Abstract
One of the authors previously developed an unconstrained and noninvasive measurement system of heartbeat and respiration using an acoustic sensor enclosed in an air pillow or an air mat. The system measured instantaneous heartbeat and respiration periods by applying Kalman filters and maximum likelihood method to the dynamic models of the heartbeat and respiration signals extracted from the sensor output. To improve the respiration measurement accuracy especially in a weak respiration case, the paper proposes a measurement system taking into account not only the fundamental mode of the heartbeat, but also the harmonic component of the heartbeat and the two side band components of the heartbeat signal modulated by the respiration.
