Range Enhancement of a 60 GHz FMCW Heart Rate Radar Using Fabry–Perot Cavity Antenna

Abstract

This paper presents a bistatic 60 GHz frequency-modulated continuous-wave (FMCW) radar system for non-contact heart rate monitoring, utilizing high-gain Fabry–Perot cavity (FPC) antennas at both the transmitter and receiver. Each FPC antenna integrates a partially reflective surface (PRS) and a metallic ground plane to form a resonant cavity. Compared to conventional patch arrays of the same aperture, the FPC antenna improves the antenna gain from 4.1 dBi to 8.1 dBi at the transmitter and from 3.9 dBi to 7.8 dBi at the receiver, resulting in an overall link budget enhancement of approximately 7.9 dB. This dual high-gain configuration theoretically increases the maximum detection range by a factor of 2.48. The proposed radar system was implemented and experimentally validated under indoor conditions using both calibration targets and human participants. Active measurement results confirm that the bistatic radar equipped with FPC antennas extends the reliable heart rate detection distance by approximately 2.27 times compared to a conventional system, closely matching the theoretical prediction. These results confirm the practicality and effectiveness of FPC antennas in extending both the range and reliability of millimeter-wave vital sign detection systems.

1. Introduction

Non-contact vital sign monitoring has emerged as a key technology in healthcare, remote patient observation, and emergency response scenarios, offering advantages in convenience, hygiene, and patient comfort. Among various sensing approaches, millimeter-wave frequency-modulated continuous-wave (FMCW) radar has attracted significant attention due to its high spatial resolution, compact antenna size, and robustness to environmental variations [1,2,3,4,5,6,7,8,9]. In particular, 60 GHz radar systems enable sub-millimeter displacement sensitivity, making them well suited for detecting the minute chest wall motions induced by human heartbeat activity.

Despite these advantages, conventional FMCW vital sign radars face critical limitations in detection range. The micro-displacement signals corresponding to heartbeats typically exhibit amplitudes on the order of hundreds of micrometers, resulting in extremely low radar cross-section (RCS) values. As a result, the received signal power decreases rapidly with increasing distance, leading to a sharp drop in signal-to-noise ratio (SNR) and reduced measurement reliability. Although advanced signal processing techniques and higher transmit power have been explored to mitigate this problem, these approaches alone often fail to achieve significant range extension, especially under regulatory limits on equivalent isotropically radiated power (EIRP).

One of the promising candidates for improving the radar link budget is the use of high-gain antennas. Various approaches have been proposed, including transmitarray antennas based on metasurfaces [10,11,12,13,14], which can achieve high gain but often require complex metallic pattern designs and precise fabrication. In contrast, Fabry–Perot cavity (FPC) antennas have drawn interest for their ability to provide narrow beams and substantial gain enhancement while maintaining a relatively low-profile structure and a simple structure [15,16,17,18]. The FPC antenna consists of a partially reflective surface (PRS) and a metallic ground plane, forming a resonant cavity that enhances forward radiation. For the same aperture size, FPC antennas can achieve higher gain compared to the conventional patch antennas, offering clear benefits for long-range vital sign detection such as heart rate.

Although previous works have investigated FPC antennas for remote vital sign detection at 60 GHz, they primarily focus on beam scanning or complex multi-layer structures [19,20]. This study proposes a bistatic 60 GHz FMCW radar system utilizing a fixed-beam, high-gain FPC antenna with a commercially fabricated PRS, offering a simpler design, ease of integration, and practical performance enhancement. Each FPC antenna achieves a gain improvement exceeding 3 dB compared to a reference 2 by 1 patch array, resulting in a combined link budget enhancement of about 8 dB. Experimental validation under indoor conditions demonstrates that this configuration more than doubles the maximum reliable heart rate detection range relative to the conventional system, while maintaining stable phase response and high SNR. Furthermore, the improved spatial selectivity of the FPC antennas effectively suppresses multipath-induced clutter, enhancing detection robustness in realistic environments.

While this study focuses on the experimental validation of bistatic FMCW radar with FPC antenna, it is important to place our approach in the context of other modern vital-sign monitoring techniques. Continuous-wave Doppler radars with beam scanning, MIMO FMCW systems with digital beamforming, and multi-antenna coherent combining have all demonstrated strong potential for improving robustness, range, and adaptability. Furthermore, the integration of advanced algorithms such as MUSIC/DR-MUSIC, wavelet decomposition, and HRV extraction provides significant additional performance gains. Compared with these approaches, the proposed method offers clear advantages in simplicity and hardware efficiency, providing immediate gain enhancement without introducing additional algorithmic or computational complexity. However, the trade-off is the absence of spatial diversity and adaptive processing capabilities available in multi-antenna and algorithm-driven systems.

The paper is organized as follows. Section 2 describes the design procedure and simulation results of the FPC antennas. Section 3 details experimental results and discussion of bistatic FMCW radar system. In Section 4, the concluding remarks are given.

2. Design of Fabry–Perot Cavity (FPC) Antenna

Figure 1 shows the FPC by a ground and PRS with excitation inside a cavity. The directivity (D) of the FPC antenna is given by [15]

$$D={\displaystyle \frac{1-{\Gamma }_{PRS}^2}{1+{\Gamma }_{PRS}^2-2{\Gamma }_{PRS}\mathit{cos}\left({\phi }_{PRS}+{\phi }_{GND}-\frac{4\pi }{\lambda }d\right)}}{f}^2$$

(1)

where Γ_PRS is the reflection magnitude of PRS and d is a distance between PRS and GND. ${\phi }_{PRS}$ and ${\phi }_{GND}$ are the reflection phase of the PRS and the ground plane, respectively. f is the radiation pattern of the source. From Equation (1), ${\phi }_{PRS}+{\phi }_{GND}-{\displaystyle \frac{4\pi }{\lambda }}d$ and Γ_PRS should be 2πN (N = integer) and close to 1, respectively, to achieve the maximum directivity. Then, the height (d) of the FPC antenna is calculated by Equation (2) and can be determined by the sum of reflection phases of the ground plane and the PRS.

$$d=\left({\phi }_{PRS}+{\phi }_{GND}\right){\displaystyle \frac{\lambda }{4\pi }}+N{\displaystyle \frac{\lambda }{2}}$$

(2)

In this study, we designed the FPC antenna to enhance the antenna gain of bistatic 60 GHz FMCW heart rate radar system without increasing the number of array elements. The reference antenna system consists of a 2 × 1 patch array antenna for both the transmitter (Tx) and receiver (Rx), as shown in Figure 2a. The antenna size is 3 mm × 4 mm, and an inset cut technique is utilized to achieve good impedance matching. The feed lines of both Tx and Rx antennas are connected bottom lines through via to connect with the RF circuit on the bottom layer. In addition, due to the opposing orientation of the antennas, a half-wavelength phase delay feed line was implemented in the Rx antenna array to ensure proper phase alignment, as shown in Figure 2a. As illustrated in Figure 2b, the full-wave simulated S-parameters clearly indicate that a resonance occurs at 60 GHz and the good isolation below—40 dB at the resonant frequency. The simulation results are obtained by ANSYS Desktop. To implement the antenna along with the RF circuit realized on the PCB, both antennas were fabricated on an FR4 substrate, which is known to have a relatively high loss tangent of 0.02. Due to this material characteristic, the realized gain of both the Tx and Rx antennas is limited to approximately 4 dBi at operation frequency, even though each antenna is implemented as a 2 × 1 patch array. To improve the antenna performance without expanding the physical size of the antenna or increasing the number of elements, we introduced the PRS with a dielectric constant of 4.3, forming an FPC antenna structure. The PRS was fabricated using a commercial PCB substrate (ZYF-430CA), which has a thickness of 1.6 mm and a loss tangent of 0.0035. The dimension of the PRS is 23 mm × 25 mm, which is the same as the size of the PCB board. Since the PRS has a reflection phase of −150° at 60 GHz, it was placed above the patch antenna with a spacing of 2.7 mm, which was calculated according to Equation (2). Moreover, since the PRS has a reflection coefficient of 0.6 at 60 GHz, a gain enhancement of approximately 6 dB can be theoretically and ideally predicted according to Equation (1). The reflection magnitude and phase of the PRS were obtained using ANSYS Desktop simulation, assuming normal incidence from the far field. The reflection coefficient of the PRS varies from 0.57 to 0.62 across the 58~62 GHz frequency range, while the reflection phase ranges from −144° to −156°. As indicated by Equation (1), an increase in the permittivity of the PRS leads to greater gain enhancement. Nevertheless, for ease of fabrication, a PRS with a relative permittivity of 4.3 was employed in this study. The effectiveness of the proposed FPC antenna design was validated through simulation. The S-parameters showed nearly the same characteristics as in the case without the PRS. Figure 3 shows the full-wave simulated far-field radiation patterns of the Tx and Rx patch arrays without and with PRS. The results clearly demonstrate a significant gain improvement of approximately 4 dB over the conventional patch array, while maintaining acceptable impedance matching.

Table 1. Frequency-dependent gain enhancement against chirp frequencies.

Gain Enhancement	58 GHz	59 GHz	60 GHz	61 GHz	62 GHz
Tx antenna	3.9 dB	4.2 dB	4 dB	3.7 dB	3.5 dB
Rx antenna	4.3 dB	4.1 dB	3.9 dB	3.8 dB	4 dB

summarizes the frequency-dependent gain enhancement, confirming that the FPC maintains high performance throughout the chirp bandwidth.

3. Experimental Results and Discussion of Bistatic FMCW Radar System

Figure 4 presents a block diagram of bistatic FMCW radar system designed for non-contact detection of heartbeat and respiration. The system operates by transmitting a frequency-modulated chirp signal f_TX(t) from a TX antenna towards the chest. The chest’s micro-movements, due to cardiac and respiratory activity, modulate the reflected signal. In the bistatic configuration, the transmitted and received signals are handled by separate antennas, allowing spatial separation between the TX and RX modules. The signal generator produces a linear chirp signal, which is amplified by a power amplifier before transmission through the TX antenna. The reflected signal f_RX(t), modulated by the displacement caused by vital signs, is captured by the RX antenna. It then passes through low noise amplifier, and is mixed with a reference copy of the transmitted signal using a mixer to obtain the intermediate frequency (IF) signal. This IF signal contains phase and frequency shifts corresponding to the target’s motion. Subsequently, the IF signal is passed through a filter to suppress noise and out-of-band components, and then amplified via a gain amplifier. The analog signal is converted to digital using an analog-to-digital converter (ADC). Finally, the digital signal is processed using a digital signal processing (DSP) unit, which extracts the heartbeat and respiration rates by analyzing periodic patterns in the phase or frequency domain. Figure 5 shows the FMCW radar module for real-time heartbeat detection. In addition, Figure 5b presents the FPC Tx and Rx antennas with a 2.7 mm spacing between the PRS with a dielectric constant of 4.3 and the ground. Figure 6 is the measurement setup of the FMCW radar system for non-contact heartbeat detection. The radar module is oriented toward the human target, which seated at a distance in front of the radar system. In addition, the height of the radar module was adjusted to be at the same level as the subject’s heart, in order to maximize signal reflection. A total of three human subjects participated in the experiment and the transmit power of the radar system is set to 12 dBm. The measurement environment was arranged to minimize clutter reflectors, with no large metallic or reflective surfaces present within the radar’s field of view. The maximum detection distance was evaluated in the configurations without and with the PRS, and the results were compared to analyze their characteristics. According to the Friis transmission Equation (3), the received power (P_r) at a distance (d) can be expressed.

$$P_r=P_t{G}_t{G}_r{\left({\displaystyle \frac{\lambda }{4\pi d}}\right)}^2$$

(3)

where P_t is the transmitted power, G_t and G_r are the gains of the transmitting and receiving antennas, respectively. In our simulation results, the transmit and receive antenna gains were increased from 4.1 dBi to 8.1 dBi and from 3.9 dBi to 7.8 dBi at 60 GHz, respectively, by employing FPC antennas with PRS. This corresponds to an approximate 4 dB gain enhancement per antenna, effectively more than doubling the linear gain factor of each. Since the received power is directly proportional to the product of G_t and G_r, the overall link gain was improved by nearly 8 dB, equating to a fourfold increase in the total link budget. According to the Friis transmission equation, the maximum detectable distance (d_max) is proportional to the square root of the received power. Based on this, the theoretical detection range is expected to increase by a factor of approximately 2.48. Experimental validation closely matched this prediction, as the actual measured detection range improved by approximately 2.27 times after implementing the FPC antennas compared to the conventional system, increasing from 1.1 m to 2.5 m. To quantitatively support the claimed improvement in detection range,

Table 2. Heartbeat detection success at different distances with and without FPC antennas.

Distance (m)	2 × 1 Patch Array (Success Rate)	FPC Antenna (Success Rate)
0.5	Detected (100%)	Detected (100%)
1	Detected (100%)	Detected (100%)
1.5	Not detected (8%)	Detected (100%)
2	Not detected (0%)	Detected (100%)
2.5	Not detected (0%)	Detected (100%)

summarizes the heartbeat detection success at various distances for both the conventional antenna and the proposed FPC antenna configuration. A heartbeat signal was regarded as successfully detected when the reconstructed waveform exhibited a clear periodic pattern corresponding to cardiac motion, and the signal quality was sufficient to allow reliable heart rate estimation. As shown, the conventional antenna failed to detect heartbeat signals beyond 1.1 m, whereas the FPC antenna enabled successful detection up to 2.5 m. These results correspond with the observed waveform presence in Figure 7 and confirm the practical range extension enabled by the FPC structure. This result clearly demonstrates the strong dependence of radar performance on antenna gain, confirming the effectiveness of FPC antennas in extending the system’s operational range.

4. Conclusions

In this paper, a bistatic 60 GHz FMCW radar system incorporating high-gain FPC antennas was proposed and experimentally demonstrated for non-contact heart rate monitoring. By utilizing PRS and ground plane, the FPC antennas significantly enhanced the overall link performance compared to conventional patch arrays. Experimental results verified that the proposed configuration effectively extends the reliable detection range while maintaining stable phase response and strong signal quality. Additionally, the improved spatial selectivity of the FPC antennas contributed to the suppression of multipath interference, resulting in enhanced robustness under realistic indoor conditions. These findings suggest that FPC antennas offer a practical and effective solution for improving the performance of millimeter-wave vital sign monitoring systems.