Adaptive Nonlinear Least Squares Framework for Contactless Vital Sign Monitoring

The respiratory and heart rates are critical physiological parameters, conventional contact-based monitoring techniques may cause discomfort epidermal damage, being therefore inadequate for long-term monitoring. Despite recent advances, accurate contactless vital sign is still challenging in practical scenarios, especially relation to rate estimation. In this work, we propose a comprehensive framework processing frequency-modulated continuous-wave radar systems evaluate its performance with real data imitating common working conditions an office environment. First, improve the signal-to-noise ratio before estimation, novel slow-time phase correlation processing, which allows early integration of energy at nearby range bins. Subsequently, present adaptive nonlinear least squares that explores harmonic structure existing recovered displacement signal. An additional Kalman filter stage designed select among multiple estimates from different search regions, thus conferring adaptivity robustness against interference noise. This approach largely provides within predefined error intervals, capable tracking true breathing values even during continuous small body movements. final accuracy root mean square have shown enhanced outperforming spectral estimation other recently proposed methods almost all scenarios.