User Localization System Based On Fingerprint Database And Measurement Report

The remarkable progress of China's mobile communication technology makes location-based services(LBS)based on wireless location have a greater development space.Compared with GPS and other satellite positioning systems,the terminal wireless positioning based on wireless networks has the advantages of low cost,convenience and speed,and has outstanding performance in areas such as emergency rescue,network optimization,and epidemic prevention and control.This paper mainly studies outdoor user positioning based on fingerprint database and Measurement Report(MR),develops and accomplish a user positioning system platform.Specifically,this paper focuses on the rapid and efficient establishment of fingerprint database in offline stage and more reasonable and accurate matching algorithm in online stage.The main contents of this paper are divided into the following three points:1.A method for constructing an outdoor fingerprint database based on the improved Universal Kriging(UK)interpolation is proposed.In this paper,the fitting function model in the Ordinary Kriging interpolation algorithm is improved to make the entire Universal Kriging algorithm model more accurate and robust.By simulating the reference signal receiving power(RSRP)of a small number of drive test points and the location parameters of the base station,a complete outdoor fingerprint database is constructed by using other interpolation algorithms and the improved Universal Kriging interpolation method,and the performance of each algorithm is compared.The simulation results show that the proposed algorithm improves the accuracy of the whole fingerprint database and reduces the consumption of human and material resources.2.An improved weighted K-nearest neighbor(WKNN)method for matching during the online stage is proposed.This paper analyzes the relationship between RSRP and location distance,and then a weighted K-nearest neighbor algorithm based on signal similarity and spatial location is proposed.In this algorithm,the approximate position distance(APD)is obtained by making full use of the location distance between reference points(RP)and weighted Euclidean distance(WED),and the user's location is estimated by the weighted k-nearest neighbor algorithm based on the approximate position distance(APD-WKNN).In addition,the positioning performance of the common matching algorithm and APD-WKNN algorithm are compared.Simulation results show that the proposed algorithm has higher matching accuracy.3.The user positioning system based on fingerprint database and MR is accomplished.The system establishes an outdoor user positioning system through the MR data collected from the base station side and the improved fingerprint database positioning algorithm.The process of this system is mainly divided into three steps.The first one is to associate user information through MR data and signaling data;the second is to complete the establishment of the fingerprint database through a small amount of field-measured data in the offline phase;the third is to match the input MR data in the fingerprint database after associating with the user in the online stage,so as to complete the positioning of the user.The positioning error of most users is less than or equal to 50m,and the user trajectory can be described by Kalman filter.