Research On Indoor Positioning Technology Based On Wireless Channel State Information

With the continuous development of the Internet of Things and mobile devices,people put forward urgent requirements for higher positioning accuracy.Global Positioning System(GPS)which can provide high-precision positioning service outdoors is not suitable for indoor positioning because the satellite signal is blocked by the wall,Therefore,it is of great significance to study positioning technology suitable for indoor positioning.WIFI indoor positioning technology with the characteristics of convenience,economy and easy deployment stands out among many indoor positioning technologies,and has become a research hotspot nowadays.In addition,with the successful acquisition of Channel State Information(CSI)in physical layer in WIFI devices,it overcomes many shortcomings of indoor WIFI location technology based on RSSI.Therefore,it is of great practical significance to explore the indoor positioning technology based on CSI.In this paper,the indoor positioning technology based on CSI is taken as the research object,and the existing indoor positioning technology based on CSI is summarized and analyzed comprehensively.The Cramer-Rao Lower bound of indoor positioning error based on CSI is obtained.The indoor positioning is carried out using CSI in the LOS/NLOS scenario.The specific innovations are as follows:(1)In ideal Line-Of-Sight non-multipath scenario,the Cramer-Rao Lower Bound of the ranging error and positioning error is obtained by using the modified path-loss free space propagation model proposed by predecessors,and in the case of range-of-sight multi-path scenario,it is proved theoretically that the ranging model proposed by predecessors has biased estimation,and the Cramer-Rao Lower Bound of the ranging error and positioning error can not be obtained.(2)In view of the fact that the previous ranging model can not find the Cramer-Rao Lower Bound in multipath scenarios,this paper proposes a logarithmic linear unbiased ranging model.The Cramer-Rao Lower Bound of ranging error and positioning error can be obtained by this model.Then,through simulation analysis,which factors are related to the Cramer-Rao Lower Bound of ranging error and positioning error obtained from logarithmic linear unbiased ranging model,which provides theoretical reference for reducing the Cramer-Rao Lower Bound and optimum positioning accuracy in practical application when positioning through the ranging model.(3)Based on the logarithmic linear ranging model,a new ranging method is proposed,and the accuracy of the ranging model is verified by experiments.The parameters of the logarithmic linear ranging model are estimated from the ranging sample data offline,and then the distance measurement is realized online based on the estimated ranging model.The experimental results show that the ranging accuracy of the model is higher than that of the existing CSI-based ranging.(4)This paper combines CSI amplitude ranging and phase angle to locate,which increases the amount of information available and improves the positioning accuracy.For CSI phase angle calculation,this paper proposes a phase pre-processing scheme,including phase extension and phase error elimination,and finally obtains the angle of arrival by improved MUSIC algorithm.The experimental results show that the positioning error is less than 0.8m in 90% cases,and the positioning accuracy is better than that of the traditional single AP positioning method.