Millimeter wave radar heart rate signal enhancement method and system based on harmonic reconstruction

Abstract

The invention provides a millimeter wave radar heart rate signal enhancement method and system based on harmonic reconstruction, and the method comprises the steps: S1, obtaining an echo signal, acting on a target, of an FMCW signal transmitted by a millimeter wave radar, and carrying out the preprocessing, and obtaining a preprocessed echo signal; s2, phase information of the preprocessed echo signals is obtained, filtering processing is carried out, and initial heart rate signals are extracted; s3, recovering low-order and high-order heart rate harmonics of the initial heart rate signal to obtain a delay self-mixing signal; s4, the heart rate value is obtained by detecting frequency spectrum information of the delay self-mixing signals.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of radar signal processing, and in particular to a millimeter wave radar heart rate signal enhancement method and system based on harmonic reconstruction. BACKGROUND

[0002] Millimeter wave radar technology has unique advantages in heart rate monitoring: all-weather, non-contact, no privacy invasion, and stronger universality and comfort compared to other monitoring methods such as rPPG (rPPG) method, ECG (ECG) method, etc.

[0003] However, since the human respiratory rate is generally 0.1-0.5Hz, the heartbeat is generally 0.8-2.0Hz, and the respiratory signal generally has high-order harmonic components; and the respiratory displacement is generally 4-12mm, and the heartbeat is generally 0.2-0.5mm. As can be seen, when using millimeter wave radar technology to monitor heart rate, the signal is easily disturbed by the respiratory signal, affecting the accuracy of the monitoring result. SUMMARY

[0004] Therefore, in order to at least partially solve at least one of the above-mentioned technical problems, the present application provides a millimeter wave radar heart rate signal enhancement method and system based on harmonic reconstruction, and the technical solution is as follows:

[0005] According to an embodiment of the present application, a millimeter wave radar heart rate signal enhancement method based on harmonic reconstruction is provided, comprising: operation S1: obtaining the echo signal of the FMCW signal emitted by the millimeter wave radar acting on the target and pre-processing to obtain a pre-processed echo signal; operation S2: obtaining the phase information of the pre-processed echo signal and performing filtering processing to extract an initial heart rate signal; operation S3: recovering the low-order and high-order heart rate harmonics of the initial heart rate signal to obtain a delay self-mixing signal; and operation S4: obtaining a heart rate value by detecting the frequency spectrum information of the delay self-mixing signal.

[0006] According to the embodiment of the present application, in operation S1: the echo signal is converted into an intermediate frequency signal and then sampled by ADC to obtain an ADC signal; the ADC signal is subjected to distance dimension Fourier transform to obtain a distance domain signal; the target delay information related to the sampling point number corresponds to the distance unit number corresponding to the distance dimension Fourier transform, and the distance unit number is denoted as R, the reflection energy of different distance targets is separated by FFT, and each distance unit corresponds to a specific distance interval; N consecutive frames of distance dimension Fourier transform data are recorded in time sequence to form a pre-processed echo signal S, the pre-processed echo signal is a slow time-distance complex matrix with a dimension of N x R, and N and R are positive integers.

[0007] According to an embodiment of the present invention, operation S2 includes: selecting complex data of the slow time dimension of the target location distance cell r from the preprocessed echo signal S, denoted as S. r The target phase φ is calculated based on the arctangent principle. r :

[0008] φ r =arctan(imag(S r ),real(S r ));

[0009] Among them, imag(S r ) represents taking the imaginary part of the complex number, real(S) represents taking the real part of the complex number, and arctan(·) represents performing the arctangent operation on the complex number.

[0010] Phase expansion operations are used to eliminate periodic phase jumps in the target phase, resulting in target phase data carrying information about the target's respiration and heartbeat. Based on the frequency difference between respiration and heartbeat, a bandpass filter is used to filter out heart rate harmonic signals. These heart rate harmonic signals include some higher-order heart rate harmonics.

[0011] According to an embodiment of the present invention, in operation S3, the delayed self-mixing signal Represented as:

[0012] ;

[0013] ;

[0014] in The self-mixing term influence factor related to heart rate signal characteristics, where Indicates the first The second harmonic, m≥1, k≥0; n represents noise, its influence is ignored, then the delayed self-mixing signal Represented as:

[0015] .

[0016] The initial heart rate signal has a heart rate harmonic order range of m to m+k. After delay self-mixing, the harmonic order range of the delayed self-mixed signal is expanded to 0 to 2(m+k), where m≥1 and k≥0, thereby increasing the low-order harmonic components and higher-order harmonic components.

[0017] According to an embodiment of the present invention, in operation S4, the delayed self-mixing signal is subjected to Fourier transform to obtain spectral information, the first spectral peak is detected as the fundamental frequency peak and the frequency information is recorded, and the heart rate value is calculated based on the frequency information.

[0018] According to an embodiment of the present application, a harmonic reconstruction based millimeter wave radar heart rate signal enhancement system is provided, comprising: a preprocessing unit, which obtains echo signals of a target acted on by FMCW signals transmitted by a millimeter wave radar and performs preprocessing to obtain preprocessed echo signals; a filtering unit, which obtains phase information of the preprocessed echo signals and performs filtering processing to extract an initial heart rate signal; a self-mixing unit, which recovers low-order and high-order heart rate harmonics of the initial heart rate signal to obtain a delayed self-mixing signal; and a spectrum detection unit, which obtains a heart rate value by detecting spectrum information of the delayed self-mixing signal. BRIEF DESCRIPTION OF DRAWINGS

[0019] The above and other objects, features and advantages of the present application will become more apparent from the following description of embodiments of the present application taken with reference to the accompanying drawings, in which:

[0020] Figure 1 A flowchart of a harmonic reconstruction based millimeter wave radar heart rate signal enhancement method according to an embodiment of the present application.

[0021] Figure 2 A composition schematic diagram of a harmonic reconstruction based millimeter wave radar heart rate signal enhancement system according to an embodiment of the present application.

[0022] Figure 3 A radar echo signal schematic diagram containing a stationary human target according to an embodiment of the present application.

[0023] Figure 4 A target phase data schematic diagram of an echo signal carrying breathing and heartbeat information according to an embodiment of the present application.

[0024] Figure 5 An initial heart rate signal spectrum schematic diagram obtained by a band-pass filter according to an embodiment of the present application.

[0025] Figure 6 A heart rate signal spectrum schematic diagram obtained after delay self-mixing according to an embodiment of the present application. DETAILED DESCRIPTION

[0026] The present application provides a harmonic reconstruction based millimeter wave radar heart rate signal enhancement method and system, which utilizes high-order harmonic components rich in heart rate for self-mixing. From spectrum analysis, the spectrum of the self-mixing signal comes from sum frequencies or difference frequencies of each harmonic frequency. Due to the high consistency of the interval between adjacent harmonic frequencies, the harmonic of the mixed signal has new sum frequency and difference frequency harmonics, thus enriching the harmonic components of the original signal. The method realizes heart rate rhythm feature enhancement by generating heart rate harmonic components, making heart rate calculation more accurate and robust.

[0027] The basic process of the non-contact heart rate monitoring based on the millimeter wave radar is that the radar transmits millimeter waves to a target area, the millimeter waves are modulated by the surface micro-displacement vibration caused by the respiration and heartbeat of the subject, and then returned to the radar receiver, and the respiration and heartbeat rhythm and other sign information are extracted after analysis and processing. Since the human respiration frequency is generally 0.1-0.5 Hz, the heartbeat is generally 0.8-2.0 Hz, and the respiration signal generally has a high-order harmonic component; and the respiration displacement is generally 4-12 mm, and the heartbeat is generally 0.2-0.5 mm. As can be seen, when extracting the heartbeat signal, it is easy to be disturbed by the respiration signal. A common method is to select to avoid the frequency spectrum interval where the respiration signal is located, and to filter the high-order harmonic interval by using the rich harmonic characteristics of the heartbeat impact, such as the invention patent with the authorized publication number CN115813363B "a heart rate estimation method based on heartbeat second harmonic reconstruction" and the invention patent with the publication number CN119014840A "a millimeter wave radar human heart rate measurement method and system based on heartbeat harmonic analysis". The technical solutions involved in the above two patents can weaken the influence of respiration and improve the robustness and accuracy of heart rate calculation by avoiding the heart rate fundamental frequency interval seriously disturbed by respiration and selecting high-order harmonics of heart rate for analysis. However, in actual scenarios, this method also faces the risk of failure, mainly because the intensity of each harmonic of the heartbeat is not uniform, and even some harmonics are missing, so that the heart rate is determined by the harmonics, which may cause errors.

[0028] In view of the deficiencies and problems of the prior art, the present application provides a millimeter wave radar heart rate signal enhancement method and system based on harmonic reconstruction. The method avoids the respiration frequency interval, reconstructs other order harmonics by using high-order harmonics, and strengthens the heart rate signal, thereby improving the accuracy and robustness of heart rate extraction.

[0029] To make the purpose, technical scheme and advantages of the present application clearer, the present application is further described in detail below with reference to the drawings and in conjunction with specific embodiments.

[0030] In the embodiments of the present application, a millimeter wave radar heart rate signal enhancement method based on harmonic reconstruction is provided, as shown in Figure 1 The millimeter wave radar heart rate signal enhancement method comprises:

[0031] Operation S1: obtaining the echo signal of the FMCW signal transmitted by the millimeter wave radar acting on the target and pre-processing to obtain a pre-processed echo signal;

[0032] Operation S2: obtaining the phase information of the pre-processed echo signal and performing filtering processing to extract an initial heart rate signal;

[0033] Operation S3: restoring the low-order and high-order heart rate harmonics of the initial heart rate signal to obtain a delay self-mixing signal; and

[0034] Operation S4: obtaining a heart rate value by detecting spectrum information of the delay self-mixing signal.

[0035] According to the embodiment of the present application, operation S1 includes:

[0036] After converting the echo signal into an intermediate frequency signal, the ADC sampling is performed to obtain an ADC signal;

[0037] The ADC signal is subjected to a range dimension Fourier transform to obtain a range domain signal;

[0038] The target delay information related to the sampling point number corresponds to the distance unit number corresponding to the range dimension Fourier transform, and the distance unit number is denoted as R. The reflection energy of different distance targets is separated by FFT, and each distance unit corresponds to a specific distance interval. N consecutive frames of range dimension Fourier transform data are recorded in time sequence to form a preprocessed echo signal S. The preprocessed echo signal is a slow time-range complex matrix with a dimension of N×R, and N and R are positive integers.

[0039] Specifically, based on a frequency-modulated continuous wave (FMCW) radar, an FMCW signal is transmitted to a target space, and an echo signal carrying target vibration information is converted into an intermediate frequency signal (IF) by a receiving channel, and then ADC acquisition is performed. The ADC signal is subjected to a range dimension Fourier transform (Range-FFT) to convert a time domain signal into a range domain signal. The target delay information related to the sampling point number corresponds to the distance unit number corresponding to the Range-FFT, and the distance unit number is denoted as R. The reflection energy of different distance targets is separated by FFT, and each distance unit corresponds to a specific distance interval. N consecutive frames of Range-FFT data are recorded in time sequence to form a slow time-range complex matrix S (dimension: N×R). Figure 3 A radar echo signal containing a stationary human target is shown. The target is in the 17th distance unit, the horizontal coordinate is the slow time (seconds), and the vertical coordinate is the distance unit number.

[0040] According to the embodiment of the present application, operation S2 includes: selecting slow time dimension complex data of a distance unit r where the target is located from the preprocessed echo signal S, denoted as S r , and calculating the target phase φ r according to the arctangent principle:

[0041] φ r =arctan(imag(S r ),real(S r ));

[0042] wherein, imag(S r) represents the imaginary part of the complex data, real(S) represents the real part of the complex data, and arctan(·) represents the inverse tangent operation on the complex data.

[0043] Further, the periodic jump of the target phase is eliminated by a phase unwrapping operation, and target phase data carrying target breathing and heartbeat information is obtained.

[0044] Specifically, when the periodic jump is eliminated by phase unwrapping, the following operation is performed:

[0045] Ф r = unwrap(φ r );

[0046] where unwrap(·) represents the unwrapping operation on the phase. Since the heartbeat usually contains the breathing signal, the above operation obtains the target phase data Ф r , which contains the breathing and heartbeat information.

[0047] Further, according to the frequency difference between the breathing and the heartbeat, a heart rate harmonic signal is filtered out by a band-pass filter, and an initial heart rate signal is obtained; the heart rate harmonic signal includes part of the high-order heart rate harmonics. Figure 4 The target phase data carrying the breathing and heartbeat information is shown, and it can be seen that the weak heartbeat signal is submerged in the breathing signal. In order to calculate the heart rate subsequently, the breathing and heartbeat signal separation needs to be performed, according to the characteristics that the breathing and heartbeat frequencies are different (breathing: 0.1-0.5Hz, heartbeat: 0.8-2.0Hz) and the heartbeat signal has rich harmonics, the present application filters out the high-order heart rate harmonics Ф r_hh = IIR_BF(Ф r ), where IIR_BF(·) represents the band-pass filtering of the input signal by the designed IIR filter. Figure 5 The spectrum of the initial heart rate signal after band-pass filtering is shown (the harmonic order is marked in red), and it can be seen that the heart rate harmonic order is concentrated in 3-20.

[0048] According to the embodiment of the present application, in operation S3, the low-order and high-order heart rate harmonics of the initial heart rate signal obtained by the above operation are recovered, and a delay self-mixing signal is obtained.

[0049] The heart rate signal has the characteristics that the multiple harmonics and the harmonic spacing are strictly consistent, and the Ф r_hh containing multiple heart rate harmonic signals is delay self-mixed to generate other low-order and high-order heart rate harmonics, and a delay self-mixing signal is obtained. The delay self-mixing signal is represented as:

[0050] ;

[0051] ;

[0052] Wherein, g is the self-mixing term influence factor related to the heart rate signal characteristics, according to the heart rate signal characteristics, here g takes 2000, and the value can be adjusted according to the actual application. Wherein Indicates the n-th harmonic; n represents the noise, then:

[0053] ;

[0054] Since n is mainly noise, it is relatively small, and its influence is ignored, then the delay self-mixing signal obtained after mixing is:

[0055] ;

[0056] Wherein, The heart rate harmonic of the initial heart rate signal, The output of the self-mixing part is the sum and difference frequency of the input harmonic And , that is , wherein The harmonic order of the initial heart rate signal ranges from m to m+k, and the harmonic order of the delay self-mixing signal obtained after delay self-mixing ranges from 0 to 2(m+k), m, k are positive integers, m≥1, k≥0, thereby increasing the low-order harmonic component and the higher-order harmonic component. As can be seen, the input harmonic range is expanded after processing. Figure 6 The delay self-mixing heart rate signal spectrum obtained after delay self-mixing is shown, the harmonic order is marked in red, and it can be seen that the heart rate harmonic order is expanded from 3-20 to 1-35.

[0057] According to the embodiment of the application, in operation S4, the delay self-mixing signal is subjected to Fourier transform to obtain frequency spectrum information, the first frequency spectrum peak is detected as a frequency fundamental frequency peak and the frequency information is recorded, and the heart rate value is calculated according to the frequency information.

[0058] Specifically, the obtained delay self-mixing signal is subjected to Fourier transform to obtain frequency spectrum information, the first peak of the frequency spectrum is detected, which is a heart rate fundamental frequency peak, and the frequency is recorded, then the heart rate value is , here the heart rate value is in the form of BPM, that is, the number of heartbeats per minute.

[0059] Another embodiment of the application also provides a millimeter wave radar heart rate signal enhancement system based on harmonic reconstruction, as shown in Figure 2 The millimeter wave radar heart rate signal enhancement system 100 comprises:​

[0060] a preprocessing unit, which obtains an echo signal of a FMCW signal transmitted by a millimeter wave radar and acting on a target and performs preprocessing to obtain a preprocessed echo signal;

[0061] a filtering unit, which obtains phase information of the preprocessed echo signal and performs filtering processing to extract an initial heart rate signal;

[0062] a self-mixing unit, which recovers low-order and high-order heart rate harmonics of the initial heart rate signal to obtain a delayed self-mixing signal; and

[0063] a spectrum detection unit, which obtains a heart rate value by detecting spectrum information of the delayed self-mixing signal.

[0064] The millimeter wave radar heart rate signal enhancement method and system based on harmonic reconstruction of the present application introduces a delayed self-mixing algorithm in heart rate calculation. By avoiding the respiratory frequency interval, using the high-order harmonic delayed self-mixing of the heart rate, not only the original high-order harmonic components are retained, but also other low-order and higher-order harmonic components are added, thereby strengthening the heart rate signal. The results are: first, the generated fundamental frequency heart rate makes the heart rate calculation more convenient; second, the rich harmonic components make the impact characteristics of the heart rate more obvious. Finally, the purpose of improving the accuracy and robustness of heart rate extraction is achieved.

[0065] Thus far, the embodiments of the present application have been described in detail with reference to the accompanying drawings. It should be noted that the implementation modes not shown or described in the drawings or the main text are known to those skilled in the art, and are not described in detail. In addition, the definitions of the elements and methods described above are not limited to the various specific structures, shapes or modes mentioned in the embodiments, and can be simply changed or replaced by those skilled in the art.

[0066] Unless specifically described or steps must occur in sequence, the order of the above steps is not limited to the above list, and can be changed or rearranged according to the desired design. And the above embodiments can be mixed and used with each other or with other embodiments based on design and reliability considerations, that is, the technical features in different embodiments can be freely combined to form more embodiments.

[0067] The above specific embodiments further illustrate the purpose, technical solutions and advantages of the present application. It should be understood that the above description is only a specific embodiment of the present application and is not intended to limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. A method for enhancing a heart rate signal of a millimeter wave radar based on harmonic reconstruction, comprising: Operation S1: obtaining echo signals of a target acted on by FMCW signals transmitted by a millimeter wave radar and pre-processing to obtain pre-processed echo signals; Operation S2: obtaining phase information of the pre-processed echo signals and performing filtering processing to extract an initial heart rate signal; Operation S3: recovering low-order and high-order heart rate harmonics of the initial heart rate signal to obtain a delay self-mixing signal; and Operation S4: obtaining a heart rate value by detecting spectral information of the delay self-mixing signal. 2.The method for enhancing a heart rate signal of a millimeter wave radar based on harmonic reconstruction according to claim 1, wherein in operation S1, the following steps are included: converting the echo signals into intermediate frequency signals and then performing ADC sampling to obtain ADC signals; performing distance dimension Fourier transform on the ADC signals to obtain distance domain signals; target delay information related to the number of sampling points corresponds to the number of distance units corresponding to the distance dimension Fourier transform, denoted as R, the reflection energy of different distance targets is separated by FFT, and each distance unit corresponds to a specific distance interval; N consecutive frames of distance dimension Fourier transform data are recorded in time sequence to form a pre-processed echo signal S, the pre-processed echo signal is a slow time-distance complex matrix with a dimension of N×R, N and R are positive integers. 3.The method for enhancing a heart rate signal of a millimeter wave radar based on harmonic reconstruction according to claim 2, wherein in operation S2, the following steps are included: Selecting the complex data in the slow time dimension of the target distance cell r from the pre-processed echo signal S, denoted as S r , calculating the target phase φ according to the arctangent principle r : φ r = arctan(imag(S r ),real(S r )) ; Among them, imag(S r ) represents taking the imaginary part of the complex number, real(S) represents taking the real part of the complex number, and arctan(·) represents performing the arctangent operation on the complex number.

4. The harmonic reconstruction based millimeter wave radar heart rate signal enhancement method of claim 3, further comprising, in operation S2: eliminating the periodic jump of the target phase by phase unwrapping operation to obtain target phase data carrying target breathing and heartbeat information.

5. The harmonic reconstruction based millimeter wave radar heart rate signal enhancement method of claim 4, further comprising: According to the frequency difference between breathing and heartbeat, a heart rate harmonic signal is filtered out by a band-pass filter. 6.The method for enhancing a heart rate signal of a millimeter wave radar based on harmonic reconstruction according to claim 5, wherein the heart rate harmonic signal includes part of high-order heart rate harmonics.

7. The harmonic reconstruction based millimeter wave radar heart rate signal enhancement method of claim 6, delay self-mixing signal is represented as: ； ； wherein is a factor of the influence of the self-mixing term related to the heart rate signal characteristics, wherein denotes the mth harmonic, m ≥ 1, k ≥ 0; n denotes the noise, which is neglected, then the delayed self-mixing signal is represented as: 。 8.The method for enhancing a heart rate signal of a millimeter wave radar based on harmonic reconstruction according to claim 7, wherein the initial heart rate signal has a heart rate harmonic order range of m to m+k, the harmonic order range of the delay self-mixing signal obtained by delay self-mixing is expanded to 0 to 2(m+k), m≥1, k≥0, thereby increasing low-order harmonic components and higher-order harmonic components. 9.The method for enhancing a heart rate signal of a millimeter wave radar based on harmonic reconstruction according to claim 1, wherein in operation S4, Fourier transform is performed on the delay self-mixing signal to obtain spectral information, a first spectral peak is detected as a frequency fundamental peak and frequency information is recorded, and a heart rate value is calculated according to the frequency information. 10.A system for enhancing a heart rate signal of a millimeter wave radar based on harmonic reconstruction, comprising: a pre-processing unit configured to obtain echo signals of a target acted on by FMCW signals transmitted by a millimeter wave radar and pre-process to obtain pre-processed echo signals; a filtering unit configured to obtain phase information of the pre-processed echo signals and perform filtering processing to extract an initial heart rate signal; a self-mixing unit configured to recover low-order and high-order heart rate harmonics of the initial heart rate signal to obtain a delay self-mixing signal; and a spectral detection unit configured to obtain a heart rate value by detecting spectral information of the delay self-mixing signal. ​

Images