An indoor geolocation algorithm based on CSI and affine propagation clustering

Abstract

An indoor geolocation algorithm based on WLAN CSI and affine propagation clustering is proposed in this paper. In this algorithm, the wireless network adaptor Intel 5300 and 802.11n CSI Tools is used to collect the CSI data, and the PCA algorithm is used to reduce the dimension of the CSI data from 30 to 3, based on which the CSI fingerprint database is established. And the affine propagation clustering algorithm is used to group the reference nodes into clusters and then the WKNN is adopted to realize precise indoor geolocation. Experiment results show that the algorithm in this paper has a better performance in the location accuracy, 74% of the testing results has the accuracy of 1m, and 95% of the testing results has the accuracy of 2m.