mmWave Frequency Modulated Continuous Wave Radar-based Human Action Recognition

Abstract

Human motion recognition is a field where human activities are extracted from raw time-series signals obtained through sensors. Due to privacy concerns or personal comfort issues, recently, radar technology is being used as an alternative. In this study, a human motion recognition model utilizing mmWave radar data collected from individuals has been developed. The proposed model utilizes micro-Doppler and range-Doppler spectrograms as input. The results obtained successfully demonstrate the model's ability to accurately identify movements such as arm raising, walking and stand up in real-time for healthy individuals.