Human respiration and heartbeat monitoring method based on millimeter wave radar time domain undersampling

Abstract

The invention discloses a human respiration and heartbeat monitoring method based on millimeter-wave radar time domain undersampling, which comprises the following steps: modeling thoracic cavity fluctuation caused by respiration and thoracic cavity micro-motion caused by heartbeat, and deducing a micro-Doppler signal mathematical analytical expression corresponding to human respiration and heartbeat by combining a millimeter-wave radar signal system; on the basis of defining a radar echo modulation mechanism, a time domain undersampling method is provided, and a Doppler signal can be accurately extracted from a time domain difference frequency signal by taking a radar modulation frequency as a sampling frequency; different filters are designed according to frequency bands of the respiration signals and the heartbeat signals, and Doppler signals corresponding to respiration and micro-Doppler signals corresponding to heartbeat are separated from the proposed signals; and calculating the respiration Doppler signal frequency and the heartbeat micro Doppler signal frequency to obtain the human respiration frequency and the heartbeat frequency so as to realize human respiration and heartbeat monitoring. According to the invention, the breathing and heartbeat frequencies of the human body can be accurately inverted under the non-contact condition.

Description

Technical Field

[0001] The present invention relates to a millimeter wave radar signal processing technology, and in particular to a human breathing and heartbeat monitoring method based on millimeter wave radar time domain undersampling. Background Art

[0002] Effective monitoring of human breathing and heartbeat is an important means to ensure health and can effectively prevent various lung and heart diseases. However, traditional human breathing and heartbeat monitoring mainly uses contact photoelectric means such as electrocardiogram and phonocardiogram. The human body needs to be connected to the measurement equipment for a long time, which not only limits the application scenarios but also is expensive. Existing non-contact measurement methods such as laser and WIFI have problems such as low accuracy and high hardware equipment requirements.

[0003] As a detection component with ultra-high distance resolution, millimeter-wave ultra-wideband radar can not only effectively measure the human respiratory and heartbeat frequency by accurately detecting the chest rise and fall, but also has the characteristics of non-contact. Therefore, it is of great research significance to carry out research on non-contact human respiratory and heartbeat monitoring methods based on millimeter-wave ultra-wideband radar. However, the existing human respiratory and heartbeat monitoring methods based on millimeter-wave ultra-wideband radar are all based on the phase change of the human chest echo signal. After obtaining the complete difference frequency data, it is necessary to determine the distance unit where the target is located and perform multiple phase inversion calculations based on fast Fourier transform. The calculation is complicated and the measurement error is large. There is an urgent need for a respiratory and heartbeat measurement method with small calculation amount and high accuracy. Summary of the invention

[0004] Purpose of the invention: The present invention provides a method for monitoring human respiration and heartbeat based on millimeter-wave radar time-domain undersampling, so as to overcome the problems existing in existing respiration and heartbeat monitoring methods, such as contact, high cost, large amount of calculation, and limited accuracy.

[0005] Technical solution: The method for monitoring human respiration and heartbeat based on millimeter-wave radar time domain undersampling described in the present invention comprises the following steps:

[0006] (1) Model the chest movement caused by breathing and the chest micro-motion caused by heartbeat, and calculate the mathematical analytical expression of chest displacement;

[0007] (2) Combined with the millimeter-wave ultra-wideband radar signal system, the mathematical analytical expression of the respiratory and heartbeat micro-Doppler signal is derived, and the respiratory and heartbeat micro-Doppler signal is simulated;

[0008] (3) Analyze the mathematical analytical relationship between the intermediate frequency signals of different modulation periods, clarify the working principle of time domain undersampling, use the time domain undersampling method to extract the Doppler component in the intermediate frequency signal, and obtain the time domain sampling signal of the Doppler signal modulated by micro-Doppler;

[0009] (4) Based on the different frequency ranges of the Doppler frequency corresponding to the chest movement during breathing and the micro-Doppler frequency corresponding to the chest micro-movement during heartbeat, a corresponding filter is designed to separate the respiratory Doppler signal and the heartbeat micro-Doppler signal;

[0010] (5) The frequency spectrum of the respiratory Doppler signal and the heartbeat micro-Doppler signal is calculated using Fourier transform to obtain the human respiratory frequency and heartbeat frequency, thereby realizing human respiratory and heartbeat monitoring.

[0011] Furthermore, the process of modeling the chest cavity fluctuation caused by breathing in step (1) is as follows:

[0012] The human breathing signal is modeled as a simple sinusoidal signal, and the chest fluctuation caused by the breathing signal is modeled as:

[0013] x r (t) = A r sin(2πf r t)

[0014] Among them, A r and f r are the amplitude and frequency of human breathing respectively. The initial radial distance from the radar to the human chest is R0. The instantaneous radial distance between the chest wall and the radar when only breathing is obtained is:

[0015] x r (t) = R0 + A r sin(2πf r t).

[0016] Furthermore, the process of modeling the chest micro-movement caused by heartbeat in step (1) is as follows:

[0017] The pulse signal generated by the ventricle is an exponential signal t0 is the pulse duration constant. This signal is passed through a second-order Butterworth filter with the cutoff frequency of the filter set to f0. The displacement after filtering is:

[0018]

[0019] Among them, ω0 is the angular frequency corresponding to the filter cutoff frequency f0, A h is the human heart rate, x h The signal in (t) corresponds to one heartbeat, and then this signal is multiplied by 1 / f h The periodic repetition of f h is the human heart rate, and we get the chest micro-motion signal model caused by continuous heartbeat.

[0020] Furthermore, the implementation process of step (2) is as follows:

[0021] The chest wall displacement caused by breathing and heartbeat signals is modeled as a superposition state, and the chest wall displacement caused by breathing and heartbeat is obtained as:

[0022]

[0023] The radar echo delay is expressed as:

[0024] τ=2x r (t) / c

[0025] Among them, c is the propagation speed of electromagnetic waves; τ changes with the chest displacement caused by breathing and heartbeat; the chest displacement caused by breathing will produce Doppler modulation, and the chest displacement caused by heartbeat will produce additional micro-Doppler modulation on the Doppler signal. It is only necessary to extract the Doppler signal from the radar's difference frequency signal to further solve the breathing and heartbeat frequencies.

[0026] Furthermore, the implementation process of step (2) is as follows:

[0027] The frequency of the radar signal is:

[0028]

[0029] Where, ΔF M is the modulation bandwidth, f m =1 / T M is the modulation frequency, T M is the modulation period, f0 is the carrier frequency;

[0030] The frequency of the received signal is:

[0031]

[0032] Using the relationship between frequency and phase, φ=2π∫fd t , get the instantaneous phase difference between the transmitted signal and the received signal; in a single linear frequency modulation cycle, calculate the instantaneous phase of the transmitted signal and the received signal respectively;

[0033] when When , the phase of the transmitted signal is:

[0034]

[0035] when When , the phase expression of the transmitted signal is:

[0036]

[0037] Where, C is a constant;

[0038] From the continuity of the signal phase, we know that but

[0039] when When , the phase of the received signal is:

[0040]

[0041] when When , that is, the signal is in the regular area, the instantaneous phase difference between the transmitted signal and the received signal is:

[0042] Δφ1=|φ R -φ T1 |=2π(τΔF M f m t+f0τ)

[0043] The difference frequency signal obtained by mixing the transmission signal and the echo signal is in the form of cos(Δφ n ), then the difference frequency signal in the regular area is:

[0044] cos(Δφ1)=cos2π(ΔF M f m τt+f0τ);

[0045] The difference frequency signal cos(Δφ1) in the regular region is written in the form of Fourier trigonometric series:

[0046]

[0047] in,

[0048]

[0049] make:

[0050]

[0051] Then A n =a n cos(2πf0τ), B n =b n sin(2πf0τ), A0=a0 cos(2πf0τ);

[0052] The Fourier trigonometric series decomposition of the regular region of the difference frequency signal is:

[0053]

[0054] With f s =f m / p(p∈N * ) frequency to sample e(t), then:

[0055]

[0056] in, The frequency of is much smaller than the frequency of the Doppler signal cos(2πf0τ), and e(k) can be regarded as Modulated Doppler signal;

[0057] Ensure that s =f m / p≥2f d (p∈N * ), with f s =f m / p(p∈N * ) is the sampling rate, the intermediate frequency signal is undersampled in the time domain, and the Doppler signal is extracted, f d Represents the Doppler frequency.

[0058] Furthermore, the implementation process of step (4) is as follows:

[0059] In order to eliminate the Doppler signal modulated by the chest micro-motion caused by the heartbeat d The static interference and noise interference in (t) are firstly d (t) Sliding average denoising process, get s dn (t); The Doppler frequency range of the chest fluctuation caused by breathing is 0.13Hz~0.4Hz, while the micro-Doppler frequency range of the chest micro-motion caused by heartbeat is 0.82Hz~3.3Hz; two bandpass filters H with frequency ranges of 0.1Hz~0.5Hz and 0.6Hz~4Hz are set respectively r and H h ; will s dn (t) are respectively sent to the designed filter H r and H h , filter H r The output is the Doppler signal s corresponding to the chest rise and fall during breathing r (t), filter H r The output is the micro-Doppler signal s corresponding to the micro-movement of the chest during the heartbeat h (t).

[0060] Furthermore, the implementation process of step (4) is as follows:

[0061] The Doppler signal s obtained by separation r (t) and micro-Doppler signal s h (t), perform Fourier transform respectively to obtain the spectrum S of the two signals r (f) and S h (f), find the frequency corresponding to the spectrum peak, denoted as f r and f h, then the human respiratory rate RESP is 60·f r (beats / minute), the human heart rate HR is 60 f h , realizing human breathing and heartbeat monitoring.

[0062] Beneficial effects: Compared with the prior art, the present invention has the following beneficial effects: the present invention performs under-sampling on the time domain echo at a very low sampling rate, and can directly extract accurate Doppler signals, and then separate the respiratory signal and the heartbeat signal, to achieve respiratory frequency and heart rate measurement, not only does it not need to obtain complete difference frequency data and multiple fast Fourier transforms, the hardware cost and the amount of calculation are greatly reduced, and starting from the chest rise and fall caused by breathing and heartbeat, a micro-Doppler signal model is established, and higher accuracy can be obtained while clarifying the mechanism; the present invention can invert the accurate frequency of human breathing and heartbeat by measuring the Doppler frequency and micro-Doppler frequency corresponding to the chest rise and fall in a non-contact situation. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 is a flow chart of the present invention;

[0064] Figure 2 It is the time domain waveform and spectrum of chest displacement caused by breathing;

[0065] Figure 3 The time domain waveform and spectrum of chest displacement caused by heartbeat;

[0066] Figure 4 It is the time domain waveform and spectrum of chest displacement caused by breathing and heartbeat;

[0067] Figure 5 It is the time domain waveform and spectrum of the 77GHz linear frequency modulation radar intermediate frequency signal;

[0068] Figure 6 It is the time domain waveform of the Doppler signal extracted based on the time domain undersampling method;

[0069] Figure 7 is the time domain waveform of the Doppler signal after average denoising;

[0070] Figure 8 It is the time domain waveform and spectrum of the Doppler modulation signal caused by the chest rise and fall during breathing;

[0071] Fig. 9 It is the time domain waveform and spectrum of the micro-Doppler modulation signal caused by the chest micro-movement during heartbeat. DETAILED DESCRIPTION

[0072] The present invention is further described in detail below with reference to the accompanying drawings.

[0073] like Figure 1As shown, the present invention proposes a non-contact human breathing and heartbeat monitoring method based on millimeter wave ultra-wideband radar time domain undersampling, comprising the following steps:

[0074] Step 1: Model the chest movement caused by breathing and the chest micro-movement caused by heartbeat, calculate the mathematical expression of chest displacement, combine the millimeter wave ultra-wideband radar signal system, derive the mathematical expression of respiratory and heartbeat micro-Doppler signals, and simulate to obtain respiratory and heartbeat micro-Doppler signals.

[0075] Millimeter-wave ultra-wideband radar has an ultra-high distance resolution and can effectively detect chest displacement caused by breathing and heartbeat. Human respiratory movements include inhalation and exhalation. During inhalation, the lungs expand and the chest cavity expands. During exhalation, the lungs contract and the chest wall contracts. The chest wall displacement caused by breathing is a simple expansion and contraction cycle. The amplitude of chest wall displacement caused by breathing is 4mm to 12mm, and the frequency is 0.13Hz to 0.4Hz. This displacement is relatively large relative to the wavelength of 77GHz and 60GHz radars, and the frequency is low. From the radar's perspective, it is a low-speed target and is easier to detect.

[0076] The human respiratory signal is modeled as a simple sinusoidal signal, and the chest wall displacement caused by the respiratory signal is modeled as:

[0077] x r (t) = A r sin(2πf r t)

[0078] Among them, A r and f r are the amplitude and frequency of human breathing respectively. The initial radial distance between the radar and the human chest is R0. It can be obtained that when only breathing occurs, the instantaneous radial distance between the chest wall and the radar is:

[0079] x r (t) = R0 + A r sin(2πf r t)

[0080] Human heart activity includes the contraction and relaxation of the atrial muscles and ventricles. The heartbeat causes chest wall displacement at a frequency of 0.82Hz to 3.3Hz, with an amplitude of less than 0.6mm, which is approximately one-tenth of breathing. It is an extremely weak signal and is easily affected by respiratory signals and other interferences.

[0081] Compared with the respiratory signal, the mechanism of the heartbeat is more complicated, and the heartbeat signal model is also more complicated. It can be modeled as a sine wave, a half-cycle sine pulse, a Gaussian pulse sequence, and two adjacent pulses, etc., but these signal models have large deviations from the actual signal. The present invention proposes an improved heartbeat signal model, which can be briefly described as follows: During the heartbeat, when the ventricle is in the contraction phase, it transmits a short pulse signal, which is filtered by bones and body tissues. This process is equivalent to filtering. The filtered pulse signal is transmitted to the chest wall, generating a perceptible chest wall displacement. The heartbeat signal is modeled based on this process. First, it is assumed that the pulse signal generated by the ventricle is an exponential signal. Where t0 is the pulse duration constant. This signal is passed through a second-order Butterworth filter with the cutoff frequency of the filter set to f0. The displacement after filtering is:

[0082]

[0083] Among them, ω0 is the angular frequency corresponding to the filter cutoff frequency f0, A h is the human heart rate, x h The signal in (t) corresponds to one heartbeat, and then this signal is multiplied by 1 / f h The periodic repetition of f h is the human heartbeat frequency, and the continuous-time heartbeat signal model is obtained.

[0084] The chest wall displacement caused by breathing and heartbeat signals is modeled as a superposition state, and the chest wall displacement caused by breathing and heartbeat is obtained as:

[0085]

[0086] The radar echo delay can be expressed as:

[0087] τ=2x r (t) / c

[0088] Where c is the propagation speed of electromagnetic waves, and τ changes with the chest displacement caused by breathing and heartbeat. The chest displacement caused by breathing will produce Doppler modulation, and the chest displacement caused by heartbeat will produce additional micro-Doppler modulation on the Doppler signal. It is only necessary to extract the Doppler signal from the difference frequency signal of the linear frequency modulation radar to further solve the breathing and heartbeat frequency.

[0089] Step 2: Analyze the mathematical analytical relationship between the intermediate frequency signals of different modulation periods, clarify the working principle of time domain undersampling, use the time domain undersampling method to extract the Doppler component in the intermediate frequency signal, and obtain the time domain sampling signal of the Doppler signal modulated by micro-Doppler.

[0090] Taking sawtooth linear frequency modulation radar as an example, the frequency expression of the transmitted signal can be expressed as:

[0091]

[0092] Where, ΔF M is the modulation bandwidth, f m =1 / T M is the modulation frequency, T M is the modulation period, and f0 is the carrier frequency.

[0093] The frequency expression of the received signal can be expressed as:

[0094]

[0095] Using the relationship between frequency and phase, φ=2π∫fd t , we can get the instantaneous phase difference between the transmitted signal and the received signal. In a single linear frequency modulation cycle, we can calculate the instantaneous phase of the transmitted signal and the received signal respectively.

[0096] when When , the phase expression of the transmitted signal is:

[0097]

[0098] when When , the phase expression of the transmitted signal is:

[0099]

[0100] Where C is a constant.

[0101] From the continuity of the signal phase, we know that but

[0102] when When , the phase expression of the received signal is:

[0103]

[0104] when When , that is, the signal is in the regular area, the instantaneous phase difference between the transmitted signal and the received signal is:

[0105] Δφ1=|φ R -φ T1 |=2π(τΔF M f m t+f0τ)

[0106] The difference frequency signal obtained by mixing the transmission signal and the echo signal is in the form of cos(Δφ n), then the difference frequency signal expression in the regular area is:

[0107] cos(Δφ1)=cos2π(ΔF M f m τt+f0τ)

[0108] Although the single-cycle difference frequency signal consists of regular and irregular regions, due to T M >>τ, in the process of signal analysis, the regular area is mainly studied.

[0109] The difference frequency signal cos(Δφ1) in the regular area can be written in the form of Fourier trigonometric series, as shown below:

[0110]

[0111] in:

[0112]

[0113]

[0114] For simplicity of writing, let:

[0115]

[0116] Then A n =a n cos(2πf0τ), B n =b n sin(2πf0τ), A0=a0 cos(2πf0τ).

[0117] The Fourier trigonometric series decomposition expression of the regular area part of the difference frequency signal is:

[0118]

[0119] With f s =f m / p(p∈N * ) frequency to sample e(t), then:

[0120]

[0121] in, The frequency of is much smaller than the frequency of the Doppler signal cos(2πf0τ), and e(k) can be regarded as Modulated Doppler signal.

[0122] Therefore, with f s =f m / p(p∈N *) is the sampling rate, and the Doppler signal can be extracted by undersampling the intermediate frequency signal in the time domain. However, it is still necessary to ensure that f s =f m / p≥2f d (p∈N * ), otherwise the extracted Doppler signal will be aliased. Here, f d Represents the Doppler frequency.

[0123] Step 3: Use the time domain undersampling method to extract the obtained s d (t) is the Doppler signal modulated by the chest micro-motion caused by the heartbeat. It is necessary to separate the Doppler signal caused by the chest ups and downs during breathing and the micro-Doppler signal caused by the chest micro-motion during heartbeat.

[0124] In order to eliminate d The static interference and noise interference in (t) are firstly d (t) Sliding average denoising process, get s dn (t), and then set two band-pass filters H with frequency ranges of 0.1Hz to 0.5Hz and 0.6Hz to 4Hz respectively. r and H h , the Doppler frequency range of the chest fluctuation caused by breathing is 0.13Hz~0.4Hz, and the micro-Doppler frequency range of the chest micro-movement caused by heartbeat is 0.82Hz~3.3Hz, so s dn (t) are respectively sent to the designed filter H r and H h , filter H r The output is the Doppler signal s corresponding to the chest rise and fall during breathing r (t), filter H r The output is the micro-Doppler signal s corresponding to the micro-movement of the chest during the heartbeat h (t).

[0125] Step 4: Separate the Doppler signal s r (t) and micro-Doppler signal s h (t), perform Fourier transform respectively to obtain the spectrum S of the two signals r (f) and S h (f), find the frequency corresponding to the spectrum peak, denoted as f r and f h , then the human respiratory rate RESP (Respiratory Rate) is 60·f r (beats / minute), the human heart rate HR (Heart Rate) is 60 f h (times / minute), thereby realizing human breathing and heartbeat monitoring.

[0126] In order to verify the effectiveness of the scheme of the present invention, the following simulation experiments are carried out. The system parameters are: the chest wall displacement amplitude caused by breathing is 6 mm, the frequency is 0.28 Hz, the chest wall displacement amplitude caused by heartbeat is 0.6 mm, the frequency is 1.20 Hz, the linear frequency modulation radar carrier frequency is 77 GHz, the modulation period is 25 ms, the frequency deviation is 300 MHz, and the time domain undersampling frequency is consistent with the modulation frequency at 40 Hz.

[0127] Figure 2 It is the time domain signal and spectrum of chest displacement caused by breathing. Figure 3 The time domain signal and spectrum of chest displacement caused by heartbeat, Figure 4 The time domain signal and spectrum of the chest displacement caused by breathing and heartbeat; it can be found that the amplitude and frequency of the chest displacement signal are consistent with the preset ones. Figure 5 The time domain waveform and spectrum of the 77GHz linear frequency modulation radar intermediate frequency signal are shown below. Figure 6 It is the time domain waveform of the Doppler signal extracted based on the time domain undersampling method. Figure 7 Yes Figure 6 The result after averaging and denoising the waveform in Figure 8 The time domain waveform and spectrum of the respiratory signal obtained after passing through the bandpass filter. Fig. 9 The time domain waveform and spectrum of the heartbeat signal obtained after passing through the bandpass filter. It can be found that the frequency of the breathing signal is 0.2835Hz, and the frequency of the heartbeat signal is 1.2024Hz, which is basically consistent with the theoretical value. The corresponding breathing frequency is 17 times / minute, and the corresponding heartbeat frequency is 72 times / minute, which is in line with the normal frequency range of human breathing and heartbeat.

[0128] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A human breathing and heartbeat monitoring method based on millimeter wave radar time domain undersampling, characterized in that: The following steps are involved: (1) Model the chest movement caused by breathing and the chest micro-motion caused by heartbeat, and calculate the mathematical analytical expression of chest displacement; (2) Combined with the millimeter-wave ultra-wideband radar signal system, the mathematical analytical expression of the respiratory and heartbeat micro-Doppler signal is derived, and the respiratory and heartbeat micro-Doppler signal is simulated; (3) Analyze the mathematical analytical relationship between the intermediate frequency signals of different modulation periods, clarify the working principle of time domain undersampling, use the time domain undersampling method to extract the Doppler component in the intermediate frequency signal, and obtain the time domain sampling signal of the Doppler signal modulated by micro-Doppler; (4) Based on the different frequency ranges of the Doppler frequency corresponding to the chest movement during breathing and the micro-Doppler frequency corresponding to the chest micro-movement during heartbeat, a corresponding filter is designed to separate the respiratory Doppler signal and the heartbeat micro-Doppler signal; (5) The frequency spectrum of the respiratory Doppler signal and the heartbeat micro-Doppler signal is calculated using Fourier transform to obtain the human respiratory frequency and heartbeat frequency, thereby realizing human respiratory and heartbeat monitoring.

2. According to claim 1, a method for monitoring human respiration and heartbeat based on millimeter wave radar time domain undersampling is characterized in that: The process of modeling the chest cavity fluctuation caused by breathing in step (1) is as follows: The human breathing signal is modeled as a simple sinusoidal signal, and the chest fluctuation caused by the breathing signal is modeled as: x r (t)=A r sin(2πf r t) Among them, A r and f r are the amplitude and frequency of human breathing respectively. The initial radial distance from the radar to the human chest is R0. The instantaneous radial distance between the chest wall and the radar when only breathing is obtained is: x r (t)=R0+A r sin(2πf r t)。 3. The human breathing and heartbeat monitoring method based on millimeter wave radar time domain undersampling according to claim 1 is characterized in that: The process of modeling the chest micro-movement caused by heartbeat in step (1) is as follows: The pulse signal generated by the ventricle is an exponential signal t0 is the pulse duration constant. This signal is passed through a second-order Butterworth filter with the cutoff frequency of the filter set to f0. The displacement after filtering is: Among them, ω0 is the angular frequency corresponding to the filter cutoff frequency f0, A h is the human heart rate, x h The signal in (t) corresponds to one heartbeat, and then this signal is multiplied by 1 / f h The periodic repetition of f h is the human heart rate, and we get the chest micro-motion signal model caused by continuous heartbeat.

4. The method for monitoring human respiration and heartbeat based on millimeter wave radar time domain undersampling according to claim 1, characterized in that: The implementation process of step (2) is as follows: The chest wall displacement caused by breathing and heartbeat signals is modeled as a superposition state, and the chest wall displacement caused by breathing and heartbeat is obtained as: The radar echo delay is expressed as: τ=2x r (t) / c Among them, c is the propagation speed of electromagnetic waves; τ changes with the chest displacement caused by breathing and heartbeat; the chest displacement caused by breathing will produce Doppler modulation, and the chest displacement caused by heartbeat will produce additional micro-Doppler modulation on the Doppler signal. It is only necessary to extract the Doppler signal from the radar's difference frequency signal to further solve the breathing and heartbeat frequencies.

5. The method for monitoring human respiration and heartbeat based on millimeter wave radar time domain undersampling according to claim 1, characterized in that: The implementation process of step (2) is as follows: The frequency of the radar signal is: Where, ΔF M is the modulation bandwidth, f m =1 / T M is the modulation frequency, T M is the modulation period, f0 is the carrier frequency; The frequency of the received signal is: Using the relationship between frequency and phase, φ=2π∫fd t , get the instantaneous phase difference between the transmitted signal and the received signal; in a single linear frequency modulation cycle, calculate the instantaneous phase of the transmitted signal and the received signal respectively; when When , the phase of the transmitted signal is: when When , the phase expression of the transmitted signal is: Where C is a constant; From the continuity of the signal phase, we know that but when When , the phase of the received signal is: when When , that is, the signal is in the regular area, the instantaneous phase difference between the transmitted signal and the received signal is: Δφ1=|φ R -f T1 |=2π(τΔF M f m t+f0τ) The difference frequency signal obtained by mixing the transmission signal and the echo signal is in the form of cos(Δφ n ), then the difference frequency signal in the regular area is: cos(Δφ1)=cos2π(ΔF M f m τt+f0τ); The difference frequency signal cos(Δφ1) in the regular region is written in the form of Fourier trigonometric series: in, make: A n =a n cos(2πf0τ), B n =b n sin(2πf0τ), A0=a0cos(2πf0τ); The Fourier trigonometric series decomposition of the regular region of the difference frequency signal is: With f s =f m / p(p∈N * ) frequency to sample e(t), then: in, The frequency of is much smaller than the frequency of the Doppler signal cos(2πf0τ), and e(k) can be regarded as Modulated Doppler signal; Ensure that s =f m / p≥2f d (p∈N * ), with f s =f m / p(p∈N * ) is the sampling rate, the intermediate frequency signal is undersampled in the time domain, and the Doppler signal is extracted, f d Represents the Doppler frequency.

6. The method for monitoring human respiration and heartbeat based on millimeter wave radar time domain undersampling according to claim 1, characterized in that: The implementation process of step (4) is as follows: In order to eliminate the Doppler signal modulated by the chest micro-motion caused by the heartbeat d The static interference and noise interference in (t) are firstly d (t) Sliding average denoising process, get s dn (t); The Doppler frequency range of the chest fluctuation caused by breathing is 0.13Hz~0.4Hz, while the micro-Doppler frequency range of the chest micro-motion caused by heartbeat is 0.82Hz~3.3Hz; two bandpass filters H with frequency ranges of 0.1Hz~0.5Hz and 0.6Hz~4Hz are set respectively r and H h ; will s dn (t) are respectively sent to the designed filter H r and H h , filter H r The output is the Doppler signal s corresponding to the chest rise and fall during breathing r (t), filter H r The output is the micro-Doppler signal s corresponding to the micro-movement of the chest during the heartbeat h (t).

7. The method for monitoring human respiration and heartbeat based on millimeter wave radar time domain undersampling according to claim 1, characterized in that: The implementation process of step (4) is as follows: The Doppler signal s obtained by separation r (t) and micro-Doppler signal s h (t), perform Fourier transform respectively to obtain the spectrum S of the two signals r (f) and S h (f), find the frequency corresponding to the spectrum peak, denoted as f r and f h , then the human respiratory rate RESP is 60·f r (beats / minute), the human heart rate HR is 60 f h , realizing human breathing and heartbeat monitoring.

Images