Author(s)

Yanxing Liu, Xiaoqin Li, Ran Tian

Corresponding Author

Xiaoqin Li

ABSTRACT

In order to improve the indoor positioning accuracy and stability of WiFi received signal strength indicating (RSSI), we propose a channel state information (CSI) indoor fingerprint location algorithm based on adaptive Kalman filter in this paper. In the offline stage, the original data is filtered by adaptive Kalman filter algorithm with variance compensation, and then the filtered data is classified by binary K-means clustering algorithm. Then, according to the off-line data and real-time data of the point to be measured, the K nearest neighbor matching algorithm is used to estimate the location coordinates of the location point in the online stage. Finally, simulation and field experiments show that the algorithm can effectively reduce the influence of multi-path attenuation at the receiving end of the signal by using the amplitude characteristics of the corresponding subcarriers in the channel, and the positioning accuracy reaches 0.7m. In addition, the positioning results are efficient and effective.

KEYWORDS

Indoor localization, fingerprint localization, feature fingerprint, channel state information, bisecting k-means clustering algorithm