Millimeter-wave radar interferometry is superior in detecting small displacement motions owing to its short wavelength. However, it is subject to phase ambiguity as the target displacement may often exceed a quarter wavelength. In this article, a modified differentiate and cross-multiply (MDACM) technique is proposed to tackle the phase ambiguity issue for accurate reconstruction of the phase history in millimeter-wave interferometry. In addition, to resolve the imperfections of the interferometric radar system, an MDACM-based integral phase reconstruction approach is presented, which seamlessly integrates alternating current (ac)-coupling-induced distortion correction, $I/Q$ mismatch correction, and direct current (dc) offsets’ calibration. Without causing any phase ambiguity, the proposed technique acts as a black box to take in the raw $I/Q$ signals and correct all the hardware imperfections, and it outputs the desired displacement motions with micrometer accuracy. The simulation results show that the proposed technique can not only linearly recover the displacement motions across a wide range in different noise conditions without any phase ambiguity but also improve the stability by 11 times under ac-coupling-induced distortion. With a custom-designed 120-GHz interferometric radar sensor, experiments were carried out to validate various scenarios including mechanical vibrations and gesture sensing. The experimental results show that the proposed technique can accurately track not only deep subwavelength motion of only $1~\mu \text{m}$ but also multiwavelength displacement of >40 times of the wavelength at 120 GHz.

Vital sign detection using linear frequency-modulated continuous-wave (LFMCW) radar may be subject to the proximity stationary clutters. This paper presents a novel technique to synthesize the slow-time I / Q signals, which are equivalent to those in a single tone quadrature CW radar, from a single-channel LFMCW radar. It correlates the two types of radars in such a way that the proximity stationary clutters are translated to direct current (DC) offsets in the synthesized I / Q signals across slow-time. The circle-fitting based DC offsets calibration (DCcal) technique, which was developed for CW radar, can now be applied to eliminate the impact of the proximity stationary clutters in LFMCW radars for accurate vital sign detection. Moreover, the modified differentiate and cross-multiply (MDACM) algorithm can also be leveraged to eliminate the phase ambiguity issue. Thorough theoretical analysis and working principles are presented. Simulations are performed to validate the proposed technique. Moreover, exhaustive experiments are carried out with a millimeter-wave 79 GHz FMCW radar in the office environment. Mechanical vibration and vital signs are extracted with micrometer-level accuracy in the existence of proximity stationary clutters.

A novel 4-D gesture sensing technique using reconfigurable virtual array method with a 60-GHz frequency-modulated continuous wave (FMCW) radar is presented. The 4-D sensing includes the 3-D spatial positioning and the 1-D motion tracking, which combines detection in both spatial and temporal domains. The proposed technique can be used for gesture sensing for both macro gestures such as handwriting letters and numbers, hand drawing different patterns in space and micro gestures such as the fingers’ actions of virtual slider, finger tap, and finger wave. Moreover, this work presents a method to reconfigure the working modes of different transmitting and receiving channels in the multiple input and multiple output (MIMO) architecture, which effectively increases the sensing range of the radar sensor and the signal-to-noise ratio (SNR) of the beat signal. For various gestures at different distances or in different environments, the proposed technique can adaptively achieve the best detection by reconstructing the virtual array of the MIMO radar. A series of experiments are carried out in both microwave anechoic chamber and office environments to validate the proposed technique. The experimental results show that the proposed technique performs well in sensing a variety of both macro gestures andmillimeter-level micro gestures. Compared with the conventional MIMO method, the proposed technique increases the sensing range by 3.5 times, and the SNR is also greatly improved by 12 dB when the radar performs gesture sensing at the distance of 0.65 m.

This article presents a novel 5.8-GHz ISM-band dual-circularly polarized (CP) simultaneous transmit and receive (STAR) common-aperture antenna array (CAAA) for integrated sensing and communication (ISAC) in Internet of Things (IoT). A new near-field cancellation technique based on differential feeding and geometrical symmetry is proposed for dual-CP to realize high TX/RX isolation. In addition, a high-impedance surface is introduced to further improve the TX/RX isolation. Furthermore, the parasitic patches are added to enhance the gain and isolation bandwidth (BW). The simulation results demonstrate that the proposed CAAA possesses the merits of high isolation (up to 67 dB), wide isolation BW (better than 40 dB for 240 MHz BW), high gains (up to 10/10.5 dB for right-hand CP (RHCP)/left-hand CP (LHCP)), wide 1-dB gain BWs (better than 230/350 MHz BW for RHCP/LHCP), and low-axial ratios (lower than 1.4 dB). The designed CAAAs are fully characterized and the experimental results meet the simulation very well. For the first time, systematic validations for both wireless communication and radar sensing have been thoroughly carried out by integrating the proposed CAAA in the custom-designed full-duplex radio system. The experimental results not only prove the validity of the design methodology but also the feasibility of employing the proposed CAAA for ISAC, which indicates potential applications in IoT era.

The field of biomedical radar has witnessed significant advancements in recent years, paving the way for innovative and transformative applications in clinical settings. Most medical instruments invented to measure human activities rely on contact electrodes, causing discomfort. Thanks to its non-invasive nature, biomedical radar is particularly valuable for clinical applications. A significant portion of the review discusses improvements in radar hardware, with a focus on miniaturization, increased resolution, and enhanced sensitivity. Then, this paper also delves into the signal processing and machine learning techniques tailored for radar data. This review will explore the recent breakthroughs and applications of biomedical radar technology, shedding light on its transformative potential in shaping the future of clinical diagnostics, patient and elderly care, and healthcare innovation.

Cardiogram is one of the most important factors for health assessment. As a new expression of cardiograms, the Doppler cardiograms (DCGs) detected remotely via Doppler radar sensor (DRS) provide rich details of the heart motion. However, the existing techniques for measurement of DCG requires that the subject holds his/her breath to avoid the disturbance caused by respiration. To realize the accurate detection of DCGs in the presence of respiration, a high-linearity $K$ -band dc-coupled continuous-wave (CW) DRS with digital dc-tuning function was custom-designed and a novel parameterized respiratory filter (PRF) algorithm was proposed to precisely remove the respiratory signal. Both simulation and experimental results show that the proposed PRF algorithm has better DCG extraction performance than the conventional methods. Furthermore, the experimental results in normal and clinical environments show that the grained differential DCG (D-DCG) has promising potential in professional heart rate variability (HRV) analysis and heart diseases diagnosis, which means the proposed DCGs detection has the potential to be a convenient, comfortable, and reliable way for daily and clinical heart health assessment.