Calibration-free Localization Based On Fingerprinting Of Channel State Information

Location information plays a vital role in today’s intelligent IoT era.In-door localization-based services have penetrated into many aspects of mod-ern life,for example,indoor navigation in large places such as shopping malls,airports,etc.can quickly provide location guidance for users,and real-time localization technology in unmanned warehouses can more effec-tively implement automated management,and by obtaining the real-time lo-cation information of visitors in the exhibition,it is possible to analyze the degree of visitors’preference for each work.GPS can only provide local-ization service on the outdoors with accuracy of meters,and indoor local-ization require high accury.Therefore,in the exhibition hall,the error of the meter level will locate the user near the adjacent works,leading to the fallacious analysis results.For indoor localization applications in common scenarios,the localization system should be easy to deploy and low in cost.Wi-Fi is widely deployed in modern offices and homes,so it is considered to be the most promising solution,and rapid localization can be achieved using existing ubiquitous Wi-Fi infrastructure.Despite decimeter-level ac-curacy,current localization systems require labor-intensive site survey or the deployment of multiple Access Points(APs)and antennas.For this purpose to achieve a high-precision,easy-to-deploy,and low cost indoor localization system,this paper proposes a calibration-free chan-nel state fingerprint localization system that uses 2 to 3 commercial Wi-Fi infrastructures and requires only 3 antennas to achieve decimetre-level lo-calization accuracy.Therefore,this paper first analyzes the influence of the inter-antenna distance on the localization performance of the system,pro-poses a constellation diagram of the phase difference of the channel state information and the angle of arrival,and proposes a localization perfor-mance assessment method.As a result,this paper proposes a calibration-free channel state fingerprint localization system,then promote localization performance via hybrid inter-antenna distance,context-aware fingerprinting recognition and a dynamic update algorithm of the fingerprint database.Fi-nally,this article built a system prototype based on the commercial Wi-Fi infrastructure Intel 5300 NIC,and conducted comprehensive and sufficient experiments to verify the system.The experimental results show that the calibration-free channel state fingerprint localization system proposed in this paper can realize the 80%errors within 0.2m,which is more accurate than the existing designs.