Research On Indoor Location Algorithm Based On Fingerprint

The continuous improvement of the Internet of Things technology has driven the development of a number of emerging technologies.Among them,the short-distance communication technology has gradually become a research hotspot under the support and drive of the Internet of Things technology.With the popularity of smart phones,indoor navigation and positioning technology is also called an indispensable part of people's lives.This paper mainly studies and improves several typical indoor positioning algorithms.The main research contents are as follows:Firstly,several typical Wi Fi positioning models are introduced.An improved indoor positioning algorithm based on the influence factor of Wi Fi signal is proposed.The improved algorithm is mainly based on the analysis of the main influencing factors of positioning accuracy.In order to achieve higher positioning accuracy,the data points in the fingerprint database are pre-processed in the early stage of database construction.Then use the influence factor to weight the positioning results,and analyze the advantages and accuracy of the improved algorithm through experimental comparison and analysis.Secondly,the principles of KNN algorithm and Bayesian matching algorithm are introduced,and a positioning algorithm based on the joint improvement of KNN and Bayesian is proposed to effectively reduce the impact of signal strength value fluctuations on the positioning results.Before the joint improvement algorithm is implemented,the choice of K value in the KNN algorithm is tested to determine the best K value.Through experiments,the error results of the traditional weighted algorithm WKNN and the joint improved algorithm are compared,and the error accuracy changes of the traditional algorithm relative to the improved algorithm are compared.Finally,the basic flow based on the realization of pedestrian trail estimation(Pedestrian Dead Reckoning,PDR)theory is introduced,and the advantages and disadvantages of the algorithm and the main scope of application are described.Use the built-in sensor of the smartphone to detect the pedestrian's step length and frequency.In the detection of directional angle detection,the directional angle is measured by setting a threshold,and the experimental results are obtained and the PDR algorithm is realized through experiments.An algorithm model based on Wi Fi and PDR fusion positioning is introduced.The implementation process based on the extended Kalman filter fusion algorithm is described.And correct the positioning result output by the fusion algorithm,and explain the positioning result of the fusion algorithm through the error accumulation function.Experimental results show that the fusion algorithm can not only reduce the impact of Wi Fi signal strength fluctuations on the positioning results,but also reduce the cumulative error of the PDR algorithm,improve the positioning accuracy of the algorithm,and reduce the scope of the error.