Research On Indoor Positioning Method Based On Multi-source Information Fusion

With the rapid development of the internet of things,the demand for location-based services is growing.The application of wireless positioning technology in navigation,medical,logistics and other fields has produced huge economic benefits,so it has broad prospects.With the development of global positioning system(GPS)and Beidou navigation satellite(BDS),outdoor positioning technology can achieve sub-meter level positioning accuracy.However,in the indoor area such as mine,warehouse,shopping malls,the satellite signal is obscured,resulting in positioning technology based on global navigation satellite system(GNSS)can not be applied in the indoor environment,so indoor positioning technology has become a research hotspot.Due to the drawbacks of indoor positioning methods based on a single positioning source,it is difficult to balance accuracy and cost.Multiple positioning information has a certain complementarity,so fusion positioning based on multi-source information is an effective way to improve positioning performance.Based on wireless fidelity(Wi-Fi),magnetic field(MF),acceleration and angular velocity information,combined with positioning related principles and data analysis methods,this paper optimized the indoor positioning program by theoretical research and experimental testing.The main work include:(1)Aiming at the fuzzy matching situation of magnetic positioning,a method used to correct the magnetic positioning result based on the evaluation of matching ambiguity is proposed,which effectively reduces the probability of mismatching.In addition,a Wi-Fi/magnetic fusion positioning method is proposed based on the magnetic correction positioning result and the concept of extended candidate location set.Experiment results show that this method uses the corrected magnetic positioning results to select high-quality estimated location in the extended candidate locations constructed by Wi-Fi positioning,which significantly improves the positioning accuracy in area with small size.(2)The magnetic adaptability of differrnt indoor sub-regions is different,the influence on the fusion positioning is analyzed.Based on the analysis,an adaptive positioning method considering the magnetic adaptability is proposed.This method is a hierarchical positioning method.The whole positioning area is divided into several sub-regions by clustering method and the magnetic suitability of different sub-regions is judged separately.In the online positioning phase,rough regional positioning is performed firstly.According to the magnetic adaptability of the region located in,the fusion positioning strategy is adjusted flexibly.Static positioning experiment results show that the optimized adaptive method improves the positioning accuracy by 5.6%.(3)The three links of gait detection,step length estimation and orientation estimation in the pedestrain dead reckoning(PDR)method were studied by experiments.Based on the adaptive filter,a multi-source information fusion positioning method consisting of the proposed method and PDR formed.Experiment results show that the positioning result of this method has no obvious position jump,and no error accumulation phenomenon.The average positioning accuracy reaches1.09 m,and the probability of positioning error within 2m reaches 86.2%.The paper contains 41 figures,12 tables and 82 references.