Doppler Shift Measurement Using Complex-Valued CSI of WiFi in Corridors

Abstract

This paper presents a novel approach to measure indoor Doppler shift (which is generally neglected) based on Channel State Information (CSI) using WiFi. Our approach is motivated by the fact that the CSI provided by a commercially available measurement tool is complex-valued, and therefore the Fast Fourier Transform (FFT) of CSI is not conjugate-symmetric. After establishing a CSI acquisition system, we process CSI data guided by theoretical derivation under different corridor scenarios and finally obtain the Doppler shift of different relative motions. Our results demonstrate that this method of measuring Doppler shift could be applied to further research on crowd counting or human activity recognition based on WiFi.