Interference Suppression Based Gesture Recognition Method with FMCW Radar

Abstract

In this paper, we propose a gesture recognition method based on interference suppression with Frequency Modulation Continuous Wave (FMCW)radar. We analyze the sampling point and the time needed for snapshot of radar signal, and estimate the range and doppler parameter of gesture. Afterward, we design a complete background and target interference suppression method for the characteristics of radar signal interference, then use Time Sequential Inflated 3 Dimensions(TS- I3D)network for gesture feature extraction and classification. The experimental results show that the suppression method has effectively solved the problems caused by non-gesture interference, and the average accuracy rate of gesture recognition is 96.67 %.