Research On TOA Localization Method In Non-line-of-sight Environments

The problem of target localization has been widely concerned by people in recent years,and the accurate location of targets plays a very important role in both military and civilian systems.Radio target localization mainly depends on information such as time-of-arrival(TOA)and time-difference-of-arrival(TDOA).However,due to the complex and changeable indoor environment,the measurement information based on TOA/TDOA contains none-line-of-sight(NLOS)error,so it is difficult for traditional sensor localization technology to obtain accurate and effective location information in indoor environment.Therefore,the localization method of suppressing NLOS error has become a hot and difficult point in this field.This paper mainly studies the method of TOA localization in NLOS environment,and the main work is summarized as follows:(1)The accuracy of cooperative localization can be severely degraded in NLOS environments.To mitigate the NLOS errors,the cooperative localization problem based on TOA under the mixed line-of-sight(LOS)/NLOS conditions is addressed.By studying the topological relationship between nodes,a TOA NLOS mitigation cooperative localization algorithm based on the topological unit is proposed.This algorithm is implemented under the classical multidimensional scaling framework.The adjacent topological unit of NLOS measurements are successfully identified by using the LOS matrix,and the NLOS measurements are re-estimated using topological units.The least-square method is used to transform the relative coordinates into absolute coordinates depending on the location of the anchor nodes.Compared to the existing methods,by employing the topological unit,this algorithm only requires the number of LOS anchor nodes to be 2 in the two-dimensional plane,and the better localization performance can be achieved.Simulation results show that the proposed method works well for both the sparse and dense NLOS environments.(2)This work addresses the TOA localization problem under mixed LOS/NLOS environments.In order to alleviate the problem that the existing robust methods usually perform poorly in sparse NLOS environments,we introduce a ”balancing parameter”related to NLOS error in the modeling process,and establish a new robust weighted least square(RWLS)problem to jointly estimate the source location and NLOS balancing parameter.Because the localization model can not be solved directly,by transforming the localization problem into a generalized trust region sub-problem,although this kind of problem is still non-convex in general,it can be solved accurately by a bisection procedure.This method does not need any statistical information of NLOS errors,nor does it need to identify NLOS paths.Simulation results show that this method can effectively improve the localization performance of the robust method in sparse NLOS environments,and has a stable ability to suppress NLOS errors in sparse and dense NLOS environments,so it can always provide a feasible solution.(3)A simulation platform of NLOS error suppression TOA localization method based on MATLAB is developed.The platform integrates most of the existing NLOS error suppression methods,and the performance of the method can be verified by simulation data and measured data,and then the positioning performance of the method is shown by calculating RMSE and CDF,and the practical functions such as real-time viewing of the corresponding positioning node data and displaying the positioning scene can be seen in real time,and the serial port data can also be used for real-time positioning development.