A Single Node High Precision Indoor Localization Method Based On Channel State Information

With the rapid popularization and development of mobile Internet,indoor location has become a hot spot.Because the indoor wireless environment is complex,the traditional fingerprint-based indoor localization system need a large number of access points to achieve satisfactory results.In order to solve this problem,this thesis analyzes and discusses the possibility of using single access point to build indoor localization system,designs and develops the error elimination algorithm and location algorithm based on single node localization system,also build testing system in the actual indoor environment for practical testing and simulation verification.Firstly,this thesis analyzes the model of the traditional fingerprint indoor localization system and the typical single-node indoor localization system,discusses the advantages and disadvantages of the above two systems.According to the past researches,the positioning method of using channel state information is introduced in this thesis to achieve higher precision localization system.According to the theoretical analysis of the channel state information,this thesis analyzed the possibility and advantages of the channel state information applied to indoor positioning system.And then,the phase error model of the channel state information is established by analyzing the statistical distribution of the received signals.Based on the model,a new error elimination algorithm is proposed in this thesis.Secondly,this thesis analyzes and demonstrates the method of enhancing the resolution of multi-signal classification algorithm,and designs an improved spatial spectrum estimation algorithm by frequency hopping and spatial smoothing.The new Chinese residual theorem algorithm combined with the power delay distribution is proposed to solve the high precision propagation distance.Also,this thesis designs new localization methods based on triangle law and Taylor series,and drives the Cramer-Rao lower bound(CRLB)of positioning error when the measured distance error is normal distribution.Finally,aiming to evaluate the performance of the proposed algorithms,this thesis designs and implements a single node localization system on Linux platform in the actual environment,performs the simulation and testing of the proposed algorithms in different demonstration environment.Experimental results demonstrate that the distribution of cumulative error is 67.5% within 0.6-meter locating error.It denotes the algorithms proposed in this thesis fulfill the high precision localization requirement.