Non-Contact Instantaneous Heart Rate Extraction System Using 24-GHz Microwave Doppler Sensor

High Accuracy Heartbeat Detection from CW-Doppler Radar Using Singular Value Decomposition and Matched Filter

“…Hence, researchers have proposed various approaches using more complex signal processing methods [ 24 , 25 , 26 , 27 , 28 ]. Petrovic et al [ 24 ] focused on the harmonic vibration of the chest surface and proposed extracting the harmonic signal of heartbeats from the vibration of a chest surface acquired using a CW-DRS with a bandpass filter.…”

“…To extract heartbeats from chest vibration using MF, a template of the heartbeat must be prepared. Many researchers proposed algorithms to adaptively generate a template from the output of radar [ 26 , 27 , 28 ]. In other studies [ 26 , 27 ], the template was created using polynomial approximation to remove breathing and body motion from measurement signals extracted in a 2 s time window.…”

“…Many researchers proposed algorithms to adaptively generate a template from the output of radar [ 26 , 27 , 28 ]. In other studies [ 26 , 27 ], the template was created using polynomial approximation to remove breathing and body motion from measurement signals extracted in a 2 s time window. Lv et al [ 26 ] evaluated the effectiveness of the proposed method through simulations and actual measurements and successfully eliminated the effects of harmonics of breathing and body motion.…”

“…However, they remarked that this method requires a 2 s visual selection, where no body motion is present, to enable a template to be created. Izumi et al [ 27 ] proposed a technique that combines an MF with the Burg method, in which the time–frequency characteristics of measurement signals are measured. This method requires a high sampling frequency and sufficient data length to generate a template in the time–frequency domain.…”

Analysis on frequency-dependency of conductive signal transmission channel for wearable devices

“…In particular, sensor network technology has recently been applied to human body related fields, and is being developed as a machine to machine (M2M) human body sensor network technology in which sensors are disposed around a human body to provide information on the health status of users. U-healthcare devices have been developed that use a patch-type device such as AirBit to measure the exact pulse and heart rate and transmit it to a smart device using a wireless channel such as Bluetooth or Wi-Fi [6,7,8,9,10]. As these various wearable healthcare sensor devices are developed, various communication methods for optimum data transfer among these devices are being studied.…”