A millimeter wave radar detection method for identifying multiple-living respiration and heart rate

Abstract

The application discloses a millimeter wave radar detection method for identifying multiple life breath and heart rate. The method comprises the following steps: collecting breathing signals and heartbeat signals by using a millimeter wave radar for multiple targets, wherein the collected breathing signals and heartbeat signals are signals of the targets in different distance and different angle scenes relative to the millimeter wave radar; pre-processing the collected breathing signals and heartbeat signals to obtain corresponding frequency spectrum signals, and determining target positions through peak value detection, and then separating the breathing signals and the heartbeat signals; for the separated breathing signals and heartbeat signals, performing spectrum peak search on the frequency spectrum graphs respectively to obtain estimated breathing frequency and heartbeat frequency. The application can effectively adapt to the variable and complex home environment, and improve the accuracy and generalization ability of the breath and heart rate monitoring.

Description

Technical Field

[0001] This invention relates to the field of radar signal detection technology, and more specifically, to a millimeter-wave radar detection method for identifying the breathing and heart rate of multiple living organisms. Background Technology

[0002] Millimeter-wave radar detects respiration and heartbeat by detecting phase changes caused by chest displacement due to breathing and heartbeat. The principle behind millimeter-wave radar is that human respiration and heartbeat cause displacement changes on the surface of body parts. The chest cavity, in particular, is close to the lungs and heart; respiration and heartbeat cause significant displacement within the chest cavity, which in turn affects the phase change of the radar's received echo signal. Therefore, the chest cavity is commonly used as a location for monitoring vital signs such as respiration and heartbeat. Millimeter-wave radar enables remote, real-time detection of vital signs and timely detection of abnormalities in targets. Compared to ECG (electrocardiogram) and respiratory camera detection, millimeter-wave radar offers advantages such as non-contact operation, high privacy, and suitability for remote monitoring.

[0003] Existing millimeter-wave radar methods for detecting breathing and heartbeat typically involve three steps: First, a high-frequency millimeter-wave signal is emitted by the radar system, and the echo signal reflected from the human body is received. The echo signal contains the target object's position, velocity, and other characteristics. The echo signal is then filtered and denoised, typically using a bandpass filter to remove frequency components irrelevant to the target signal, thus reducing interference noise. Range-Doppler analysis of the echo signal yields the reflected signals of different targets. The Constant False Alarm Rate (CFAR) detection algorithm is typically used to distinguish the target from background noise. CFAR ensures accurate target detection even in complex environments (such as multi-target interference). Using the detected target information (e.g., time delay, frequency shift), the target's spatial location (including range and angle) is determined. This step is usually achieved through Direction of Arrival (DOA) estimation.

[0004] Then, after target detection, the radar system continuously monitors the target's phase changes to capture the minute displacements caused by breathing and heartbeats. When a person breathes, their chest or abdomen causes periodic, minute movements relative to the radar, and a heartbeat also results in minute displacement changes. By analyzing the phase changes of these displacements, information about physiological signals can be obtained. Due to the periodic characteristics of breathing and heartbeats, the phase signal in radar echoes typically exhibits periodic changes. Phase demodulation techniques can be used to extract these periodic phase changes from the echo signal, allowing for further analysis of the frequency components within the signal.

[0005] Finally, using Minimum Variance Distortionless Response (MVDR), frequency domain analysis is performed on the phase-processed signal to identify and extract multiple signal sources (such as respiratory and heartbeat signals), providing accurate frequency estimates. Then, based on the frequency and amplitude information of the extracted respiratory and heartbeat signals, specific physiological parameters, such as respiratory rate and heart rate, are calculated. This data can serve as an important basis for monitoring human health.

[0006] In existing technologies, patent application CN118592921A (a millimeter-wave radar detection method for heart rate and respiratory rate based on CNN fusion features) describes a method for detecting heart rate and respiratory rate using millimeter-wave radar based on CNN fusion features. This method utilizes real-time one-dimensional radar signals obtained by monitoring the vital signs of a target person through a millimeter-wave radar sensor, performs time-frequency analysis to obtain two-dimensional time-frequency signals, and then uses these signals to detect the target person's heart rate and respiratory rate. By leveraging the key information about human vital signs contained in the multi-dimensional signals, it achieves accurate and reliable detection of heart rate and respiratory rate, addressing the shortcomings of existing contact-based detection methods that restrict user activity, cause discomfort, and have limited application to specific populations. This method aims to achieve rapid and accurate detection of human heart rate and respiratory rate, improving the accuracy and reliability of the detection results. However, in practical applications, the detection effect is still insufficient for clinical testing.

[0007] Analysis reveals that existing millimeter-wave radar detection technology has the following main shortcomings: (1) Millimeter-wave radar has the problem of inaccurate target identification when detecting multiple people's breathing and heart rate in complex environments. Traditional detection algorithms are usually based on a single distance for target detection, but when multiple targets are at the same or close distance, the system cannot distinguish them based on distance alone, resulting in ineffective breathing and heart rate monitoring. Therefore, the number of targets cannot be predicted, making accurate detection more difficult. In addition, there are a lot of clutter in complex environments, such as tables, walls, chairs, and electrical appliances. These objects also reflect millimeter-wave signals, which overlap with the reflected signals of human targets, making it impossible for the system to distinguish human targets from other objects. Due to these problems, the accuracy of millimeter-wave radar in detecting multiple people's breathing and heart rate in complex environments is low, limiting its widespread application in daily life.

[0008] (2) The key to monitoring respiration and heart rate using millimeter-wave radar lies in accurately extracting respiratory and heart rate signals. Currently, the method for extracting these two signals is usually to first separate the respiratory and heart rate signals using filters, and then use the same algorithm to extract these two signals separately. However, since the amplitude of the heart rate signal is smaller than that of the chest cavity displacement caused by respiration, the higher harmonic frequencies of respiration may fall into the frequency range of the heart rate signal. In particular, the amplitudes of the second, third, and fourth harmonics are relatively large, which will cause certain interference to the frequency analysis of the heart rate signal, thus affecting the accurate detection of respiratory rate and heart rate.

[0009] In summary, traditional millimeter-wave radar-based methods for detecting respiratory and heart rates in multiple individuals employ a single-spectrum estimation algorithm. This algorithm is limited by its ability to separate and extract features from respiratory and heartbeat signals. While it may be more accurate at extracting respiratory signals, it may be less accurate at extracting heartbeat signals, or vice versa. This is particularly true for multi-person and long-distance detection, where accuracy is low. In recent years, traditional methods for detecting respiratory and heartbeat signals have improved accuracy, but this is limited to single-person detection within 1.5 meters. For long-distance multi-person detection, the accuracy and stability of these methods still face challenges, especially under the influence of signal interference and multipath effects, which can lead to significant errors in the detection results. Summary of the Invention

[0010] The purpose of this invention is to overcome the shortcomings of the prior art and provide a millimeter-wave radar detection method for identifying the respiration and heart rate of multiple living organisms. This method includes the following steps: For multiple targets, millimeter-wave radar is used to collect respiratory and heartbeat signals. The collected respiratory and heartbeat signals are the signals of the target at different distances and angles relative to the millimeter-wave radar. The collected respiratory and heartbeat signals are preprocessed to obtain the corresponding spectral signals, and the target location is determined by peak detection, thereby separating the respiratory and heartbeat signals. For the separated respiratory and heartbeat signals, spectral peak search is performed on the spectrograms to obtain the estimated respiratory and heartbeat frequencies.

[0011] Compared with existing technologies, the advantages of this invention lie in its proposed method for extracting respiratory and heartbeat signals from multiple individuals based on millimeter-wave radar. First, a peak detection algorithm based on a threshold is used to determine the target's location. Then, phase is extracted and processed. Finally, multi-signal classification and two-dimensional fast Fourier transform are combined to extract respiratory and heartbeat signals separately. This invention detects the respiratory and heart rates of multiple individuals using millimeter-wave radar, determining whether their vital signs are abnormal. This helps medical personnel and family members promptly understand their health status, ensuring rapid rescue measures can be taken in case of abnormalities. Accurate detection can be achieved even in complex environments such as indoors.

[0012] Other features and advantages of the invention will become clear from the following detailed description of exemplary embodiments of the invention with reference to the accompanying drawings. Attached Figure Description

[0013] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments of the invention and, together with their description, serve to explain the principles of the invention.

[0014] Figure 1 This is a flowchart of a millimeter-wave radar detection method for identifying the breathing and heart rate of multiple living organisms according to an embodiment of the present invention; Figure 2 This is a schematic diagram of an experimental scenario according to an embodiment of the present invention; Figure 3 This is a schematic diagram of peak detection according to an embodiment of the present invention; Figure 4 This is a schematic diagram of amplitude spectrum and time-distance heatmap according to an embodiment of the present invention. Detailed Implementation

[0015] Various exemplary embodiments of the present invention will now be described in detail with reference to the accompanying drawings. It should be noted that, unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps set forth in these embodiments do not limit the scope of the invention.

[0016] The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit the invention or its application or use.

[0017] Techniques, methods, and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and equipment should be considered part of the specification.

[0018] In all the examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values.

[0019] It should be noted that similar labels and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be discussed further in subsequent figures.

[0020] To address the issues of inaccurate target detection and low accuracy of existing millimeter-wave radar in detecting the breathing and heart rate of multiple living individuals, this invention provides a millimeter-wave radar detection method for identifying the breathing and heart rate of multiple living beings. Combined with... Figure 1 and Figure 2 As shown, the method includes the following steps: Step S110: For multiple targets, use millimeter-wave radar to collect breathing signals and heartbeat signals. The collected breathing signals and heartbeat signals are signals of the targets at different distances and angles relative to the millimeter-wave radar.

[0021] See Figure 3 The experimental scenario shown illustrates the detection of two targets. Two subjects (labeled Target 1 and Target 2) were positioned at different distances and angles relative to the millimeter-wave radar. The distances between the subjects and the radar were 0.5m, 1.0m, 1.5m, 2.0m, and 3.0m, with angles set to -60°, -45°, -30°, -15°, 0°, 15°, 30°, 45°, and 60°. The experiment first determined the initial distances to the subjects and then adjusted their angles while maintaining these distances.

[0022] The parameter settings for the millimeter-wave radar in the experiment are shown in Table 1 below.

[0023] Table 1: Radar Parameter Settings

[0024] Before the experiment, participants were required to ensure their breathing and heart rate were stable. At the start of the experiment, participants remained seated while millimeter-wave radar transmitted signals to them to collect breathing and heart rate signals for 30 seconds. A total of 48 sets of data were collected.

[0025] Step S120: The collected respiratory and heartbeat signals are preprocessed to obtain the corresponding spectral signals, and the target position is determined by peak detection, thereby separating the respiratory and heartbeat signals.

[0026] In step S120, the acquired signal is subjected to target detection and phase processing to separate the respiratory signal and heartbeat signal, as detailed in the following steps.

[0027] Step S11: The raw AD data of the intermediate frequency signal is acquired by the millimeter-wave radar and saved in binary format. The experiment collected 30 seconds of data, [number of frames]. Frame, each frame of data sampling In the preprocessing stage, the original AD data is processed into a 128×600×8 dimensional matrix. A 128-point 1D-FFT is performed on each column of the third dimension to obtain a 128×600×8 dimensional 1D-FFT spectrum. j is the imaginary unit, and the 1D-FFT formula is as follows: (1) Step S12: Compare 128 data points in the 1D-FFT (One-Dimensional Fast Fourier Transform) amplitude spectrum of the intermediate frequency signal, find the point with the largest amplitude, and record it as... This point can be considered the location of a potential target. Next, one-third of this maximum value is taken as the threshold for peak detection, denoted as . Then, within the effective area (e.g., within 3 meters), for each peak point... By comparing and combining the time-distance heatmaps, the precise location of the target was finally determined. As shown in Figure 4, where... Figure 4 (a) is the 1D-FFT amplitude spectrum of the intermediate frequency signal (first 3 meters), with the horizontal axis representing distance (m) and the vertical axis representing amplitude (coordinate value * 10). 4 ), Figure 4 (b) is a time-distance heatmap, with the horizontal axis representing time (s) and the vertical axis representing distance (m).

[0028] By using a peak detection algorithm based on one-third threshold of maximum echo intensity, combined with distance and angle information, the target can be accurately located, effectively reducing the interference of random human motion noise on target localization and tracking.

[0029] Step S13: After determining the frequency point corresponding to the target, it can be converted into a range cell. Based on the effective cell range of the target, the intermediate frequency signal phases of the two range cells occupied by the two targets are first extracted, phase summed, and then phase expanded. The phase extraction formula is: (2) in, It refers to phase, and the unit is usually radians. It is the real part of the complex number X. It is the imaginary part of the complex number X.

[0030] The phase expansion formula is: (3) in, Indicates will Round to the nearest integer. This represents the phase of the nth data point.

[0031] Step S14: Calculate the first-order difference of the phase.

[0032] First-order difference refers to the difference between two sets of data in a given data sequence. In this context, the difference between each data point and the previous data point is represented by the following formula: (4) in, It is the first The values ​​of each data point. It is the first The difference between points represents the amount of change between adjacent data points.

[0033] Step S15: After obtaining the first-order difference result, perform a moving average filter on the result.

[0034] Moving average filtering is one of the simplest and most commonly used digital filtering methods for smoothing signals and reducing noise, especially suitable for time series or discrete signals. The moving average reduces random fluctuations by calculating the average of adjacent data points. The formula is as follows: (5) in, It is the window size (i.e., how many points to take each time to calculate the average). It is the present and the future One sample. It is the filtered output.

[0035] By employing phase accumulation technology, the detectability and reliability of respiratory and heartbeat signals are significantly improved under low signal strength conditions, while suppressing external noise and interference. This ensures that target signals can be accurately identified and analyzed in complex signal environments, thereby improving the overall signal processing effect and accuracy.

[0036] Step S16: Separate the breathing signal and heartbeat signal using a breathing filter and a heartbeat filter.

[0037] For example, the breathing filter uses an elliptic bandpass filter with a frequency of 0.1 to 0.5 Hz. The heartbeat filter uses an elliptic bandpass filter with a frequency of 0.8 to 2.0 Hz. See Table 2 for the parameter settings of the breathing filter and Table 3 for the parameter settings of the heartbeat filter.

[0038] Table 2: Parameter settings for the breathing filter

[0039] Table 3: Parameter settings for the heartbeat filter

[0040] Step S130: For the separated respiratory signal and heartbeat signal, perform spectral peak search of the spectrogram to obtain the estimated respiratory frequency and heartbeat frequency.

[0041] In step S130, the respiratory rate and heart rate are estimated by analyzing the separated respiratory and heart rate signals respectively. See the following description for details.

[0042] (1) Respiratory signal processing For the respiratory signal, the MUSIC (spatial spectrum estimation algorithm) algorithm is applied to the bandpass filtered respiratory signal, and the peak search is performed on the MUSIC algorithm spectrum. The frequency corresponding to the maximum peak is the respiratory frequency.

[0043] In one embodiment, respiratory rate is estimated according to the following steps: Step S21: Using the respiratory signal after elliptic bandpass filtering as the observation sample, estimate the autocorrelation matrix of the sample. .

[0044] Assume the order of the autocorrelation matrix is The number of observed samples is The observed sample sequence is as follows: (6) The autocorrelation matrix of the observed samples is calculated as follows: (7) Step S22: For the estimated autocorrelation matrix Eigenvalue decomposition is performed to extract basis vectors of the noise subspace.

[0045] Assume the number of targets to be detected is Then, after eigenvalue decomposition, Chinese correspondence The normalized eigenvectors of the smallest eigenvalues ​​constitute a set of basis vectors for the noise subspace.

[0046] Step S23: Use the MUSIC (spatial spectrum estimation algorithm) algorithm in combination with spectral peak search to perform spectrum estimation.

[0047] For example, the pseudo-power spectrum expression of the MUSIC algorithm is: MUSIC (8) exist Internal change The drawn curve MUSIC The peak position corresponds to the estimated signal frequency. Among them, , called the angular frequency vector, represents the signal's response in the frequency domain. These are the eigenvectors of the noise subspace. It is the normalized angular frequency to be searched. The autocorrelation matrix is... The normalized eigenvectors corresponding to the smallest eigenvalues ​​constitute the noise subspace.

[0048] (2) Heartbeat signal processing Step S31: After filtering the heartbeat signal, the following is obtained: The target signal is denoted as . A two-dimensional fast Fourier transform (2D-FFT) is performed on the signal.

[0049] First, construct a two-dimensional signal matrix, which can be a two-dimensional array composed of signal data from multiple time periods: (9) here For the frequency domain representation of the signal, and These are spatial coordinates in the frequency domain.

[0050] Step S32: Calculate the spectrum.

[0051] Specifically, the spectrum It is the amplitude spectrum of the signal: (10) Step S33: Peak search.

[0052] Specifically, for the 2D-FFT spectrogram Perform a spectral peak search to find the position of the peak with the largest amplitude. The position of the largest peak corresponds to the main frequency component of the signal. Estimate the heart rate using the frequency position of the largest peak.

[0053] In one embodiment, heart rate It can be estimated using the following formula: (11) in, It is the sampling interval of the signal on the frequency axis. It is the amplitude of body surface displacement caused by heartbeat.

[0054] In summary, using the MUSIC algorithm for spectral estimation of respiratory signals, even when signal frequencies are extremely close, the MUSIC algorithm can still provide high-precision frequency estimation, thereby achieving effective signal separation and identification. Simultaneously, employing the 2D-FFT algorithm for spectral estimation of heartbeat signals effectively reduces the impact of respiratory signal harmonics on heartbeat signals, ensuring the accuracy and stability of heartbeat signals.

[0055] To further verify the actual effect of the present invention, multiple sets of experiments were conducted in an indoor environment. The following are the experimental results of multiple sets of respiratory rate and heart rate, see Table 4 and Table 5.

[0056] Table 4: Comparison of radar-monitored respiratory rate and actual value (bpm)

[0057] Table 5: Comparison of radar-monitored heart rate and actual value (bpm)

[0058] It should be noted that, without departing from the spirit and scope of this invention, those skilled in the art can make appropriate changes or modifications to the above embodiments. For example, when performing target detection, to exclude irrelevant targets, the threshold-based peak detection algorithm described above can be used. Alternatively, multiple potential target phase information can be extracted simultaneously and denoised. If the final waveform is too smooth, it can be determined to be a static object, thus excluding this potential target and confirming other waveforms as human targets. As another example, the above-mentioned use of the MUSIC algorithm to extract respiratory signals and the 2D-FFT algorithm to extract heartbeat signals can also be replaced by the MVDR algorithm. The MVDR algorithm maximizes the energy of the target signal and suppresses interference from noise and other signal sources by adjusting the array weighting coefficients. When extracting respiratory and heartbeat signals, the MVDR algorithm can accurately separate the target signal, avoiding interference from other signals. Furthermore, peak thresholds, filter parameter settings, radar parameter settings, etc., can all be appropriately adjusted according to actual needs or performance requirements.

[0059] In summary, compared with the prior art, the present invention has the following advantages: (1) By combining a threshold-based peak detection algorithm with angular distance measurement technology, precise target localization can be achieved. The threshold peak detection algorithm can accurately identify peaks in the signal by setting an appropriate threshold, thereby effectively extracting the features of the target signal, filtering background noise, and enhancing positioning accuracy. Compared with existing technologies, it enhances the sensitivity to signal peaks and can effectively suppress false signals and noise, making target localization more accurate and reliable, especially in complex or noisy environments.

[0060] (2) Since the target's volume usually occupies multiple distance units, the respiratory and heartbeat signals of the target are enhanced by accumulating phase information. In this way, the detection intensity of the signal can be effectively improved, ensuring clearer target localization and signal extraction. Compared with traditional methods, this phase accumulation method makes signal extraction clearer and more accurate, especially effective in low signal-to-noise ratio environments, and significantly improves the quality of signal extraction.

[0061] (3) Moving average filtering suppresses high-frequency noise and eliminates transient interference caused by equipment or environment, making the signal more stable and reliable. Moving average filtering can remove environmental noise, equipment noise, and physiological interference (such as muscle tremors, motion artifacts, etc.). By smoothing the signal waveform and reducing abrupt changes, the signal becomes more stable, which is helpful for subsequent analysis and processing.

[0062] (4) Using an IIR elliptic bandpass filter of 0.1-0.5Hz to bandpass filter the signal can effectively preserve the breathing signal while filtering out the heartbeat signal and other noise, thereby effectively reducing the subsequent spectrum estimation error and improving the accuracy of signal processing.

[0063] (5) Since respiratory signals generate many harmonics, when the frequencies of these harmonics are close to those of the heartbeat signal, the harmonics of respiratory signals may be misinterpreted as heartbeat signals. Therefore, when the harmonic frequencies are close to those of the heartbeat signal, the harmonics of respiratory signals can significantly interfere with the heartbeat signal. To address this, an IIR elliptic bandpass filter of 0.8-2.0 Hz can be used to bandpass filter the signal, thereby effectively preserving the heartbeat signal and filtering out interference.

[0064] (6) The MUSIC algorithm is used to estimate the spectrum of the respiratory signal, and 2D-FFT is used to estimate the spectrum of the heartbeat signal. Compared with existing technologies, this combination can provide high-precision frequency estimation when the signal frequencies are very close, ensuring effective signal separation and identification. Especially in complex frequency environments, it can provide more accurate respiratory rate and heart rate data, avoiding the spectral aliasing phenomenon that may occur in traditional methods.

[0065] In summary, this invention effectively reduces the impact of environmental noise on measurement results by precisely filtering echo signals generated by surrounding objects (such as tables, chairs, and walls), thereby enabling accurate detection of the distance and angle of the target object. This method maintains high detection accuracy even in complex environments. Furthermore, by applying a series of filtering techniques to remove noise interference and combining multiple advanced algorithms to extract respiratory and heart rate signals, the detection accuracy of respiratory rate and heart rate is significantly improved. It outperforms existing technologies in terms of robustness and generalization ability, effectively adapting to varied and complex home environments and providing high-precision respiratory and heart rate monitoring functions.

[0066] This invention can be a system, method, and / or computer program product. A computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for causing a processor to implement various aspects of the invention.

[0067] Computer-readable storage media can be tangible devices capable of holding and storing instructions for use by an instruction execution device. Computer-readable storage media can be, for example, but not limited to, electrical storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any suitable combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable compact disc read-only memory (CD-ROM), digital multifunction disc (DVD), memory sticks, floppy disks, mechanical encoding devices, such as punch cards or recessed protrusions storing instructions thereon, and any suitable combination thereof. The computer-readable storage media used herein are not to be construed as transient signals themselves, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media (e.g., light pulses through fiber optic cables), or electrical signals transmitted through wires.

[0068] The computer-readable program instructions described herein can be downloaded from computer-readable storage media to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external storage device. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to the computer-readable storage media in the respective computing / processing device.

[0069] The computer program instructions used to perform the operations of this invention may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, Python, etc., and conventional procedural programming languages ​​such as "C" or similar languages. The computer-readable program instructions may be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuitry, such as programmable logic circuitry, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), is personalized by utilizing state information from the computer-readable program instructions. This electronic circuitry can execute the computer-readable program instructions to implement various aspects of the invention.

[0070] Various aspects of the present invention are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer-readable program instructions.

[0071] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that, when executed by the processor of the computer or other programmable data processing apparatus, they create means for implementing the functions / actions specified in one or more blocks of the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium that causes a computer, programmable data processing apparatus, and / or other device to operate in a particular manner; thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing aspects of the functions / actions specified in one or more blocks of the flowchart and / or block diagram.

[0072] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to perform the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0073] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of an instruction containing one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions. It will be known to those skilled in the art that implementation in hardware, implementation in software, and implementation using a combination of software and hardware are equivalent.

[0074] The various embodiments of the present invention have been described above. These descriptions are exemplary and not exhaustive, and are not limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical application, or technical improvements to the embodiments in the market, or to enable others skilled in the art to understand the embodiments disclosed herein. The scope of the invention is defined by the appended claims.

Claims

1. A millimeter-wave radar detection method for identifying respiration and heart rate in multiple living organisms, comprising the following steps: Data acquisition steps: For multiple targets, use millimeter-wave radar to collect respiratory and heartbeat signals. The collected respiratory and heartbeat signals are the signals of the target at different distances and angles relative to the millimeter-wave radar. Target detection and phase processing steps: The collected respiratory and heartbeat signals are preprocessed to obtain the corresponding spectral signals, and the target position is determined by peak detection, thereby separating the respiratory and heartbeat signals; Signal extraction steps: For the separated respiratory and heartbeat signals, perform spectral peak search of the spectrograms to obtain the estimated respiratory and heartbeat frequencies.

2. The method according to claim 1, characterized in that, The target detection and phase processing steps include: The collected respiratory and heartbeat signals were processed into matrices of a set dimension, and then converted into 1D-FFT spectrum results using one-dimensional fast Fourier transform. By comparing the data points in the 1D-FFT spectrum results, the point with the largest amplitude is found and denoted as... The location of potential targets is used as a reference. Then, within the set effective area, each peak point is compared with the set peak detection threshold, and combined with the time-distance heatmap, the frequency point corresponding to the target is determined. The frequency point corresponding to the target is converted into a range cell. The intermediate frequency signal phase of two range cells occupied by multiple targets is extracted, the phase is accumulated, and the phase is expanded. Based on the phase expansion processing result, the first-order difference of the phase is calculated to obtain the first-order difference result; The first-order difference result is subjected to moving average filtering to obtain the filtered result. The filtered result is used to separate the breathing signal and the heartbeat signal using a breathing filter and a heartbeat filter.

3. The method according to claim 1, characterized in that, In the signal extraction step, the frequency of the respiratory signal is estimated according to the following sub-steps: Using the separated heartbeat signals as observation samples, estimate the autocorrelation matrix of the samples. ; For autocorrelation matrix Eigenvalue decomposition is performed to extract basis vectors from the noise subspace; The power spectrum was calculated using the MUSIC algorithm, and in Internal change The power spectrum curve is obtained, and then the signal corresponding to the peak position is determined by the power spectrum curve, which is used as the frequency of the estimated respiratory signal.

4. The method according to claim 1, characterized in that, In the signal extraction step, the frequency of the heartbeat signal is estimated according to the following sub-steps: The separated heartbeat signal As the target signal, a two-dimensional fast Fourier transform is performed to obtain a two-dimensional array composed of signal data from multiple time periods, represented as follows: in, For the frequency domain representation of the signal, and These are spatial coordinates in the frequency domain; Calculate the spectrum of the signal using the following formula. : Based on spectrogram Perform a spectral peak search to find the peak position with the largest amplitude, and estimate the heart rate by using the frequency position of the largest peak.

5. The method according to claim 3, characterized in that, The power spectrum curve is calculated based on the following formula: MUSIC in, It is an angular frequency vector. The autocorrelation matrix is... The normalized eigenvectors corresponding to the smallest eigenvalues ​​constitute the noise subspace. It is the normalized angular frequency to be searched. These are the eigenvectors of the noise subspace. It refers to the window size.

6. The method according to claim 2, characterized in that, The peak detection threshold is set to one-third of the maximum peak value.

7. The method according to claim 2, characterized in that, The breathing filter uses an elliptic bandpass filter from 0.1 Hz to 0.5 Hz, and the heartbeat filter uses an elliptic bandpass filter from 0.8 Hz to 2.0 Hz.

8. The method according to claim 4, characterized in that, The heart rate is estimated using the following formula: in, It is the sampling interval of the signal on the frequency axis. It is the amplitude of body surface displacement caused by heartbeat.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 8.

10. A computer device comprising a memory and a processor, wherein a computer program capable of running on the processor is stored in the memory, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 8.

Images