Human Activity Classification with Frequency Modulated Continuous Wave Radar Using Deep Convolutional Neural Networks

Abstract

Radar is an attractive sensor for human activity classification due to its ability to work in low lighting conditions, its invariance to the environment and its ability to operate through obstacles. Human activity classification finds applications in human-computer interfaces, user-intent understanding and contextual-aware smart homes. Radar reflections from humans produce unique micro-Doppler and range signatures that can be used to recognize target movements. In this paper, we study such features obtained from a compact short-range 60-GHz frequency modulated continuous wave radar for various human activities. We present novel Convolutional Neural Network architectures to leverage these features for human activity classification.