Research And Realization Of High Precision Indoor Positioning Method

Accurate and real-time indoor positioning has great market value and important application prospects.Due to the GPS in the indoor environment affected by many factors,so it can't provide high-precision positioning.How to accurately locate in the indoor environment has become a hot spot of research and application.With the wide deployment of wireless local area network(WLAN),the existing wireless network is used as the basic equipment,so that the indoor positioning can achieve precise positioning without any additional equipment.In recent years,researchers have adopted CSI to positioning,which a more fine-grained than RSSI.In this paper,a indoor localization method based on CSI and sparse representation is proposed.The system consists of two stages,offline and online.offline is used to collect CSI data.Online is used to position.All of the two stages of the CSI is collected in advance,to remove some of the unstable and bad data,to ensure the effectiveness of CSI.In order to improve the online positioning rate and reduce the computational complexity of the algorithm,the CSI fingerprint training step is added after the offline CSI.The positioning scheme combined with fingerprint matching method makes up the shortage of the previous indoor positioning scheme.The sparse representation algorithm is used to weaken the influence of the noise and unstable variables in the CSI fingerprint.Therefore,it improves the positioning accuracy.In this paper,it uses the common route,the computer system and the wireless network card driver that is customized to build the experimental environment.Finally,we use MATLAB to simulate the scheme.The test results show that the algorithm can effectively realize the indoor positioning without the need of other specific devices.The average accuracy is 0.12m,the accuracy can reach 91%.