Ultrasonic circular array sensing for human arm motion classification

Abstract

Classification of human arm movements is an important problem in healthcare and well-being applications. In this paper, an ultrasonic Uniform Circular Array (UCA) Doppler sensing method is proposed for classifying arm movements from a given set. The method uses velocity and angular information, derived from Doppler frequencies and direction-of-arrival (DoA) by processing the signals received at the UCA, and employs a Bayesian classifier to distinguish between movements. The performance of the sensing method is evaluated using experimental datasets. The proposed ultrasonic UCA Doppler sensor and processing methods provide a compact solution for human arm movement classification.