A system is described that enables a computing system (e.g., a mobile phone, a smartwatch, a tablet computer, etc.) to passively detect a user’s sleep duration. That is, without a user configuring the computing system into a sleep mode or otherwise inputting a sleep duration, the computing system may, after receiving explicit permission from the user, monitor various contextual signals to automatically determine the user’s sleep duration. The computing system may passively capture various data using sensors (e.g., accelerometers, ambient light sensors, microphones, etc.) in the computing system and analyze the captured data to estimate a user’s sleep duration. For example, the computing system may analyze accelerometer data to determine when a user is moving and how much the user is moving, analyze audio data captured by a microphone to determine if the audio captured is indicative of sleep, and/or analyze ambient light data to determine ambient light conditions. Such sensor data may be periodically generated and analyzed to generate sleep information for s series of time intervals. Based on the analysis of such sensor data, the computing system may determine whether the user was asleep when the computing system generated the sensor data. In some examples, the computing device may further classify sleep stages (e.g., rapid eye movement (REM) stage, light sleep stage, deep sleep stage, etc.) using the generated sensor data (e.g., classify sleep stages using user’s breathing rate, heart rates, or movement, etc.).
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Chen, Kathy; Gadh, Tajinder; and Stogaitis, Marc, "PASSIVE SLEEP DETECTION", Technical Disclosure Commons, (November 06, 2020)