| With the fast development of mobile Internet technique,positioning services attracts people’s more attention.In the outdoor environment,Global positioning system(GPS)with cellular network technology can achieve faster and more accurate positioning,which has been widely used in various industries.In modern society,according to the survey that people spend at least 80% of whole day in the indoor environment every day.Therefore,how to provide convenient location service in the indoor environment has become an urgent problem to be solved.Researchers have already come up with some indoor positioning solutions: Bluetooth,ultra-wideband,radio frequency,WiFi signal and light tracking,but each positioning technology requires external conditions.For example,wireless location technology requires additional costly facilities,and the signal will be interrupted by the complicated room factors,resulting in reflection,refraction,multi-path effect,which can not guarantee high accuracy.So only one single positioning program could not meet location service requirements for a variety of indoor scene.In recent years,the researchers found that the unique reinforced concrete structure of modern buildings will produce geomagnetic anomaly field in the local area,and its characteristic information will remain stable for a long time.In theory,it can be used to locate with the magnetic field characteristic vector uniquely at different positions.However,when the indoor geomagnetic matching is realized in a large range,there is a problem merges that the initial process filtering convergence is slow and similar magnetic field causes the failure of positioning.On the other hand,in the fast development of wireless communication technique,large indoor spaces have many WiFi hot spots.Using traditional signal propagation model based on positioning technology,researchers can get a higher positioning accuracy,but need more expensive hardware facilities and more ideal conditions of the indoor scene.The location fingerprint algorithm based on WiFi signal strength(RSSI)is strong and easy to operate,but the positioning accuracy is low.Taking into account the advantages and disadvantages of geomagnetic and WiFi positioning schemes,this paper presents an indoor positioning method that combines indoor geomagnetism and WiFi signal strength.In large positioning area,the wireless signal strength is used roughly to find which small area the target is in and then help geomagnetic filtering in next matching.This paper firstly describes the characteristics of geomagnetic field and indoor positioning principle with magnetism,and then introduces several principles and methods of common indoor positioning technology.Then this paper summarizes the characteristics of magnetic field environment inside the reinforced concrete structure.In order to reduce the error of magnetic field measurement,an improved idea is put forward which combines improved limiting filtering and average filtering to process original observation data.And using experiment of hard magnetic compensation for magnetic sensor to reduce impact of measurement noise.Experiments show that the Kriging interpolation algorithm has high accuracy in the construction of indoor magnetic maps.In this paper,the fusion of geomagnetic and RSSI localization algorithm is divided into two stages: building and matching.In the first phase,the magnetic field strength and RSSI observations of the positioning area are collected in advance and the database is established.In the second phase,using the proposed method to determine in which small area the target is through the RSSI observation fingerprint,followed by the next step of geomagnetic matching.The fusion algorithm reduces the initial searching range of random particles in geomagnetic matching,reducing the number of particles and the convergence time,and improves system positioning efficiency.In order to avoid the interference caused by the similarity field in the geomagnetic map,which will possibly lead to drift phenomenon in the matching of the geomagnetic particle filter algorithm,this paper proposed an improved algorithm of error circle constraint to improve the positioning success rate for estimating object position.The improved algorithm is compared with the classical geomagnetic filtering algorithm through the experiment.The results indicate that the proposed algorithm can enhance the positioning accuracy of the system and reduce the convergence time of filtering and enhance the stability of the system.It can be used as an indoor positioning scheme. |
