From Tumor Targeting to Speech Monitoring: Accurate Respiratory Monitoring Using Medical Continuous-Wave Radar Sensors

Abstract

Using microwaves to detect small physiological movements such as respiration and heartbeat dates back to the 1970s [1]. It is realized by detecting the phase information in the received radar signals, which is caused by Doppler shift due to the moving chest wall. The principle is similar to the radar guns used by police officers to detect over-speed vehicles. Based on the form of the transmit signal, there are basically two types of radars: continuous-wave (CW) radar and ultrawideband (UWB) radar. The CW radar falls into three subcategories: single-tone, stepped frequency (SFCW), and frequency-modulated CW (FMCW). Each category of radars has its specific advantages. The singletone CW radar has a simple system architecture that allows high-level chip integration [2]?[4]. It also has high accuracy (submillimeter) in relative displacement measurement [5]?[6]. Unfortunately, because no instant bandwidth is transmitted, single-tone CW radars do not carry range (i.e., absolute distance between the radar and the subject) information. FMCW radars are able to detect range information [7]?[8] but normally require a very large bandwidth and more sophisticated signal processing to realize high-accuracy relative displacement measurement. Researchers also have successfully integrated the FMCW radar on silicon chips [9]?[11]. SFCW radars carry some advantages of both singletone CW radars and FMCW radars and thus have been successfully used in applications such as fall detection [23]. In addition, a hybrid radar system combining the advantages of the single-tone and FMCW radars was reported in [12]. UWB biomedical radars have veryhigh-range resolution due to its wideband nature [13]. The state of the art shows that UWB pulse radars have been efficiently implemented on silicon [74] and have been successfully applied to the accurate detection of respiratory rate and apnea in adults and infants [75].