Research On Csi Single-node Indoor Positioning Method Based On Time Difference Of Arrival And Angle Of Arrival Estimation

With the rapid development of wireless networks and the widespread popularity of mobile smart terminal devices,the demand for indoor positioning services is also increasing.People spend most of their time indoors.Indoor technology has been widely used in various fields such as indoor emergency rescue,commercial navigation,personnel monitoring,smart home,and medical security.The main bottleneck for the large-scale promotion and application of indoor positioning technologies is that most indoor positioning technologies require dedicated sensors or infrastructure,and the hardware cost and complexity of deployment are too high.Wi-Fi infrastructure and terminal equipment have been widely used in various indoor commercial and home environments.The indoor positioning system based on Wi-Fi signals does not require additional infrastructure and dedicated hardware,is easy to use widely,and has low equipment costs.Therefore,Wi-Fi Fi indoor positioning has become a hot research topic.Compared with Received Signal Strength Indicator(RSSI),Wi Fi Channel State Information(CSI)contains more fine-grained useful information,so this paper studies the use of CSI to achieve Wi Fi indoor positioning.Considering the coverage of Wi Fi signals,the available access points in indoor environments are often limited,and it is very likely that only a single access point can be used.Therefore,in order to improve the availability of Wi Fi indoor positioning,this article mainly studies the single-node(access point)CSI indoor positioning method.First,this paper proposes a new CSI single-node indoor positioning method based on Time Difference of Arrival(TDOA)and Angle of Arrival(AOA)estimation.The single access point AOA/TDOA hybrid positioning method proposed in this paper does not need to change the existing hardware and Wi-Fi underlying communication protocol,only through the antenna resource configuration and spatial distribution optimization design of the single access point,combined with the mobile terminal Multi-antenna AOA and TDOA estimation,giving user positioning results.This article proposes a new single-access point antenna array placement method,which divides multiple antennas into two groups,and the antenna array with a relatively close distance is used to obtain the AOA signal from the user end to the access point antenna array;A group of antennas with a long distance are used for TDOA signal acquisition.Compared with TOA estimation,TDOA estimation does not need to maintain accurate clock synchronization,and the delay difference is estimated for single access point and multiple antennas sharing the same clock,avoiding most of the phase error factors and ensuring higher TDOA estimation accuracy.Secondly,in view of the time difference of arrival and angle of arrival estimation,this paper proposes to use spatial smoothing multiple signal classification algorithm(MUSIC)to estimate TOA for single antenna and AOA and TOA joint estimation for multiple antenna arrays respectively.The measured multi-antenna CSI signal matrix is ??spatially smoothed,and the virtual array element MUSIC model is constructed by using different sub-carrier phase shifts to increase the number of available array elements and improve the super-resolution estimation ability of multipath signals.Then,a Gaussian model clustering algorithm is proposed for direct path recognition and statistical processing of multipath signals to improve the accuracy of TDOA estimation.The clustering algorithm clusters the estimated values of TOA and AOA obtained by each CSI data packet into clusters,calculates the total variance of the estimated value data of TOA and AOA belonging to each path cluster,and measures each path as The possibility score of the direct path,the path with the highest possibility score is regarded as the direct path.The average value of the identified direct path corresponding clusters is used as the final direct path TOA and AOA estimates.Finally,the estimated TOA of the two groups of antennas is differenced to obtain the corresponding TDOA estimate.Finally,for the AOA/TDOA mixed signal fusion positioning problem,a weighted least squares positioning algorithm is proposed.This article first constructs the coordinate observation equation of the target to be located based on TDOA and AOA,and gives the direct analytical solution of the target to be located based on the principle of least square method.Since the actual AOA and TDOA estimates are affected by the external environment and measurement errors,the directly calculated coordinates may have large positioning errors,which need to be further optimized.This paper proposes to use the weighted least squares method for AOA/TDOA hybrid positioning,and constructs the weighted least squares observation and fusion positioning estimation model.According to the weighted least squares method,for AOA or TDOA positioning information with higher estimation accuracy,adaptively set and increase the corresponding fusion weight to improve the fusion positioning accuracy.This paper verifies the AOA/TDOA performance and mixed-signal weighted least squares positioning performance through a large number of simulation experiments and some actual experimental data,and confirms the feasibility and effectiveness of the single access point positioning proposed in this paper.