Research On Fingerprint Positioning Technology In Heterogeneous Network Environment

With the rapid development of mobile Internet technology,more and more mobile Internet business needs to be based on the location information of mobile terminals.Therefore,the topic of precise positioning of mobile terminals has been the research hotspot in recent years.At the same time,with the development of information fusion technology,heterogeneous network fusion has become an irreversible technology trend.Under the background of the above technology,this thesis focuses on the research of mobile terminal location in WLAN/Bluetooth heterogeneous wireless network environment.Firstly,this thesis summarizes the research background of the subject and the research status of the related subject at home and abroad,and introduces the positioning technology commonly used in indoor wireless positioning.The basic principle of geometric wireless localization algorithm based on distance measurement and positioning algorithm based on position fingerprint are introduced.This thesis focuses on the advantages and existing problems of the Bayesian probabilistic position fingerprinting algorithm compared with the traditional deterministic algorithm in positioning accuracy.It is found that the location fingerprinting algorithm based on Bayesian is more accurate than traditional NN,KNN and WKNN.However,there are too many sampling points in the process of establishing the position fingerprint database,which affects the efficiency of the collection.And the complexity of the Bayesian posterior probability matching algorithm is too high to increase the positioning speed and other issues.To solve the problems found in the previous research,this thesis proposes a method of location fingerprint localization for WLAN/Bluetooth heterogeneous network based on interpolation algorithm and online matching algorithm,which reduces the workload of the location fingerprint positioning algorithm in the database establishment phase.This method establishes position fingerprint database by a linear interpolation method combining wireless signal propagation loss model.And the vector similarity matching algorithm and improved Bayesian fingerprint matching algorithm are used for position fingerprint matching in the WLAN and Bluetooth wireless network environment.And then the weighted vector similarity localization algorithm and the optimized Bayesian posteriori probability localization algorithm are used to determine the target position.The experimental feasibility and the improvement of the positioning performance of the method are verified by experiments and Matlab data simulation.The simulation results show that positioning accuracy of the heterogeneous network localization algorithm used in this thesis is improved compared with the single network.The main content of this thesis is of significance to do research on the positioning of target in WLAN/Bluetooth heterogeneous network.