Research On Indoor Localization Based On WLAN Fingerprint Positioning

With the continuous development of digital technology,various industries have a strong demand for high-precision indoor positioning systems.Indoor positioning technology has many application prospects in many areas such as commercial applications,disaster rescue,and drone control.The positioning accuracy of the satellite positioning system is affected by the multipath and signal attenuation caused by obstacles,resulting in poor indoor positioning results.Thanks to the continuous development of the Internet industry,wireless network access equipment is widely deployed,and the use of wireless routers to achieve indoor positioning can effectively reduce deployment costs.The main research content of this subject is based on the location fingerprint positioning technology,focusing on the optimal sampling parameters of the fingerprint map,the algorithm improvement in the actual environment,and the correction problem after positioning.First of all,this paper uses simulation experiments to study the parameter selection of fingerprint map.The result of fingerprint positioning is greatly affected by the arrangement of sampling points and the sampling interval.Through simulation experiments,this paper determines the best acquisition method in the offline acquisition stage.At the same time,the relationship between the sampling interval and the positioning accuracy is studied.Secondly,research and improve the existing problems of the KNN algorithm under the fingerprint map of the actual environment.Through the location fingerprint map constructed in the actual scene,the ideal KNN algorithm is improved in application.In the case of small signal,the influence of the limit of the received signal strength threshold on the positioning accuracy is explored.Based on the analysis of the positioning error in the open scene and the corridor scene,a fast algorithm of corridor segmentation fingerprint map is proposed,which greatly improves the positioning speed.Aiming at poor positioning in harsh environments,an improved KNN algorithm with extended data using threshold is proposed to improve positioning accuracy in harsh environments.Finally,the problem of positioning path correction and fingerprint image correction is studied.The Kalman filter algorithm is applied to realize the correction of the positioning trajectory,and the performance of the Kalman filter in this scenario is analyzed.Aiming at the inaccuracy of the fingerprint map caused by the time-varying signal in the environment,an error correction model is established through the perceptron,and a small number of supplementary collection points are used to improve the positioning accuracy of the fingerprint map at different times.