Research On TDOA Cooperative Location Algorithm Based On Error Decision

In recent years,the rapid development of key technologies such as communication and the Internet of Things has made location-based services widely used in many fields as a strategic emerging industry,and positioning and navigation have gradually become an inseparable part of daily life.Indoor wireless positioning technology is a supplement to the current mature and stable satellite positioning system,wireless positioning signal in the process of positioning base station and terminal transmission is easily affected and interfered with by complex environment,resulting in reflection,refraction and other phenomena,such phenomena caused by non-line-of-sight transmission will lead to time,incidence angle and other positioning measurement There are large errors,which in turn affect the final positioning solution results.In order to reduce the interference of error signals,improve indoor positioning performance,achieve error suppression and provide new optimization schemes for indoor positioning services,this paper studies the non-line-of-sight error judgment strategy and time difference positioning algorithm of wireless positioning signals in complex indoor scenarios.The specific research content and results are as follows:1、aiming at the identification problem of NLOS error in indoor environment,without prior ranging information,this paper introduces a support vector machine to identify the ranging residuals,and uses the normalized residual processing method to eliminate the influence of the problem of excessive residual residuals in multi-base station positioning and solving in the identification process.At the same time,in order to improve the recognition performance of the algorithm,the advantages and disadvantages of the quadratic term of the normalized residual function are used as the classification parameters to generalize the recognition ability of the vector machine,and finally the polling re-recognition method is used to further optimize the purity of the LOS base station.In addition,in extreme cases,when the number of LOS base stations is insufficient to complete the positioning solution requirements,this paper uses the NLOS correction algorithm of grid voting to perform NLOS correction on the identified ranging values,and weights the LOS and NLOS signals according to the Laita principle to complete the grid voting election,so as to improve the robustness of the non-line-of-sight error judgment strategy.2,aiming at the problem that the accuracy and stability of the traditional time difference positioning algorithm need to be improved,this paper proposes an optimized sparrow search algorithm and a Chan collaborative positioning strategy,and after the traditional Chan solution obtains the preliminary position prediction coordinates,the optimized sparrow search algorithm is used to iteratively find a better coordinate position based on the initial coordinate divergence.The inverse learning strategy is introduced into the sparrow search algorithm,which makes the optimization speed of the proposed algorithm faster and avoids the algorithm falling into local optimum,and the adaptive speed update factor is introduced in the process of sparrow position update to make its search results more delicate and more accurate.3,in order to verify the accuracy and validity of the above research content and enable the positioning results to be visualized,this paper builds a demonstration system based on 5G positioning.The whole positioning system can be divided into positioning transceiver platform,positioning solution platform,positioning demonstration platform.With the positioning solution platform integrated with the positioning algorithm proposed in this paper as the core,cross-platform communication is realized through the HTTP protocol,so that the ranging information of the positioning transceiver platform is solved on the positioning solution platform,and finally displayed on the mobile positioning demonstration platform.Based on the above platform,the performance of the non-lineof-sight error judgment strategy and the optimized time difference localization algorithm proposed in this paper is verified,and the results show that the algorithm has improved the NLOS recognition accuracy,positioning stability,positioning solution convergence speed and positioning accuracy to a certain extent.