Research On Indoor Wireless Positioning Technology Based On UWB

With the development of the information age,people's demand for location-based services is increasing,which makes indoor positioning one of the most popular research topics.Among them,high-precision,low-latency indoor positioning technology has become the focus of research and has attracted more and more attention from experts,scholars and related industries.Ultra-wideband(UWB)signals are suitable for indoor high-precision positioning due to their high temporal resolution and strong penetrating ability.Based on UWB technology,this paper analyzes the signal waveform,generation mode and indoor channel model,discusses the ranging and positioning mode suitable for it,and mainly studies the Cramer Rao Lower Bound(CRLB)of positioning error variance and the iterative cooperative positioning algorithm combined with it.Firstly,this paper analyses the principle of three commonly used wireless positioning methods,points out the shortcomings based on field strength and angle ranging.By analyzing the CRLB of the ranging error based on TOA,it is found that the large bandwidth of UWB can obtain higher ranging accuracy and high time resolution,so this paper uses UWB location based on TOA.This paper studies some commonly used wireless location algorithm used based on TOA and points out its application conditions,advantages and disadvantages.Then,the CRLB of the positioning error variance is selected as the standard to measure the positioning accuracy.When the coordinates of base stations are known,the Fischer information matrix can be used to calculate the CRLB of location in white Gaussian noise.The numerical value of the CRLB depends on the number of base stations and relative geometric relationship of base stations location.Then,an iterative collaboration method based on least squares estimation(LSE)in multi-target localization is studied.Using the distance information between tags,the tag is treated as a virtual “base station” to refresh the tag estimate through iteration.This method has low complexity and simple implementation.It can achieve higher accuracy than non-cooperative methods.However,the above algorithms have some limitations: when estimating the initial position of each tag,because of the existence of errors,the credibility of the initial estimate is relatively low.Tag location information carries the amount of errors.When it is added to the calculation for iteration,it can be found that with the increase of iteration times,errors accumulate continuously,resulting in the accuracy can not be further improved,or even may be reduced.Especially when there are some tags with large estimation error,the impact of initial error may be great.In view of the problem of cumulative errors,a cooperative localization methodbased on CRLB to determine the confidence level of the estimated value is proposed.The node information with large error in initial estimation is screened out and removed in iteration.When the geometric relationship of the base station location is determined,the CRLB distribution in the location area is known.The CRLB represents the lower variance limit of the error at that point.So the smaller the CRLB value,the higher the positioning accuracy,the higher the confidence level of the estimated coordinates.Therefore,under certain conditions,CRLB is used to measure the confidence level of the tag position estimation.CRLB estimation nodes higher than the given threshold value are screened in the collaboration.Under certain strategies,the estimation information is removed in the iteration to reduce the accumulation of errors.After a lot of simulation experiments,this method has higher accuracy than the original LSE iterative collaboration algorithm,especially with the increase of iteration times,which shows that the accumulation of initial estimation error in the iteration is well controlled.