Abstract:Sleep monitoring plays an important role in many medical applications, including SIDS prevention, care of patients with pressure ulcers, and assistance to patients with sleep apnea, where studies have shown that autonomous and continuous monitoring of sleep postures provides useful information for lowering health risk. Existing systems are designed based on electrocardiogram, cameras and pressure sensors, which are expensive to deploy, intrusive to privacy, or uncomfortable to use. This paper presents TagSheet, the first sleep monitoring system based on passive RFID tags, which provides a convenient, non-intrusive, and comfortable way of monitoring the sleeping postures. It does not require attaching any tag directly to a patient’s body. Tags are taped under a bed sheet. With a combination of hierarchical recognition, image processing and polynomial fitting, the proposed system identifies body postures based on the observed variation caused by the patient body to the backscattered signals from tags. The system does not require any personalized data training, making it plug-n-play in use. One additional advantage is that the system can also estimate the patient’s respiration rate. This is particularly helpful in assisting patients with sleep apnea. We have implemented a prototype system, and experiments show that the system performs posture identification with an accuracy up to 96.7% and in the meantime it measures the respiration rate with a small error of about 0.7 bpm (breath per minute).

TagSheet: Sleeping Posture Recognition with an Unobtrusive Passive Tag Matrix