Research On Indoor Integrated Positioning Method Based On Heterogeneous Wireless Networks

In recent years,the development of wireless sensor network technology is changing with each passing day.It has gradually become the focus of public attention due to its advantages of simple deployment,low cost,ease of use,high efficiency,and high safety factor.It has been widely used in military,environmental monitoring,stock market forecasting,target tracking,image processing and other fields,and its branch-wireless indoor positioning and location service technology is currently one of the core issues of current research.The current mainstream indoor positioning technologies mainly include UWB technology,ZigBee positioning technology,WIFI positioning technology,radio frequency identification technology,and Bluetooth positioning technology.In recent years,some emerging positioning technologies have emerged,such as narrow-band Internet of Things positioning technology,visible light positioning technology,and ultra-long-distance low-power data transmission and positioning technologies.Relatively speaking,WIFI technology is the most widely used,its signal propagation distance and penetration ability are relatively strong,and UWB positioning technology has advantages such as higher positioning accuracy,multi-path resolution and stable high-speed signal transmission,etc.The field is unique.Combining the advantages of WIFI and UWB positioning technologies,this project has designed an indoor positioning system.The system consists of a monitoring platform,four base stations and a positioning tag.Among them,the system monitoring platform adopts the(Exxos4412)Samsung quad-core processor Exynos4412 as the main control chip.With the help of the embedded Linux operating system,the main functions of monitor display,touch control,position coordinate calculation,and send and receive instructions are implemented.In the system,the positioning base station and the positioning tag all use TI's CC3200 as the core chip to realize data collection,processing and uploading.In view of the defects of these two positioning technologies,this paper proposes a heterogeneous wireless network location algorithm based on WIFI technology and UWB technology.Firstly,the location location tag of the positioning tag is roughly obtained through the WIFI location fingerprint fast location algorithm;then the hybrid particle swarm optimization algorithm based on the ultra-wideband technology is adopted,and the initial particle is selected with this region as the boundary condition,and the number of iterations of the algorithm is reduced while the positioning result is improved.Robustness,speed and positioning accuracy.For the traditional K-means clustering location fingerprinting algorithm,the problem of slow positioning speed and low positioning accuracy in the positioning estimation stage,this paper removed the fingerprints with lower similarity and introduced new weighting coefficients,improved the positioning algorithm and proposed a An Improved K-means Clustering Algorithm for Fingerprint Location.In the positioning process,the traditional particle swarm optimization algorithm is used,there is a problem that it is easy to fall into the local extreme point,the convergence process is slow and the accuracy is low near the extreme point.This paper combines the simulated annealing algorithm and the particle swarm optimization algorithm to establish A set of weighted models for reducing the NLOS error of anchor nodes was incorporated into the algorithm,and the idea of"single wins" was incorporated into the algorithm.The Metropolis criteria was used to eliminate particles with poor fitness.Compared with the existing hybrid particle swarm optimization algorithms,the simulation and test data show that the proposed algorithm has a higher positioning accuracy and a faster convergence speed,and can better suppress the influence of non-line-of-sight errors.Finally,according to the algorithm,a set of monitoring platform application program based on embedded Linux operating system was designed,including data receiving application,data processing application,fingerprint database building application.In the laboratory scenario,actual positioning tests were performed.The positioning system designed in this project can achieve positioning accuracy of 15cm.