Contact-less indoor activity analysis using first-reflection echolocation

• Published in: IEEE International Conference on Communications (2003) pp. 1 - 6
• Main Authors: Biswas, Subir, Harrington, Brandon, Hajiaghajani, Faezeh, Rui Wang
• Format: Conference Proceeding Journal Article
• Language: English
• Published: IEEE 01.05.2016
• Subjects: Acoustics; Activity Classification; Biomedical monitoring; Classifiers; Computation; Doppler effect; Echo-located Distance Computation; Electronics; First-reflected Echolocation; Human Activity Monitoring; Monitoring; Moving object recognition; Sensor arrays; Signal analysis; Ultrasonic imaging; Ultrasound; Ultrasound Echolocation; Video data; Walking
• ISSN: 1938-1883
• DOI: 10.1109/ICC.2016.7510731

This paper presents an ultrasound echolocation-based approach for human activity recognition in indoor settings. The key novelty of the proposed approach is to perform activity analysis using distance estimated through "first-reflection echolocation". The distance to the nearest obstructing object is computed using the first reflected ultrasound signal. All subsequent reflected signal components from other distant objects are ignored. This leads to an extremely simple signal (i.e., time-series distance data) analysis approach with very low computational complexity. Especially so, when compared with the existing approaches in literature in which full reflected signal analysis, often with Doppler Shift computation, is performed for activity classification. It is demonstrated that for the goal of isolating workplace sedentary behavior, the proposed approach can differentiate between sitting, standing, and walking (i.e., in-office pacing) with more than 80% accuracy. This was validated with different classifiers applied on data collected from multiple subjects in multiple sessions. Recorded video was used as the ground-truth for training the classifiers.