DIRECT RF SIGNAL PROCESSING FOR HEART-RATE MONITORING USING UWB IMPULSE RADAR

Abstract

Methods, systems, and devices for direct radio frequency (RF) signal processing for heart rate (HR) monitoring using ultra-wide band (UWB) impulse radar are presented. A radar sensor is able to directly sample a received signal at RF which satisfies the Nyquist sampling rate, preserving a subject's vital sign information in the received signal. The vital sign information can be extracted directly from a raw RF signal and thus down conversion to a complex baseband is not required. The HR monitoring performance from the proposed direct RF signal processing technique provides an improvement in continuous HR monitoring as compared against existing methods using a complex baseband signal and/or other measurement techniques.

Description

FIELD OF THE DISCLOSURE

This application is related to monitoring of vital signals, and in particular to improved systems and methods for continuously monitoring a heart rate of a subject.

BACKGROUND

Heart disease remains the number one cause of death in the United States, according to recent statistics published by the American Heart Association. When it comes to early diagnosis and prevention of cardiovascular diseases, there are few measures more helpful than a subject's resting heart rate (HR). HR is also one of the most important vital sign parameters that can be used to inspect other potential health issues. For example, HR monitoring is important for elderly care, in-hospital monitoring, baby monitoring, and so on.

With the advancement of mobile technologies, increasing numbers of smart devices enter a consumer's daily life, including low-cost smart devices (e.g., smart phones, fitness trackers, smart wearable devices, Bluetooth electrocardiogram (ECG) sensors). Many of these devices can conveniently provide important physiological data, including HR. Not only can a normal resting HR vary from subject to subject, but if tracked over time it can also elicit some important data regarding a subject's heart health and fitness. For example, studies have shown that a rise in the resting HR of a subject can be linked to inadequate exercise, increased stress, obesity, and tobacco consumption. By changing the subject's life routine accordingly, the elevated HR can be lowered. In general, a low HR is desirable since it indicates that the heart is able to pump more blood with each heartbeat with greater efficiency.

SUMMARY

The present disclosure relates to using radar technology to measure the resting heart rate (HR) of a subject over time remotely. Various techniques exist to detect vital signs, and in particular a subject's heartbeat. In addition, various vital sign detection algorithms have been developed to improve the heartbeat detection performance. However, these algorithms' ability to monitor HR over time is not carefully studied in realistic situations.

In this regard, embodiments of the present disclosure include methods, systems, and devices for direct radio frequency (RF) signal processing for HR monitoring using ultra-wide band (UWB) impulse radar. A radar sensor is able to directly sample a received signal at RF which satisfies the Nyquist sampling rate, preserving a subject's vital sign information in the received signal. The vital sign information can be extracted directly from a raw RF signal and thus down conversion to a complex baseband is not required. The HR monitoring performance from the proposed direct RF signal processing technique provides an improvement in continuous HR monitoring as compared against existing methods using a complex baseband signal and/or other measurement techniques.

An exemplary embodiment relates to a method for continuously measuring HR. The method includes transmitting a series of radar impulses toward a subject and receiving an RF response signal corresponding to the series of radar impulses. The method further includes sampling the RF response signal to generate a sampled response signal at an RF band and measuring a HR of the subject from the sampled response signal at the RF band.

Another exemplary embodiment relates to a vital sign detection device. The device includes a radar transmitter configured to transmit a series of radar impulses and an RF receiver configured to receive an RF response signal to the series of radar impulses. The device further includes a processing circuit coupled to the RF receiver and configured to estimate a vital sign of a subject based on the RF response signal without converting the RF response signal to a baseband.

Another exemplary embodiment relates to a method of remotely monitoring a HR. The method includes transmitting a plurality of radar impulse signals to a subject over a distance between 0.5 meters (m) and 2.5 m and receiving an RF response signal to the plurality of radar impulse signals. The method further includes continuously monitoring a HR of the subject by analyzing the RF response signal in a spectral domain.

Those skilled in the art will appreciate the scope of the present disclosure and realize additional aspects thereof after reading the following detailed description of the preferred embodiments in association with the accompanying drawing figures.

BRIEF DESCRIPTION OF THE DRAWING FIGURES

The accompanying drawing figures incorporated in and forming a part of this specification illustrate several aspects of the disclosure, and together with the description serve to explain the principles of the disclosure.

FIG. 1 is a graphical representation of spectral features of an exemplary radio frequency (RF) response signal, illustrating heart rate (HR) and respiration components.

FIG. 2 is a graphical representation of a synthesized ultra-wide band (UWB) pulse and a measured UWB pulse transmitted by a radar transmitter.

FIG. 3 is a graphical representation of heartbeat harmonics and respiration harmonics as a function of chest movement, illustrating an exemplary approach for extracting HR measurements from the RF response signal.

FIG. 4 illustrates relative strengths of 2^nd-order heartbeat harmonics vs. inter-modulations as a function of chest movement.

FIG. 5 is an in-phase/quadrature (I/Q) data constellation plot from a normal breathing subject but with a controlled regular breathing pattern before and after a direct current (DC)-offset correction under a traditional phase-based approach.

FIG. 6 illustrates spectrum results from the phase-based approach before DC-offset correction, the phase-based approach after DC-offset correction, and a direct RF estimation approach according to embodiments of the present disclosure.

FIG. 7A is a flow diagram illustrating an exemplary method for continuously measuring HR.

FIG. 7B is a flow diagram illustrating details of the direct RF signal processing approach for measuring the HR of the subject in the method of FIG. 7A.

FIG. 8 is a schematic diagram of an exemplary vital sign detection device applying the direct RF signal processing approach of the present disclosure.

FIG. 9 is a graphical representation of continuous HR monitoring performance at different ranges.

FIG. 10 is a graphical representation of the direct RF estimation approach compared with a reference electrocardiogram (ECG).

DETAILED DESCRIPTION

The embodiments set forth below represent the necessary information to enable those skilled in the art to practice the embodiments and illustrate the best mode of practicing the embodiments. Upon reading the following description in light of the accompanying drawing figures, those skilled in the art will understand the concepts of the disclosure and will recognize applications of these concepts not particularly addressed herein. It should be understood that these concepts and applications fall within the scope of the disclosure and the accompanying claims.

It will be understood that, although the terms first, second, etc. may be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another. For example, a first element could be termed a second element, and, similarly, a second element could be termed a first element, without departing from the scope of the present disclosure. As used herein, the term “and/or” includes any and all combinations of one or more of the associated listed items.

It will be understood that when an element is referred to as being “connected” or “coupled” to another element, it can be directly connected or coupled to the other element or intervening elements may be present. In contrast, when an element is referred to as being “directly connected” or “directly coupled” to another element, there are no intervening elements present.

The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the disclosure. As used herein, the singular forms “a,” “an,” and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will be further understood that the terms “comprises,” “comprising,” “includes,” and/or “including” when used herein specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.

Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure belongs. It will be further understood that terms used herein should be interpreted as having a meaning that is consistent with their meaning in the context of this specification and the relevant art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.

The present disclosure relates to using radar technology to measure the resting heart rate (HR) of a subject over time remotely. Various techniques exist to detect vital signs, and in particular a subject's heartbeat. In addition, various vital sign detection algorithms have been developed to improve the heartbeat detection performance. However, these algorithms' ability to monitor HR over time is not carefully studied in realistic situations.

In this regard, embodiments of the present disclosure include methods, systems, and devices for direct radio frequency (RF) signal processing for HR monitoring using ultra-wide band (UWB) impulse radar. A radar sensor is able to directly sample a received signal at RF which satisfies the Nyquist sampling rate, preserving a subject's vital sign information in the received signal. The vital sign information can be extracted directly from a raw RF signal and thus down conversion to a complex baseband is not required. The HR monitoring performance from the proposed direct RF signal processing technique provides an improvement in continuous HR monitoring as compared against existing methods using a complex baseband signal and/or other measurement techniques.

I. Signal Model

Vital sign detection using UWB impulse radar can be characterized with a signal model. Under a conventional approach, a received RF response signal is directly sampled in RF and then digitally converted to a complex baseband. A fast-time sampling interval (range) is usually on the order of nanoseconds and an output slow-time sampling interval is on the order of micro-seconds. Under the signal model, τ denotes a fast sampling time and v is a transformed frequency component while t denotes a slow cross-range sample time and f is a corresponding Fourier domain component. The vital sign of a subject at a nominal distance d₀ can be modeled as a sum of two sine waves from respiratory and cardiac activities:

V(t)=d₀+M_b sin(2πf_bt)+M_h sin(2πf_ht)  Eqn. 1

where M_b is an amplitude of respiratory activity, and M_h is an amplitude of cardiac activity. Respiration and heartbeat frequencies are represented as f_b and f_h.

In addition, the RF response signal received by an RF receiver in response to the UWB impulse radar can be modeled as a sum of a target response and a delayed, attenuated version of a transmitted pulse due to static environment:

r(t,τ)=A_Tp(τ−τ_D(t))+Σ_iA_ip(τ−τ_i)  Eqn. 2

where p(t,τ) is a generated short pulse, centered at a carrier frequency F_c. Magnitudes of the target response and multi-path components are denoted as A_T and A_i, while τ_D(t) and τ_i are corresponding delays.

The RF response signal can further be modeled as:

r _o(t,τ)=A_Tp(τ−τ_D(t))  Eqn. 3

where the multi-path components due to static environment are eliminated by mean subtraction. Under a traditional approach, the received RF response signal is then down converted to a complex baseband and is represented as:

$\begin{array}{cc}\begin{array}{c}y\left(t,\tau \right)=r_o\left(t,\tau \right)e^{-j2\pi F_c\tau }\\ =A_{T}p\left(\tau -{\tau }_D\left(t\right)\right)e^{-j2\pi F_c\tau }\end{array}& \mathrm{Eqn}.4\end{array}$

II. Direct RF Signal Processing

In aspects disclosed herein, vital sign information can be directly extracted from the RF response signal in Eqn. 3 at the nominal target distance d₀, as follows:

$\begin{array}{cc}{\tau }_D\left(t\right)=2\frac{d_0+V\left(t\right)}{c}={\tau }_0+2\frac{V\left(t\right)}{c}& \mathrm{Eqn}.5\end{array}$

where c denotes the speed of light. A time-delay variation as a function of t due to chest movement is preserved in Eqn. 3. The Fourier analysis of Eqn. 3 generates a similar result for vital sign detection to the complex baseband signal processing result with the addition of some constant coefficient.

Through forward and backward Fourier transforms with respect to t and τ, the Fourier transform of Eqn. 3 with respect to t is given as:

$\begin{array}{cc}{Y}^{RF}\left(f,\tau \right)=A_T{\sum }_{k=-\infty }^{\infty }{\sum }_{l=-\infty }^{\infty }\delta \left(f-k{f}_b-l{f}_h\right)\times {\int }_{-\infty }^{\infty }dv\left[P\left(v\right)e^{j2\pi v\left(\tau -{\tau }_0\right)}\times J_k\left(4\pi v\frac{M_b}{c}\right)J_l\left(4\pi v\frac{M_h}{c}\right)\right]& \mathrm{Eqn}.6\\ =A_T{\sum }_{k=-\infty }^{\infty }{\sum }_{l=-\infty }^{\infty }{C}_{k,l}\left(\tau \right)\delta \left(f-k{f}_b-l{f}_h\right)& \mathrm{Eqn}.7\end{array}$

where P(v) denotes a Fourier transform of the transmitted pulse p(τ) and J_k(·) denotes a Bessel function of the first kind. Therefore, the Fourier transform result of the received RF response signal in Eqn. 3 at the distance d₀ is given as:

|Y^RF(f,τ₀₎|=|A_T|Σ_k=−∞^∞Σ_l=−∞^∞|C_k,l(τ₀₎|δ(f−kf_b−lf_h)≥|Y^RF(f,τ)|  Eqn. 8

where an absolute value of a complex coefficient C_k,l(τ) achieves its maximum at the delay τ₀ and C_k,l(τ) is given as,

C _k,l(τ)=∫_−∞^∞dv[P(v)e^{j2πv(τ−τ0)}×J_k(4πvM_b/c)J_l(4πvM_h/c)]  Eqn. 9

III. Vital Sign Higher-Order Spectral Features

FIG. 1 is a graphical representation of spectral features of an exemplary RF response signal 10, illustrating HR and respiration components. In an exemplary aspect, a harmonics-based HR estimation method can be used in the RF signal domain because the vital sign information and its associated higher-order spectral features are preserved in the direct RF sampled signal in Eqn. 3 above. The main idea of this harmonics-based HR estimation algorithm is briefly discussed here. In the spectrum domain, an exemplary embodiment looks for higher-order harmonics due to cardiac activity at a relatively higher frequency region, rather than looking for the fundamental heartbeat spectrum peak in the low frequency region.

Conventional methods use the fundamental heartbeat spectrum peak in an attempt to separate the heartbeat spectrum using a pre-defined bandpass filter with frequency limit (such as 0.7 hertz (Hz) to 1.6 Hz). This method fails when the subject's resting HR is low and close to the respiration harmonics region, especially the lower-order respiration harmonics. For example, FIG. 1 illustrates a respiration peak 12 (at approximately 11 beats per minute (BPM)) and a fundamental heartbeat peak 14 (at approximately 55 BPM) in the exemplary RF response signal 10. As further illustrated, a lower-order respiration harmonic 16 (e.g., the fifth respiration harmonic 16) can be stronger than the fundamental heartbeat peak 14. Even worse, in some cases the fundamental heartbeat peak 14 and the 2^nd or the 3^rd respiration harmonics might collocate in the spectral domain. Due to these facts, this approach cannot provide reliable continuous measurement.

In contrast, aspects of the present disclosure estimate the HR based on its higher-order harmonics since they are separated by a fundamental heartbeat frequency (e.g., the frequency at the fundamental heartbeat peak 14). The higher-order harmonics are obtained through a harmonics bandpass filter (e.g., 1.5 Hz to 4 Hz), where the upper limit is trivial and the lower limit is determined by the HR statistics of human subjects (e.g., the normal resting adult HR). As illustrated in FIG. 1, when the fundamental heartbeat peak 14 collocates with the lower-order respiration harmonic 16, the HR cannot be automatically separated without an external reference signal and thus cannot be estimated with confidence.

TABLE I
VITAL SIGN SIMULATION PARAMETERS
	Chest Motion	Resp. Rate	Heartbeat Motion	Heart Rate
	0~5 mm	15 BPM	0.08 mm	70 BPM

IV. Respiration-Free Demonstration

In this section, a quantitative analysis compares the strength of the 2^nd-order harmonics of a heartbeat and its nearby possible competitors in the spectral domain, such as higher-order respiration harmonics and inter-modulations. To perform the following numerical evaluation, some realistic numbers are considered, as summarized in Table I.

A more accurate evaluation is performed to calculate the harmonics strength. Instead of using mean value approximation to evaluate Eqn. 9, the actual transmitted waveform has been taken into consideration. In this regard, FIG. 2 is a graphical representation of a synthesized UWB pulse 18 and a measured UWB pulse 20 transmitted by a radar transmitter. The synthesized UWB pulse 18 is modeled as a cosine wave with a Gaussian envelope:

$\begin{array}{cc}\begin{array}{c}p\left(\tau \right)=p_0\left(\tau \right)\cos \left(2\pi F_c\tau \right)\\ =V_{\mathrm{Tx}}{e}^{-\frac{{\tau }^2}{2{\sigma }^2}}\cos \left(2\pi F_c\tau \right)\end{array}& \mathrm{Eqn}.10\end{array}$

where V_Tx is a pulse amplitude of the Gaussian pulse envelope p₀(τ) and the Gaussian parameter determines the −10 dB bandwidth,

$\begin{array}{cc}\sigma =\frac{1}{2\pi \left(BW/2\right)\sqrt{{\log }_{10}\left(e\right)}}& \mathrm{Eqn}.11\end{array}$

Given the radar system parameters, carrier frequency F_c=7.3 GHz and operating bandwidth BW=1.4 GHz, the waveform of the synthesized UWB pulse 18 corresponds to the measured UWB pulse 20 as seen in FIG. 2.

FIG. 3 is a graphical representation of heartbeat harmonics 22, 24 and respiration harmonics 26, 28, 30, 32, 34 as a function of chest movement, illustrating an exemplary approach for extracting HR measurements from the RF response signal. This approach begins by comparing the strength of the respiration harmonics 26, 28, 30, 32, 34 to the 2^nd-order heartbeat harmonic 24. Then the strength of the higher-order inter-modulations is compared to the heartbeat harmonics of interest.

The harmonics strength is obtained by evaluating the integral in Eqn. 9 using the waveform of the synthesized UWB pulse 18 at −10 dB bandwidth with τ=τ₀. Given a fundamental HR of 70 BPM, the 2^nd-order heartbeat harmonic 24 (140 BPM) is competing against an 8^th-order respiration harmonic 30, a 9^th-order respiration harmonic 32, and a 10^th-order respiration harmonic 34, ranging from 120 to 150 BPM. Only the higher-order respiration harmonics 30, 32, 34 that are close to the 2^nd-order heartbeat harmonic 24 are considered, otherwise they can be easily filtered out.

For normal respiratory activity, the 2^nd-order heartbeat harmonic 24 is much stronger than the 8^th-order respiration harmonic 30, the 9^th-order respiration harmonic 32, and the 10^th-order respiration harmonic 34, as illustrated in FIG. 3. For different chest displacement, the fundamental heartbeat experiences interference from lower-order harmonics of the respiration (e.g., a 2^nd-order respiration harmonic 26). In general, the higher-order harmonics with an order number m=k+l≥4, where k and l are from C_k,l^B(τ₀), cannot be easily observed due to the weak vital sign signal and the background noise. The main challenge of monitoring the fundamental heartbeat over time is due to the spurious spectrum peaks such as the 2^nd-order respiration harmonic 26 or a 3^rd-order respiration harmonic 28. Fortunately, they are spectrally further away from the frequency of the 2^nd-order heartbeat harmonic 24.

FIG. 4 is a graphical representation of heartbeat harmonics 22, 24 and inter-modulations 36 as a function of chest movement. The strength of the higher-order inter-modulations 36 are compared against the strength of the 2^nd-order heartbeat harmonic 24. The associated inter-modulations 36 close to 140 BPM are C_k=−1,l=2^B(τ₀) and C_k=1,l=2^B(τ₀). Only the order numbers of the inter-modulations 36 m≤3 are considered. Under normal breathing, the 2^nd-order heartbeat harmonic 24 is the dominant spectral component in this frequency region. Therefore, locating the 2^nd-order heartbeat harmonic 24 becomes a respiration-interference free task.

V. Issues of Phase-Based Approach

A phase-based estimation approach is widely used for vital sign detection. The main motivation of this traditional approach is that the phase variation is directly related to vital sign information. Theoretically, in the spectral domain, two peaks will be observed: the larger one is due to respiration and the smaller one is heartbeat frequency. To obtain the phase information, a fast-time Fourier transform is performed on the complex baseband signal in Eqn. 4:

Y ^B(t,v)=A_TP(v+F_c)e^{−j(v+Fc)τD(t)}

Y ^B(t,0)=A_TP(v+F_c)e^{−j(Fc)τD(t)}  Eqn. 12

where in order to evaluate Y^B(t, v), v is set to 0, corresponding to summing over the range bins near the distance d₀. Since A_T is a real constant and P(F_c) is a complex constant only contributing a constant phase residual, the desired phase information T_D(t) is preserved in the exponential term Y^B(t, 0)=I(t)+jQ(t).

In a real system, phase noise is inevitable and a phase calibration procedure is required to correctly extract phase information. Since the down conversion is performed in a digital domain, the major source of phase noise is presented in direct current (DC) as D_I and D_Q in an in-phase/quadrature (I/Q) channel. Let İ(t) and {dot over (Q)}(t) denote the distorted I/Q data,

$\begin{array}{cc}\begin{array}{c}\stackrel{.}{\varphi }\left(t\right)=\mathrm{unwrapping}\left\{A\tan \left(\frac{\stackrel{.}{Q}\left(t\right)}{\stackrel{.}{I}\left(t\right)}\right)\right\}\\ =\mathrm{unwrapping}\left\{A\tan \left(\frac{A_0\sin \left(4\mathrm{\pi V}\left(t\right)/\lambda +{\phi }_0\right)+D_Q}{A_0\cos \left(4\mathrm{\pi V}\left(t\right)/\lambda +{\phi }_0\right)+D_I}\right)\right\}\end{array}\to \frac{4\mathrm{\pi V}\left(t\right)}{\lambda }+{\phi }_0& \mathrm{Eqn}.13\end{array}$

where A₀ and ϕ₀ are the nominal amplitude and the constant phase residual. If the DC-offset can be perfectly corrected, the final expression is the desired phase information. A phase error calibration procedure is often implemented, which has been shown to be one of the most accurate center tracking algorithms.

FIG. 5 is an I/Q data constellation plot 38 from a normal breathing subject but with a controlled regular breathing pattern before and after a DC-offset correction under the traditional phase-based approach. The effectiveness of the phase-based approach with DC-offset calibration has been demonstrated, but here the challenge of correctly applying this method in a realistic situation is explained. For example, it must be determined which of these data points need to be corrected at every processing window. Ideally, the data points corresponding to the same micro-motion (meaning that they belong to the same phase trajectory in the I/Q data constellation plot 38 of FIG. 5) should be corrected at each processing window.

In reality, respiratory activity, the stronger motion in the vital sign, can be quite different from breath to breath as seen in FIG. 6 and thus the calibration should be performed on a similar respiration pattern. FIG. 6 illustrates spectrum results from a phase-based approach before DC-offset correction 40, a phase-based approach after DC-offset correction 42, and a direct RF estimation approach 44 according to embodiments of the present disclosure. An automatic approach should be developed to perform this task.

Most practitioners apply the phase calibration method on fixed length data points, like 15-second or 20-second of data points, implicitly assuming that the respiratory activity is a stationary process. On the contrary, it is a time-varying and non-stationary process even for a resting subject. Without taking these facts into consideration, the phase calibration can introduce spectrum distortion as shown in the phase-based approaches 40, 42 of FIG. 6 (generated from a 20-second dataset containing irregular respiration pattern acquired from normal breathing subject). However, in this example the proposed direct RF estimation approach 44 can accurately recover the fundamental HR from the 2^nd-order heartbeat harmonic 24: 117.8/2≈59 BPM. Additionally, the phase-based approaches 40, 42 are iterative and require much more computation than the proposed direct RF estimation approach 44.

VI. Proposed Method and System

FIG. 7A is a flow diagram illustrating an exemplary method 700 for continuously measuring HR. The method includes operation 702 with transmitting a series of radar impulses toward a subject. In an exemplary aspect, a radar transmitter transmits the synthesized UWB pulse 18. The method 700 further includes operation 704 with receiving the RF response signal 10 corresponding to the series of radar impulses. The method 700 further includes operation 706 with sampling the RF response signal 10 to generate a sampled response signal at an RF band. The method 700 further includes operation 708 with measuring a HR of the subject from the sampled response signal at the RF band.

FIG. 7B is a flow diagram illustrating details of the direct RF signal processing approach for measuring the HR (operation 708) of the subject in the method 700 of FIG. 7A. In an exemplary aspect, the RF response signal 10 includes a series of captured radar frames. In operation 710, after sampling the RF response signal 10, the series of captured radar frames is organized.

Next, in operation 712, clutter suppression is performed on the acquired mixed data to remove static background. The clutter suppression may be realized by high-pass filtering. In some applications, the clutter suppression may be omitted. In operation 714, a harmonics bandpass filter is applied to the sampled RF response signal to determine a time delay of interest for the subject. The time delay of interest refers to a time delay range at one or more impulse responses (e.g., each impulse response) in which a portion of the sampled RF response signal contains the highest energy. In some examples, the time delay of interest also indicates a distance between the subject and the radar transmitter.

While remaining in the RF domain, in operation 716, motion artifacts are extracted from the sampled RF response signal. The motion artifacts are extracted by measuring changes in amplitude of the sampled RF response signal at each of the radar frames at the time delay of interest. The extracted motion artifacts can include the respiration component and the heartbeat component, as modeled above. In operation 718, the extracted motion artifacts are processed to produce vital sign data. As further described above, processing the extracted motion artifacts includes locating the 2^nd-order heartbeat harmonic 24 (and/or other higher-order harmonics) of the subject.

Although the operations of FIGS. 7A and 7B are illustrated in a series, this is for illustrative purposes and the operations are not necessarily order dependent. Some operations may be performed in a different order than that presented. Further, processes within the scope of this disclosure may include fewer or more steps than those illustrated in FIGS. 7A and 7B.

FIG. 8 is a schematic diagram of an exemplary vital sign detection device 46 applying the direct RF signal processing approach of the present disclosure. In this regard, the vital sign detection device 46 may be a circuit or circuits included in an electronic board card, such as, a printed circuit board (PCB), a server, a personal computer, a desktop computer, a laptop computer, an array of computers, a personal digital assistant (PDA), a computing pad, a mobile device, or any other device, and may represent, for example, a server or a user's computer.

The vital sign detection device 46 in this embodiment includes a processing device 48 or processor, a main memory 50 (e.g., read-only memory (ROM), flash memory, dynamic random access memory (DRAM), such as synchronous DRAM (SDRAM), etc.), and a static memory 52 (e.g., flash memory, SRAM, etc.), which may communicate with each other via a data bus 54. Alternatively, the processing device 48 may be connected to the main memory 50 and/or the static memory 52 directly or via some other connectivity means. In an exemplary aspect, the processing device 48 may be used to perform any of the methods or functions described above, such as measuring and/or monitoring the HR or other vital signs of a subject.

The processing device 48 represents one or more general-purpose processing devices, such as a microprocessor, central processing unit (CPU), or the like. More particularly, the processing device 48 may be a complex instruction set computing (CISC) microprocessor, a reduced instruction set computing (RISC) microprocessor, a very long instruction word (VLIW) microprocessor, a processor implementing other instruction sets, or other processors implementing a combination of instruction sets. The processing device 48 is configured to execute processing logic in instructions for performing the operations and steps discussed herein.

The various illustrative logical blocks, modules, and circuits described in connection with the embodiments disclosed herein may be implemented or performed with the processing device 48, which may be a field programmable gate array (FPGA), a digital signal processor (DSP), an application-specific integrated circuit (ASIC), or other programmable logic device, a discrete gate or transistor logic, discrete hardware components, or any combination thereof designed to perform the functions described herein. Furthermore, the processing device 48 may be a microprocessor, or may be any conventional processor, controller, microcontroller, or state machine. The processing device 48 may also be implemented as a combination of computing devices (e.g., a combination of a DSP and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a DSP core, or any other such configuration).

The vital sign detection device 46 further includes a radar transmitter 56 configured to transmit a series of radar impulses. Each of the series of radar impulses may be similar to the synthesized UWB pulse 18 described above. The vital sign detection device 48 further includes an RF receiver 58 configured to receive the RF response signal 10 to the series of radar impulses, which are processed by the processing device 48 as described above. In some examples, the radar transmitter 56 and the RF receiver 58 can be separate components of the vital sign detection device 46. In other examples, the radar transmitter 56 and the RF receiver 58 can be implemented in one component.

The vital sign detection device 46 may further include a network interface device 60. The vital sign detection device 46 also may or may not include an input 62, configured to receive input and selections to be communicated to the vital sign detection device 46 when executing instructions. The vital sign detection device 46 also may or may not include an output 64, including but not limited to a display, a video display unit (e.g., a liquid crystal display (LCD) or a cathode ray tube (CRT)), an alphanumeric input device (e.g., a keyboard), and/or a cursor control device (e.g., a mouse).

The vital sign detection device 46 may or may not include a data storage device that includes instructions 66 stored in a computer-readable medium 68. The instructions 66 may also reside, completely or at least partially, within the main memory 50 and/or within the processing device 48 during execution thereof by the vital sign detection device 46, the main memory 50, and the processing device 48 also constituting computer-readable medium. The instructions 66 may further be transmitted or received via the network interface device 60.

While the computer-readable medium 68 is shown in an exemplary embodiment to be a single medium, the term “computer-readable medium” should be taken to include a single medium or multiple media (e.g., a centralized or distributed database, and/or associated caches and servers) that store the one or more sets of instructions 66. The term “computer-readable medium” shall also be taken to include any medium that is capable of storing, encoding, or carrying a set of instructions for execution by the processing device and that causes the processing device to perform any one or more of the methodologies of the embodiments disclosed herein. The term “computer-readable medium” shall accordingly be taken to include, but not be limited to, solid-state memories, optical medium, and magnetic medium.

VII. Performance Demonstration

In this section, experimental results are provided. FIG. 9 is a graphical representation of continuous HR monitoring performance at different ranges. In this example, seven measurements are captured ranging from 0.3 m to 3 m. A direct RF estimation approach 70 using the harmonics-based HR estimation algorithm described above can correctly detect the HR up to 2.5 m with a maximum error of 2 BPM at 2 m. A conventional complex signal demodulation (CSD) approach 72 in the complex baseband signal domain does not provide consistent estimates and results in large errors due to strong respiration harmonics.

FIG. 10 is a graphical representation of the direct RF estimation approach compared with a reference electrocardiogram (ECG). This example demonstrates the accuracy of a continuous measurement performance 74 of the direct RF estimation approach by comparing against a standard ECG reference signal 76. During the entire recording, the radar estimates are consistent with the ECG reference signal 76 and the estimation error is well within 1 BPM.

In this disclosure, it has been shown that the directly sampled RF signal preserves the vital sign information. The disclosure further presents a harmonics-based HR estimation algorithm in the RF domain. As for heartbeat identifiability in the spectral domain, the fundamental heartbeat is interference limited and the heartbeat harmonics are signal-to-noise ratio limited. The first issue is a much more challenging task as discussed herein. The second issue can be solved by coherently combining a relatively large number of pulses to achieve the desired processing gain since the pulse repetition rate can be much higher than 1 MHz.

Those skilled in the art will recognize improvements and modifications to the preferred embodiments of the present disclosure. All such improvements and modifications are considered within the scope of the concepts disclosed herein and the claims that follow.

Claims

1. A method for continuously measuring heart rate, comprising: transmitting a series of radar impulses toward a subject;receiving a radio frequency (RF) response signal corresponding to the series of radar impulses;sampling the RF response signal to generate a sampled response signal at an RF band; andmeasuring a heart rate of the subject from the sampled response signal at the RF band.

2. The method of claim 1, wherein the heart rate is measured from the sampled response signal without a baseband conversion.

3. The method of claim 1, wherein the RF response signal comprises a plurality of captured radar frames.

4. The method of claim 3, further comprising applying a harmonics bandpass filter to the sampled response signal to identify a time delay of interest in the plurality of captured radar frames.

5. The method of claim 4, wherein measuring the heart rate comprises measuring changes in amplitude of the sampled response signal at each of the plurality of captured radar frames at the time delay of interest.

6. The method of claim 4, wherein the harmonics bandpass filter is between 1.5 hertz (Hz) and 4 Hz.

7. The method of claim 1, wherein the series of radar impulses comprises ultra-wide band (UWB) radar impulses.

8. The method of claim 1, wherein measuring the heart rate comprises locating second order harmonics of a heartbeat of the subject.

9. The method of claim 8, wherein the locating the second order harmonics of the heartbeat removes respiration interference from the sampled response signal.

10. The method of claim 1, wherein the series of radar impulses is transmitted at a distance of between 1 meter (m) and 2.5 m from the subject.

11. A vital sign detection device, comprising: a radar transmitter configured to transmit a series of radar impulses;a radio-frequency (RF) receiver configured to receive an RF response signal from the series of radar impulses; anda processing circuit coupled to the RF receiver and configured to estimate a vital sign of a subject based on the RF response signal without converting the RF response signal to a baseband.

12. The vital sign detection device of claim 11, wherein: the radar transmitter comprises an ultra-wide band (UWB) radar transmitter; andeach impulse of the series of radar impulses comprises a bandwidth between 1.0 gigahertz (GHz) and 2.0 GHz.

13. The vital sign detection device of claim 11, wherein the processing circuit is configured to estimate the vital sign of the subject by: sampling the RF response signal to generate a sampled response signal at an RF band; andmeasuring a heart rate of the subject from the sampled response signal at the RF band.

14. The vital sign detection device of claim 11, further comprising a harmonics bandpass filter, wherein the processing circuit is configured to identify a time delay of interest in the RF response signal based on an output of the harmonics bandpass filter.

15. The vital sign detection device of claim 14, wherein the processing circuit is configured to estimate the vital sign by measuring changes in amplitude of the RF response signal at the time delay of interest.

16. The vital sign detection device of claim 11, wherein the processing circuit is configured to estimate the vital sign comprising a heart rate of the subject after removing respiration motion artifacts from the RF response signal.

17. A method of remotely monitoring a heart rate, comprising: transmitting a plurality of radar impulse signals to a subject over a distance between 0.5 meters (m) and 2.5 m;receiving a radio frequency (RF) response signal from the plurality of radar impulse signals; andcontinuously monitoring a heart rate of the subject by analyzing the RF response signal in a spectral domain.

18. The method of claim 17, wherein the monitored heart rate is accurate within 2 beats per minute (BPM) as compared with an electrocardiogram (ECG) measured through contact with the subject.

19. The method of claim 17, wherein the heart rate is measured from the RF response signal without a baseband conversion.

20. The method of claim 17, further comprising applying a harmonics bandpass filter to the RF response signal and measuring changes in amplitude of the RF response signal over the plurality of radar impulse signals.