An Evaluation of Unobtrusive Sensing in a Healthcare Case Study

This paper examines the integration of Human-in-the-Loop Cyber-Physical Systems (HiTL-CPS) and Unobtrusive Sensing through a case study named iFriend. Our approach enhances the data acquisition phase of HiTLCPS by integrating unobtrusive sensing techniques to monitor real-time heart rate and breathing rate. This is achieved by leveraging Channel State Information (CSI) of Wi-Fi signals, specifically focusing on its amplitude information. This integration facilitates seamless interaction between humans and the cyber-physical environment. We detail the architecture of the iFriend system, comprising sensors, actuators, and computational units forming a closed-loop control mechanism. The unobtrusive sensing module is specifically designed to capture physiological changes without causing discomfort or interfering with daily activities, making it well-suited for healthcare applications and human-machine interaction. We assess iFriend in an experimental setting, demonstrating its feasibility, with between 80% and 90% of estimates hovering around 2.5 breaths per minute for BR or 10 beats per minute for HR, respectively.