Through simulation in the NLOS environment,the RSSI value of traditional fingerprint localization algorithms is affected by environmental changes,which seriously reduces the accuracy and reliability of localization.Therefore,by using UWB technology,high-precision ranging results can be obtained and used as fingerprint quantities.This method has higher reliability than traditional fingerprint databases.Then,based on the new fingerprint database,a fingerprint location algorithm combining dichotomous K-means and weighted KNN is proposed.After simulation,the positioning accuracy of this algorithm is significantly improved,and compared with traditional algorithms,it has stronger stability and higher positioning accuracy. |
