Vital sign monitoring method based on multilayer variational mode decomposition and reconstruction

Abstract

The invention discloses a vital sign monitoring method based on multilayer variational mode decomposition and reconstruction, which comprises the following steps: sending a radar microwave signal to a target human body, receiving a path of returned radar microwave reflection signal, and extracting frequency domain characteristics of the radar microwave reflection signal; judging whether the current human body is in a resting state or not according to the radar microwave reflection signals, and if so, extracting the radar microwave reflection signals in the resting state; carrying out respiratory signal and heart rate signal separation reconstruction on the radar microwave reflection signal in the resting state to obtain a reconstructed respiratory signal and a reconstructed heart rate signal; phase data of the reconstructed respiration signals and the reconstructed heart rate signals are extracted respectively, wavelet transformation is conducted on the phase data of the respiration signals and the phase data of the heart rate signals respectively, respiration phase characteristics and heart rate phase characteristics are obtained, and the respiration frequency and the heart rate are calculated according to the respiration phase characteristics and the heart rate phase characteristics.

Description

Technical Field

[0001] The present invention relates to the technical field of vital sign monitoring, and in particular to a vital sign monitoring method based on multi-layer variational mode decomposition and reconstruction. Background Art

[0002] Microwave radar is a non-contact sensing technology that can be used to detect human vital signs, such as breathing rate and heart rate. Generally, the radar transmitter transmits electromagnetic wave signals to the human target. When the electromagnetic wave signals touch the human body, they will be reflected back to the radar receiver. These reflected electromagnetic wave signals will produce a weak frequency shift due to human breathing and other movements. The human breathing rate and heart rate can be obtained by analyzing the phase of these frequency shifts.

[0003] In actual environments, there will be people walking around the detected person, especially in hospitals or nursing homes. The movement of people will cause great interference to microwave radar monitoring, resulting in large errors in the subsequent calculation of respiratory frequency and heart rate. In addition, when a person is lying in bed, the respiratory frequency and heart rate monitoring is most accurate only when the person is in a resting state. However, existing microwave radar monitoring is difficult to accurately distinguish between the resting or moving state of the human body, and the respiratory and heart rate signals are often mixed together, and are accompanied by interference from environmental noise and other physiological activities. The respiratory frequency and heart rate calculated directly using the microwave radar monitoring data without making any distinction are inaccurate.

[0004] In addition, the existing technology usually uses wavelet decomposition to process the phase data of microwave radar monitoring data. However, different microwave radar monitoring data correspond to different wavelet decomposition types. If only a single wavelet decomposition algorithm is used, the calculation of respiratory frequency and heart rate will also be inaccurate.

[0005] In view of the above problems in the prior art, the purpose of the present invention is to design a vital sign monitoring method based on multi-layer variational mode decomposition and reconstruction. Summary of the invention

[0006] In view of this, the purpose of the present invention is to propose a vital sign monitoring method based on multi-layer variational mode decomposition and reconstruction, which can solve the above-mentioned problems.

[0007] The present invention provides a vital sign monitoring method based on multi-layer variational mode decomposition and reconstruction, comprising:

[0008] Send a radar microwave signal to the target human body, receive a returned radar microwave reflection signal, and extract the frequency domain characteristics of the radar microwave reflection signal;

[0009] Judging whether the human body is in a resting state according to the radar microwave reflection signal, if so, extracting the radar microwave reflection signal in the resting state;

[0010] Separating and reconstructing the respiratory signal and the heart rate signal of the radar microwave reflection signal in a resting state to obtain the reconstructed respiratory signal and the heart rate signal;

[0011] The phase data of the reconstructed respiratory signal and heart rate signal are extracted respectively, and the phase data of the respiratory signal and the heart rate signal are subjected to wavelet transform respectively to obtain respiratory phase features and heart rate phase features, and the respiratory frequency and heart rate are calculated according to the respiratory phase features and the heart rate phase features.

[0012] Further, the extracting frequency domain features of the radar microwave reflection signal includes:

[0013] Performing intermediate frequency signal sampling on the radar microwave reflection signal to obtain a radar microwave digital signal;

[0014] Perform 1D FFT calculation on the radar microwave digital signal to obtain the one-dimensional frequency domain characteristics of the radar microwave digital signal.

[0015] Further, judging whether the current human body is in a resting state according to the radar microwave reflection signal includes:

[0016] If the one-dimensional frequency domain feature of the radar microwave digital signal exceeds the first body motion threshold, there is someone in the current monitoring area, otherwise there is no one in the current monitoring area;

[0017] According to the radar microwave digital signal indicating that there is someone in the current monitoring area, it is judged whether the current human body is in a resting state.

[0018] Further, judging whether the current human body is in a resting state according to the radar microwave digital signal indicating that there is a person in the current monitoring area includes:

[0019] Extracting a frequency sweep signal at a specific time interval from the distance characteristics of the radar microwave digital signal;

[0020] Perform 2D FFT calculation on the frequency sweep signal to obtain the two-dimensional frequency domain characteristics of the radar microwave digital signal;

[0021] If the two-dimensional frequency domain feature of the radar microwave digital signal exceeds the second body motion threshold, the human body is in a body motion state, otherwise the human body is in a resting state.

[0022] Furthermore, the first body movement threshold and the second body movement threshold are dynamically adjusted respectively by a constant false alarm rate algorithm.

[0023] Further, the separation and reconstruction of the respiratory signal and the heart rate signal of the radar microwave reflection signal in the resting state to obtain the reconstructed respiratory signal and the heart rate signal includes:

[0024] Performing multi-layer variational mode decomposition on the radar microwave digital signal in a static state to obtain each decomposition component at a specified number of layers;

[0025] Using known human physiological characteristics to determine the frequency range of vital signs, and filtering out a respiratory decomposition component and a heart rate decomposition component from each decomposition component according to the frequency range of the vital signs;

[0026] The filtered respiratory decomposition components and heartbeat decomposition components are superimposed and reconstructed to obtain separated respiratory signals and heart rate signals.

[0027] Furthermore, before the phase data of the respiratory signal and the heart rate signal are subjected to wavelet transformation respectively, the following steps are performed:

[0028] The reconstructed respiratory signal and heart rate signal are input into a pre-trained wavelet basis recognition model to obtain the corresponding wavelet basis type and decomposition layer number.

[0029] Furthermore, the wavelet basis recognition model is trained by the following steps:

[0030] The signal energy, frequency distribution, and instantaneous phase of several respiratory signals and heart rate signals are collected to form samples, and the corresponding wavelet basis types and decomposition layers of the samples are calibrated to form a wavelet basis recognition sample training set;

[0031] The wavelet basis recognition model is constructed by a neural network model, and the wavelet basis recognition model is trained by a wavelet basis recognition sample training set to obtain the parameters of the wavelet basis recognition model.

[0032] Further, the phase data of the respiratory signal and the heart rate signal are respectively subjected to wavelet transformation to obtain the respiratory phase feature and the heart rate phase feature, including:

[0033] According to the corresponding wavelet basis type and decomposition level, the heart rate phase data and the respiratory phase data are respectively decomposed by wavelet to obtain the heart rate spectrum and the respiratory spectrum;

[0034] The local maxima of the heart rate spectrum and the respiratory spectrum are identified respectively, and the respiratory phase features and the heart rate phase features are extracted respectively.

[0035] Further, the calculating of the respiratory frequency and the heart rate according to the respiratory phase characteristics and the heart rate phase characteristics comprises:

[0036] Performing inverse calculation on the respiratory phase characteristics and the heart rate phase characteristics to obtain reconstructed respiratory frequency and heart rate;

[0037] The reconstructed respiratory frequency and heart rate are averaged and smoothed to obtain stable respiratory frequency and heart rate.

[0038] Beneficial effects of the present invention:

[0039] First, through the use of a transmit-and-receive frequency-modulated continuous wave radar, by sampling the radar microwave reflection signal at an intermediate frequency, the accuracy of the microwave signal in the process of converting from analog to digital can be ensured, effectively eliminating the uncertainty and noise interference in the analog signal processing; the use of frequency domain features can significantly improve the ability to detect and identify targets, especially for micro-motion monitoring or monitoring of subtle displacements, frequency domain features can provide intuitive frequency and amplitude information, thereby enhancing monitoring sensitivity and accuracy.

[0040] Secondly, the two body motion thresholds can exclude the frequency domain range of non-human targets (for example, no one in the bed) or large non-periodic motion signals (for example, people walking), and exclude the frequency domain signals of large body movements (for example, people turning over, etc.). It can effectively remove strong interference signals mixed in weak respiratory and heart rate signals. It is suitable for scenarios with multiple people. When other people pass by the bed, it can ensure that the signal is not misjudged, and obtain the most accurate human resting frequency range.

[0041] Third, by using variational mode decomposition (VMD) technology, the signal is decomposed into multiple modal components (IMF) to extract signals with different frequency characteristics, and each component corresponds to the frequency component of a specific frequency band; the known human breathing and heart rate ranges are used to guide the extraction of frequency components of these signals. Breathing is usually in a lower frequency band (adults: 0.2Hz-0.33Hz), while the heart rate is in a relatively high frequency band (adults: 1Hz-1.67Hz); from the components obtained from VMD, the parts belonging to the breathing and heart rate frequency bands are extracted, and the selected modal components that meet the breathing and heart rate frequency band conditions are superimposed and merged respectively to form separated breathing and heart rate signals, so that the breathing and heart rate signals can be more accurately separated and analyzed independently, thereby improving the accuracy of physiological status monitoring.

[0042] Fourth, by using the corresponding wavelet basis and decomposition level, it helps to analyze the signal phase data at different scales, so as to accurately separate the frequency components of heart rate and breathing. By inverting the extracted phase features, the reconstructed breathing frequency and heart rate are restored, providing redundant information, which helps to verify the reliability of the results and improve the resistance to noise and anomalies. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0044] Figure 1 It is a flow chart of the main steps of the first embodiment.

[0045] Figure 2 It is an overall step flow chart of the first embodiment. DETAILED DESCRIPTION

[0046] To facilitate understanding by those skilled in the art, the structure of the present invention is now further described in detail with reference to the embodiments and the accompanying drawings. It should be understood that the steps mentioned in the present embodiment, unless otherwise specified in their order, can be adjusted in order according to actual needs, and can even be executed simultaneously or partially simultaneously.

[0047] Embodiment 1

[0048] like Figure 1 and Figure 2 As shown, an embodiment of the present invention provides a vital sign monitoring method based on multi-layer variational mode decomposition and reconstruction, comprising:

[0049] S1 sends a radar microwave signal to the target human body, receives a returned radar microwave reflection signal, and extracts the frequency domain characteristics of the radar microwave reflection signal;

[0050] S101 performs intermediate frequency signal sampling on the radar microwave reflection signal to obtain a radar microwave digital signal;

[0051] S102 performs 1D FFT calculation on the radar microwave digital signal to obtain a one-dimensional frequency domain feature of the radar microwave digital signal.

[0052] In this step, the radar is installed at an appropriate position according to the position of the human body and the bed, and the microwave radar signal is sent to the target human body through the radar. This application adopts a one-transmit-one-receive frequency-modulated continuous wave radar with only one receiving antenna and one transmitting antenna. By sampling the intermediate frequency signal of the radar microwave reflection signal, the accuracy of the microwave signal in the process of converting from analog to digital is ensured, and the uncertainty and noise interference in the analog signal processing are effectively eliminated. The one-dimensional fast Fourier transform (1DFFT) is applied to realize the rapid extraction of the frequency domain features of the microwave radar digital signal, and the useful information is separated from the microwave radar digital signal. The use of frequency domain features can significantly improve the ability to detect and identify targets, especially for micro-motion monitoring or fine displacement monitoring. The frequency domain features can provide intuitive frequency and amplitude information, thereby enhancing the monitoring sensitivity and accuracy.

[0053] S2 determines whether the current human body is in a resting state according to the radar microwave reflection signal, and if so, extracts the radar microwave reflection signal in the resting state;

[0054] S201: If the one-dimensional frequency domain feature of the radar microwave digital signal exceeds the first body motion threshold, there is someone in the current monitoring area; otherwise, there is no one in the current monitoring area;

[0055] S202 determines whether the current human body is in a resting state based on the radar microwave digital signal indicating that there is a person in the current monitoring area.

[0056] In this step, in the actual environment, there will be people walking around the detected human body, especially in the hospital or nursing home scene. The movement of people will interfere with the microwave radar monitoring, resulting in large errors in the subsequent calculation of breathing frequency and heart rate. Therefore, by setting the first body movement threshold, the frequency domain range of non-human targets (for example, no one in the bed) or large non-periodic motion signals (for example, people walking) can be excluded. The frequency domain range after the first body movement threshold is further analyzed to determine whether the human body is in a resting state.

[0057] Further, judging whether the current human body is in a resting state according to the radar microwave digital signal indicating that there is a person in the current monitoring area includes:

[0058] S2021 extracts a frequency sweep signal at a specific time interval from the distance feature of the radar microwave digital signal;

[0059] S2022 performs 2D FFT calculation on the frequency sweep signal to obtain a two-dimensional frequency domain feature of the radar microwave digital signal;

[0060] S2023 If the two-dimensional frequency domain characteristics of the radar microwave digital signal exceed the second body motion threshold, the human body is in a body motion state, otherwise the human body is in a resting state.

[0061] In this step, in order to better analyze the distance and speed information of human targets in radar microwave digital signal processing, the swept frequency signal (Chirp signal) refers to a signal whose frequency changes linearly with time. Radar systems usually use swept frequency signals because they can sweep over a wider frequency band, thereby improving the distance resolution. The two-dimensional fast Fourier transform (2D FFT) further converts the time information into speed information. By performing 2D FFT on the swept frequency signal at a specific time interval, the distance and speed of the target can be analyzed simultaneously, and noise and interference can be effectively suppressed to improve the detection sensitivity of micro-moving targets.

[0062] In addition, when a person is lying in bed, the most accurate monitoring is only when the person is in a resting state and has a breathing rate and heart rate. However, existing microwave radar monitoring is difficult to accurately distinguish between a person's resting or moving state. On this basis, a second body movement threshold is set to exclude frequency domain signals of gross body movements (for example, turning over, etc.). The remaining micro-motion periodic frequency domain signals are the frequency domain signals in a resting state. It is most accurate to determine that the person is in a resting state at this time and to monitor based on the radar microwave reflection signal obtained in a resting state.

[0063] Furthermore, the first body movement threshold and the second body movement threshold are dynamically adjusted respectively by a constant false alarm rate algorithm.

[0064] In this step, the Constant False Alarm Rate (CFAR) algorithm is a method for dynamically adjusting the detection threshold in radar and signal detection. The CFAR algorithm can automatically adjust the threshold according to the noise level in the real-time signal environment. By dynamically adjusting the threshold, the CFAR algorithm can maintain a constant false alarm rate when the noise level changes, thereby reducing false alarms (False Alarm) and missed detections (Missed Detection). Under weaker target signals and higher background noise, CFAR helps to improve the sensitivity of target detection and ensure that micro-motion signals (such as breathing and heartbeat) can be reliably detected. The constant false alarm rate algorithm processing process is a prior art and is not described in detail in this application.

[0065] S3 separates and reconstructs the respiratory signal and the heart rate signal of the radar microwave reflection signal in a resting state to obtain the reconstructed respiratory signal and the heart rate signal;

[0066] S301 performs multi-layer variational modal decomposition on the radar microwave digital signal in a static state to obtain each decomposition component at a specified number of layers;

[0067] S302 uses known human physiological characteristics to determine the frequency range of the vital signs, and selects the respiratory decomposition component and the heart rate decomposition component from each decomposition component according to the frequency range of the vital signs;

[0068] S303: Superimpose and reconstruct the filtered respiratory decomposition components and heartbeat decomposition components to obtain separated respiratory signals and heart rate signals.

[0069] In this step, in actual situations, the respiration and heart rate signals are often mixed together, and are accompanied by environmental noise and other physiological activities. VMD technology decomposes the signal into multiple modal components, so that the characteristic signals of different frequencies can be separated independently, and then the signals corresponding to the target frequency (respiration and heart rate) are extracted. Using the variational mode decomposition (VMD) technology, the signal is decomposed into multiple modal components (IMF) to extract signals with different frequency characteristics, each component corresponding to the frequency component of a specific frequency band; the known human respiration and heart rate ranges are used to guide the extraction of the frequency components of these signals. Respiration is usually in a lower frequency band (adults: 0.2Hz-0.33Hz), while heart rate is in a relatively high frequency band (adults: 1Hz-1.67Hz); from the components obtained from VMD, the parts belonging to the respiration and heart rate frequency bands are extracted, and the selected modal components that meet the conditions of the respiration and heart rate frequency bands are superimposed and merged to form separated respiration and heart rate signals, so that the respiration and heart rate signals can be more accurately separated and analyzed independently, thereby improving the accuracy of physiological state monitoring.

[0070] S4 extracts the phase data of the reconstructed respiratory signal and heart rate signal respectively, performs wavelet transform on the phase data of the respiratory signal and the heart rate signal respectively, obtains respiratory phase features and heart rate phase features, and calculates the respiratory frequency and heart rate according to the respiratory phase features and the heart rate phase features.

[0071] Furthermore, before the phase data of the respiratory signal and the heart rate signal are subjected to wavelet transformation respectively, the following steps are performed:

[0072] The reconstructed respiratory signal and heart rate signal are input into a pre-trained wavelet basis recognition model to obtain the corresponding wavelet basis type and decomposition layer number.

[0073] Specifically, the wavelet basis recognition model is trained through the following steps:

[0074] The signal energy, frequency distribution, and instantaneous phase of several respiratory signals and heart rate signals are collected to form samples, and the corresponding wavelet basis types and decomposition layers of the samples are calibrated to form a wavelet basis recognition sample training set;

[0075] The wavelet basis recognition model is constructed by a neural network model, and the wavelet basis recognition model is trained by a wavelet basis recognition sample training set to obtain the parameters of the wavelet basis recognition model.

[0076] In this step, the pre-trained wavelet basis recognition model can quickly and automatically select the optimal wavelet basis and decomposition level, avoiding the tediousness and subjectivity of the manual selection process; through the optimized wavelet basis selection, the subtle changes and main features in the signal can be effectively separated and extracted, improving the accuracy of heart rate and respiratory rate calculation.

[0077] The overall energy distribution of each signal segment reflects the amplitude characteristics of the signal, which helps to understand the strength and variation of the signal; the frequency domain characteristics of the signal, including the energy distribution of the main frequency components and the spectrum, are often obtained through Fourier transform to help identify the main oscillation frequency of the signal; the phase information of the signal is usually obtained by Hilbert transform or other phase extraction methods, which reveals the instantaneous change trend and periodic characteristics of the signal. Through the above characteristic information, a complete sample input is formed, so that the model can capture the performance of the signal in different dimensions. According to its characteristics, each sample is manually calibrated with a most suitable wavelet basis type based on actual experience. Due to the different frequencies of respiratory frequency and heart rate, the wavelet basis type selection is also different. The wavelet basis types suitable for respiratory frequency signals include: Haar wavelet, B-Spline wavelet, Biorthogonal wavelet, etc.; the wavelet basis types suitable for heart rate signals include: Daubechies wavelet, Coiflets wavelet, Symlets wavelet, etc. For each sample, the appropriate number of wavelet decomposition layers needs to be specified. The selection of the number of decomposition layers is usually related to the signal detail level, frequency range and resolution requirements, ensuring that the decomposition can properly distinguish high-frequency and low-frequency components.

[0078] Further, the phase data of the respiratory signal and the heart rate signal are respectively subjected to wavelet transformation to obtain the respiratory phase feature and the heart rate phase feature, including:

[0079] According to the corresponding wavelet basis type and decomposition level, the heart rate phase data and the respiratory phase data are respectively decomposed by wavelet to obtain the heart rate spectrum and the respiratory spectrum;

[0080] The local maxima of the heart rate spectrum and the respiratory spectrum are identified respectively, and the respiratory phase features and the heart rate phase features are extracted respectively.

[0081] Further, calculating the respiratory frequency and the heart rate according to the respiratory phase characteristics and the heart rate phase characteristics includes:

[0082] Performing inverse calculation on the respiratory phase characteristics and the heart rate phase characteristics to obtain reconstructed respiratory frequency and heart rate;

[0083] The reconstructed respiratory frequency and heart rate are averaged and smoothed to obtain stable respiratory frequency and heart rate.

[0084] In this step, using the corresponding wavelet basis and decomposition layer number helps to analyze the signal phase data at different scales, so as to accurately separate the frequency components of heart rate and respiration. Finding the local maximum in the spectrum helps to accurately identify the characteristic frequencies of heart rate and respiration rate, and enhances the accuracy of signal analysis. By inverting the extracted phase features, the reconstructed respiration frequency and heart rate are restored, and redundant information is provided, which helps to verify the reliability of the results and improve the resistance to noise and anomalies. By smoothing the reconstructed frequency, such as the moving average method, the spikes or anomalies caused by instantaneous irregular changes can be eliminated, providing a smoother result output.

[0085] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0086] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0087] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0088] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0089] It should be noted that in the claims, any reference signs placed between brackets shall not be construed as limiting the claims. The word "comprising" does not exclude the presence of components or steps not listed in the claim. The word "a" or "an" preceding a component does not exclude the presence of a plurality of such components. The invention may be implemented by means of hardware comprising several different components and by means of a suitably programmed computer. In a unit claim enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third etc. does not indicate any order. These words may be interpreted as names.

[0090] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0091] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

[0092] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0093] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms should not be understood as necessarily referring to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, unless they are contradictory.

Claims

1. A vital sign monitoring method based on multi-layer variational mode decomposition reconstruction, characterized in that: include: Send a radar microwave signal to the target human body, receive a returned radar microwave reflection signal, and extract the frequency domain characteristics of the radar microwave reflection signal; Judging whether the human body is in a resting state according to the radar microwave reflection signal, if so, extracting the radar microwave reflection signal in the resting state; Separating and reconstructing the respiratory signal and the heart rate signal of the radar microwave reflection signal in a resting state to obtain the reconstructed respiratory signal and the heart rate signal; The phase data of the reconstructed respiratory signal and heart rate signal are extracted respectively, and the phase data of the respiratory signal and the heart rate signal are subjected to wavelet transform respectively to obtain respiratory phase features and heart rate phase features, and the respiratory frequency and heart rate are calculated according to the respiratory phase features and the heart rate phase features.

2. The method for monitoring vital signs based on multi-layer variational mode decomposition and reconstruction according to claim 1 is characterized in that: The frequency domain features of the radar microwave reflection signal are extracted as follows: Performing intermediate frequency signal sampling on the radar microwave reflection signal to obtain a radar microwave digital signal; Perform 1D FFT calculation on the radar microwave digital signal to obtain the one-dimensional frequency domain characteristics of the radar microwave digital signal.

3. The method for monitoring vital signs based on multi-layer variational mode decomposition and reconstruction according to claim 2 is characterized in that: The step of judging whether the human body is currently in a resting state according to the radar microwave reflection signal comprises: If the one-dimensional frequency domain feature of the radar microwave digital signal exceeds the first body motion threshold, there is someone in the current monitoring area, otherwise there is no one in the current monitoring area; According to the radar microwave digital signal indicating that there is someone in the current monitoring area, it is judged whether the current human body is in a resting state.

4. The method for monitoring vital signs based on multi-layer variational mode decomposition and reconstruction according to claim 3 is characterized in that: The method of judging whether the current human body is in a resting state according to the radar microwave digital signal indicating that there is a person in the current monitoring area includes: Extracting a frequency sweep signal at a specific time interval from the distance characteristics of the radar microwave digital signal; Perform 2D FFT calculation on the frequency sweep signal to obtain the two-dimensional frequency domain characteristics of the radar microwave digital signal; If the two-dimensional frequency domain feature of the radar microwave digital signal exceeds the second body motion threshold, the human body is in a body motion state, otherwise the human body is in a resting state.

5. The method for monitoring vital signs based on multi-layer variational mode decomposition and reconstruction according to claim 4 is characterized in that: The first body movement threshold and the second body movement threshold are dynamically adjusted respectively by a constant false alarm rate algorithm.

6. The method for monitoring vital signs based on multi-layer variational mode decomposition and reconstruction according to claim 4, characterized in that: The step of separating and reconstructing the respiratory signal and the heart rate signal of the radar microwave reflection signal in a resting state to obtain the reconstructed respiratory signal and the heart rate signal comprises: Performing multi-layer variational mode decomposition on the radar microwave digital signal in a static state to obtain each decomposition component at a specified number of layers; Using known human physiological characteristics to determine the frequency range of vital signs, and filtering out a respiratory decomposition component and a heart rate decomposition component from each decomposition component according to the frequency range of the vital signs; The filtered respiratory decomposition components and heartbeat decomposition components are superimposed and reconstructed to obtain separated respiratory signals and heart rate signals.

7. The method for monitoring vital signs based on multi-layer variational mode decomposition and reconstruction according to claim 6, characterized in that: The phase data of the respiratory signal and the heart rate signal are respectively subjected to wavelet transformation before being executed: The reconstructed respiratory signal and heart rate signal are input into a pre-trained wavelet basis recognition model to obtain the corresponding wavelet basis type and decomposition layer number.

8. The method for monitoring vital signs based on multi-layer variational mode decomposition and reconstruction according to claim 7, characterized in that: The wavelet basis recognition model is trained by the following steps: The signal energy, frequency distribution, and instantaneous phase of several respiratory signals and heart rate signals are collected to form samples, and the corresponding wavelet basis types and decomposition layers of the samples are calibrated to form a wavelet basis recognition sample training set; The wavelet basis recognition model is constructed by a neural network model, and the wavelet basis recognition model is trained by a wavelet basis recognition sample training set to obtain the parameters of the wavelet basis recognition model.

9. The method for monitoring vital signs based on multi-layer variational mode decomposition and reconstruction according to claim 8, characterized in that: The phase data of the respiratory signal and the heart rate signal are respectively subjected to wavelet transformation to obtain the respiratory phase feature and the heart rate phase feature, including: According to the corresponding wavelet basis type and decomposition level, the heart rate phase data and the respiratory phase data are respectively decomposed by wavelet to obtain the heart rate spectrum and the respiratory spectrum; The local maxima of the heart rate spectrum and the respiratory spectrum are identified respectively, and the respiratory phase features and the heart rate phase features are extracted respectively.

10. The method for monitoring vital signs based on multi-layer variational mode decomposition and reconstruction according to claim 9, characterized in that: Calculating the respiratory frequency and heart rate according to the respiratory phase characteristics and the heart rate phase characteristics comprises: Performing inverse calculation on the respiratory phase characteristics and the heart rate phase characteristics to obtain reconstructed respiratory frequency and heart rate; The reconstructed respiratory frequency and heart rate are averaged and smoothed to obtain stable respiratory frequency and heart rate.

Images