Radar-Based Heart Rate Sensing on the Smart Glasses
Wei Huang, Paurakh L. Rajbhandary, Sam Shiu, John S. Ho, Jiang Zhu, Ben Wilson, Ye Geng
Abstract
Heart rate serves as an important indicator for evaluating health and fitness. This letter addresses the constraints of skin contact sensors by introducing an innovative noncontact heart rate monitoring approach. The method involves a wearable radar-based system seamlessly integrated into smart glasses, marking the first instance of such integration without the need for consistent skin contact, free from biases related to skin tone, ambient light interference, and hair or fabric occlusion. Radar signals preprocessed through resampling, filtering, and heart rate estimation methods are employed to extract the heart rate from vital signals. The proposed radar-based system is optimized for the best arterial signal detection while compensating for motion through an inertial measurement unit (IMU). The system’s effectiveness is confirmed through comprehensive testing involving 85 individuals, achieving a remarkable 90 <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\%$</tex-math> </inline-formula> success rate defined as mean absolute percentage error (MAPE) compared to reference device within <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${\pm}3\%$</tex-math> </inline-formula> . Test results and analyses underscore the feasibility of integrating heart rate monitoring into head-mounted devices (HMDs), laying the foundation for an enhanced user experience in future AR/VR technologies.