Are you sitting right?-Sitting Posture Recognition Using RF Signals

Abstract

Today, sedentary behaviors and bad sitting postures are the main causes of modern health musculoskeletal disorders and illnesses. Previous works either used a camera to record the image or attached wearable sensors on human body to recognize sitting postures. However, video-base approaches may face privacy issue while the wearable sensor-based approaches may cause uncomfortable to the user. This paper introduces SitR, the first sitting posture recognition system using RF signals alone. We demonstrate that with just three tags pasted to one's back, SitR can successfully recognize three habitual sitting postures. Our design exploits the correlation between the phase change of RFID tags and the sitting postures. By extracting effective features from the measured phase sequences and employing machine learning algorithm, SitR can achieve robust and high performance. We evaluated SitR through extensive experiments including 14 volunteers under 3 different scenarios. The experiment results show that SitR can recognize sitting postures with an average accuracy of 99.27%. Our system can further detect the abnormal respiration and provide sitting posture history for sedentary people.