Vital signs monitoring method and apparatus

Abstract

The specification relates to the technical field of radar vital sign monitoring, and specifically discloses a vital sign monitoring method and device, wherein the method comprises: obtaining echo data; performing 3D-FFT processing on the echo data to determine whether the echo data contains a target; in the case of determining that the target is contained, performing target detection to obtain a micro-Doppler signal corresponding to a vital sign; performing phase extraction on the micro-Doppler signal to obtain phase information; filtering the phase information by using a first band-pass filter and a second band-pass filter respectively to obtain first phase data and second phase data; performing frequency tracking on the first phase data and the second phase data respectively based on an M-Rife interpolation algorithm to obtain a first frequency value and a second frequency value; the first frequency value is used to represent a first vital sign; and the second frequency value is used to represent a second vital sign. The above scheme can improve the accuracy and efficiency of vital sign monitoring.

Description

Technical Field

[0001] This specification relates to the field of radar vital sign monitoring technology, and in particular to a vital sign monitoring method and device. Background Technology

[0002] Radar vital sign algorithms monitor respiratory rate and heart rate using radar to assess a person's health and bodily functions without contact, representing a significant direction in the development of contactless vital sign monitoring. Compared to other non-contact vital sign sensors such as infrared and video sensors, microwaves have stronger penetrating power through materials and clothing and are less affected by environmental conditions such as light, temperature, and humidity. Representative algorithms in radar vital sign monitoring include Fast Fourier Transform (FFT), Wavelet Transform (WT), and Empirical Mode Decomposition (EMD). Compared to other algorithms, Fast Fourier Transform offers better engineering practicality and faster computation speed.

[0003] However, the accuracy of current radar vital signs algorithms is low, and there is a need to improve their accuracy in instantaneous calculations.

[0004] There is currently no effective solution to the above problems. Summary of the Invention

[0005] This specification provides a method and apparatus for monitoring vital signs, in order to solve the problem of low accuracy in existing radar vital sign algorithms.

[0006] This specification provides an embodiment of a method for monitoring vital signs, including:

[0007] Acquire echo data; perform 3D-FFT processing on the echo data to determine whether the echo data contains the target;

[0008] If the echo data is determined to contain a target, target detection is performed to obtain the micro-Doppler signal corresponding to vital signs; phase extraction is then performed on the micro-Doppler signal to obtain phase information.

[0009] The phase information is filtered using a first bandpass filter and a second bandpass filter respectively to obtain first phase data and second phase data;

[0010] Based on the M-Rife interpolation algorithm, frequency tracking is performed on the first phase data and the second phase data respectively to obtain a first frequency value and a second frequency value; the first frequency value is used to characterize the first vital sign; the second frequency value is used to characterize the second vital sign.

[0011] In one embodiment, acquiring echo data includes:

[0012] Acquire echo signal;

[0013] The echo signal is analyzed using radar parameters to obtain echo data;

[0014] The echo data is processed by using an MTI filter to eliminate clutter, resulting in processed echo data.

[0015] In one embodiment, performing 3D-FFT processing on the echo data to determine whether the echo data contains a target includes:

[0016] The echo data is processed using 3D-FFT to generate a distance-angle heatmap;

[0017] The presence of a target is determined by the energy level of the highlighted portion in the distance-angle heatmap.

[0018] In one embodiment, target detection is performed to obtain micro-Doppler signals corresponding to vital signs, including:

[0019] The echo data is processed by range-FFT to generate a range bin-frame heatmap;

[0020] The distance cell with the largest amount of information in the distance cell-frame heatmap is used as the micro-Doppler signal generated by the vital signs signal.

[0021] In one embodiment, phase extraction is performed on the micro-Doppler signal to obtain phase information, including:

[0022] The phase information is obtained by processing the micro-Doppler signal using the arctangent algorithm;

[0023] The phase information is unwound to obtain the unwound phase information.

[0024] In one embodiment, frequency tracking is performed on the first phase data based on the M-Rife interpolation algorithm to obtain a first frequency value, including:

[0025] Frequency estimation is performed by interpolating the maximum amplitude spectrum line and the adjacent second-maximum spectrum line in the first phase data to obtain the first estimated frequency value;

[0026] If the first estimated frequency value is located in the central region of two adjacent quantization frequency points, the first estimated frequency value is frequency tracked to obtain the first frequency value.

[0027] If the first estimated frequency value is close to the quantization frequency point, the signal sequence is shifted and interpolated again, followed by frequency tracking to obtain the first frequency value.

[0028] In one embodiment, frequency tracking is performed on the first estimated frequency value to obtain a first frequency value, including:

[0029] Frequency tracking is performed on the first estimated frequency value to obtain at least one frequency change link;

[0030] Obtain echo data within the current unit of time up to the current moment; the echo data within the current unit of time includes multiple frames of echo data;

[0031] Determine the frequency matching point between two adjacent echo data frames in the multi-frame echo data, and calculate the link length value formed by the frequency matching point;

[0032] Calculate the lengths of multiple links within the current unit time; select the longest link among the multiple links whose initial frequency value is within the error range of the frequency value at the previous moment as the historical curve of the frequency change at the previous moment within the current unit time, and determine the tail frequency of the longest link as the first frequency value at the current moment.

[0033] In one embodiment, determining the frequency matching point between two adjacent echo data frames in the multi-frame echo data and calculating the link length value formed by the frequency matching point includes at least one of the following:

[0034] If, for a specified frequency value in the current frame, there is no frequency value in the next frame that matches the specified frequency value, the link corresponding to the specified frequency value is terminated.

[0035] For a specified frequency value in the current frame, if there is only one frequency value in the next frame that matches the specified frequency value, a connection is established between the specified frequency value and the matching frequency value in the next frame, and the link length of the connection is assigned.

[0036] For a given frequency value in the current frame, if there are at least two frequency values ​​in the next frame that match the given frequency value, at least two connections are formed based on the degree of matching between the at least two frequency values ​​and the given frequency value, and the link length corresponding to each of the at least two connections is assigned a value.

[0037] For a certain frequency value in the next frame, if there is no matching frequency value in the current frame, the frequency value is determined as a new frequency value and the new frequency value is tracked.

[0038] This specification also provides an embodiment of a vital signs monitoring device, including:

[0039] A determination module is used to acquire echo data; and to perform 3D-FFT processing on the echo data to determine whether the echo data contains a target.

[0040] The detection module is used to perform target detection when the echo data is determined to contain a target, and obtain the micro-Doppler signal corresponding to vital signs; and to extract the phase of the micro-Doppler signal to obtain phase information.

[0041] The filtering module is used to filter the phase information using a first bandpass filter and a second bandpass filter respectively to obtain first phase data and second phase data.

[0042] An interpolation module is used to perform frequency tracking on the first phase data and the second phase data based on the M-Rife interpolation algorithm to obtain a first frequency value and a second frequency value; the first frequency value is used to characterize a first vital sign; and the second frequency value is used to characterize a second vital sign.

[0043] This specification also provides a computer device, including a processor and a memory for storing processor-executable instructions, wherein the processor executes the instructions to implement the steps of the vital signs monitoring method described in any of the above embodiments.

[0044] This specification also provides a computer-readable storage medium storing computer instructions that, when executed, implement the steps of the vital signs monitoring method described in any of the above embodiments.

[0045] This specification provides a method for monitoring vital signs, which involves acquiring echo data, performing 3D-FFT processing on the echo data to determine whether the echo data contains a target, performing target detection when the echo data contains a target, obtaining micro-Doppler signals corresponding to vital signs, extracting phase information from the micro-Doppler signals, filtering the phase information using a first bandpass filter and a second bandpass filter to obtain first phase data and second phase data, and performing frequency tracking on the first phase data and second phase data based on the M-Rife interpolation algorithm to obtain a first frequency value and a second frequency value; the first frequency value is used to characterize a first vital sign; and the second frequency value is used to characterize a second vital sign. In the above scheme, considering the influence of human body micro-movements, clutter and harmonics in complex environments, when performing FFT frequency estimation on phase changes, the point with the maximum frequency amplitude may not necessarily be the history of phase changes caused by vibrations caused by breathing or heartbeat. Therefore, in this scheme, the M-Rife interpolation algorithm is used to sequentially perform M-Rife interpolation on the maximum preset number of frequency points, which can well include the actual respiratory frequency value or heartbeat frequency value, thereby improving the accuracy of vital sign detection. Attached Figure Description

[0046] The accompanying drawings, which are included to provide a further understanding of this specification and form part of it, do not constitute a limitation thereof. In the drawings:

[0047] Figure 1 A flowchart of a vital signs monitoring method in one embodiment of this specification is shown;

[0048] Figure 2 This specification shows a distance-frame heatmap after distance FFT in one embodiment;

[0049] Figure 3 A schematic diagram showing the relationship between the frequency points after M-Rife interpolation and the actual frequency points in one embodiment of this specification is shown;

[0050] Figure 4 A schematic diagram of link termination after frequency tracking is shown in one embodiment of this specification;

[0051] Figure 5 This specification shows a schematic diagram illustrating the existence of an optimal matching point and a second-best matching point in the link after frequency tracking in one embodiment.

[0052] Figure 6 This document shows a schematic diagram illustrating the assignment of link length values ​​after frequency tracking, where a unique matching point exists in the link, according to one embodiment of this specification.

[0053] Figure 7 This specification shows a schematic diagram illustrating the assignment of link length values ​​for the best and second-best matching points after frequency tracking in one embodiment.

[0054] Figure 8 This specification illustrates frequency tracking of newly emerging frequencies after frequency tracking in one embodiment;

[0055] Figure 9 This diagram illustrates multiple links obtained by frequency tracking of all frames within a unit time in one embodiment of this specification.

[0056] Figure 10 A signal processing flowchart of a vital signs monitoring method according to one embodiment of this specification is shown;

[0057] Figure 11 A detailed flowchart of a vital signs monitoring method according to one embodiment of this specification is shown;

[0058] Figure 12 A schematic diagram showing the relationship between the frequency calculated by the frequency tracking algorithm proposed in one embodiment of this specification and the actual frequency is shown.

[0059] Figure 13 A schematic diagram of a vital signs monitoring device according to one embodiment of this specification is shown;

[0060] Figure 14 A schematic diagram of a computer device according to one embodiment of this specification is shown. Detailed Implementation

[0061] The principles and spirit of this specification will now be described with reference to several exemplary embodiments. It should be understood that these embodiments are given merely to enable those skilled in the art to better understand and implement this specification, and are not intended to limit the scope of this specification in any way. Rather, these embodiments are provided to make this disclosure more thorough and complete, and to fully convey the scope of this disclosure to those skilled in the art.

[0062] Those skilled in the art will recognize that the embodiments described herein can be implemented as a system, apparatus, method, or computer program product. Therefore, the disclosure herein can be specifically implemented in the following forms: entirely hardware, entirely software (including firmware, resident software, microcode, etc.), or a combination of hardware and software.

[0063] Radar vital sign algorithms monitor respiratory rate and heart rate using radar to assess a person's health and bodily functions without contact, representing a significant direction in the development of contactless vital sign monitoring. Compared to other non-contact vital sign sensors such as infrared and video sensors, microwaves have stronger penetrating power through materials and clothing and are less affected by environmental conditions such as light, temperature, and humidity. Representative algorithms in radar vital sign monitoring include Fast Fourier Transform (FFT), Wavelet Transform (WT), and Empirical Mode Decomposition (EMD). Compared to other algorithms, FFT offers superior engineering practicality and computational speed. Therefore, the inventors of this application have designed an algorithm based on FFT, improving its accuracy in instantaneous calculations while retaining its computational speed.

[0064] The M-Rife algorithm is a frequency interpolation algorithm developed based on the Rife algorithm. The Rife algorithm performs interpolation calculations between the two largest amplitude spectral lines of a signal to obtain the instantaneous frequency of the signal. However, the estimation accuracy of this algorithm is closely related to the frequency distribution. Therefore, the M-Rife algorithm is used to perform a second Rife algorithm on the frequency estimated by the Rife algorithm to correct the influence of frequency distribution on the estimation accuracy.

[0065] The frequency tracking algorithm is a frequency estimation method performed after M-Rife algorithm interpolation, used to determine the history of respiratory rate and heart rate changes. Since the choice of initial frequency directly affects subsequent frequency estimation, this manual uses the first 20 seconds of data for initial frequency estimation and tracks the frequency change history in real time every second thereafter.

[0066] Based on this, the embodiments of this specification provide a method for monitoring vital signs. Figure 1 A flowchart of a vital signs monitoring method according to one embodiment of this specification is shown. While this specification provides method operation steps or apparatus structures as illustrated in the following embodiments or figures, more or fewer operation steps or module units may be included in the method or apparatus based on conventional or non-inventive effort. In steps or structures where there is no logically necessary causal relationship, the execution order of these steps or the module structure of the apparatus is not limited to the execution order or module structure described in the embodiments and figures of this specification. When the method or module structure is applied in a practical device or end product, it can be executed sequentially or in parallel (e.g., in a parallel processor or multi-threaded processing environment, or even a distributed processing environment) according to the method or module structure shown in the embodiments or figures.

[0067] like Figure 1 As shown, a vital signs monitoring method provided in one embodiment of this specification may include the following steps:

[0068] Step S101: Acquire echo data; perform 3D-FFT processing on the echo data to determine whether the echo data contains the target.

[0069] The methods described in this specification can be applied to computer devices. In some embodiments of this specification, acquiring echo data may include: acquiring an echo signal; analyzing the echo signal using radar parameters to obtain echo data; and performing clutter cancellation processing on the echo data using an MTI filter to obtain processed echo data.

[0070] The radar can be fixed in place, with its antenna directly facing the chest of the target object (which could be a human or other animal), ensuring that there are no other vibration signals in the surrounding environment. The designed radar parameters are then burned into the radar development board. Next, the radar development board is fixed in place and connected to a data acquisition board to obtain the data after ADC processing, facilitating subsequent testing of our method in Matlab.

[0071] The steps for connecting the radar development board and the data acquisition board to collect data are as follows: 1) Check the status of the jumper caps on the DCK board to ensure they are in the correct position; 2) Connect the Calterah RDP board to the DCK board via the J1 and J2 connectors on each board; 3) Turn on the power switch on the DCK board; 4) Connect the Micro USB port on the Calterah RDP board to the computer via a USB cable; 5) Connect the Ethernet port on the DCK board to the computer via a network cable; 6) Select the following IP address: Set the IP address to 192.168.1.100, the subnet mask to 255.255.255.0, and the default gateway to 192.168.1.1; 7) Use the Calterah's built-in GUI application to collect raw ADC / FFT data.

[0072] The collected data can be stored in the form of .dat files. The contents of the files can be read by Matlab and stored as data blocks in the form of number of sampling points × number of frames × number of antennas for easy processing later.

[0073] Radar echoes contain many fixed clutter components, including static target echoes and noise components. Generally, the target's echo power is greater than the interference clutter components. Although Constant False Alarm Rate (CFAR) can use a threshold to determine target presence, in practice, CFAR cannot completely identify the target. In this embodiment, a Moving Target Indication (MTI) filter can be used for determination. Processing with an MTI filter can remove clutter signals.

[0074] The MTI filter primarily employs pulse cancellation. For each pulse in the radar echo signal, the clutter component is the same; only moving targets experience phase changes due to their distance variations. When two consecutive pulses cancel each other out, the stationary target component is completely eliminated, leaving only the moving target component. In the embodiments described in this specification, it is necessary to detect the micro-Doppler phase information of vital signs. Therefore, MTI is proposed to identify moving target information. Then, the moving target information is subtracted from the original signal, which can reduce the interference of body micro-movements on the micro-Doppler signal analysis of vital signs to a certain extent.

[0075] In some embodiments of this specification, performing 3D-FFT processing on the echo data to determine whether the echo data contains a target may include: performing 3D-FFT processing on the echo data to generate a distance-angle heatmap; and determining whether a target exists based on the energy magnitude of the highlighted portion in the distance-angle heatmap.

[0076] The processed data is then subjected to 3D-FFT processing to generate a distance-angle heatmap. The presence of a target inside the vehicle is determined based on the energy level of the highlighted areas in the heatmap. If a target is present, further steps are performed. If no target is present, the absence of a target is directly displayed. For example, if the energy of the highlighted area is greater than a preset energy value, the presence of a target is confirmed.

[0077] Step S102: If the echo data contains a target, target detection is performed to obtain the micro-Doppler signal corresponding to vital signs; phase extraction is performed on the micro-Doppler signal to obtain phase information.

[0078] In some embodiments of this specification, target detection to obtain micro-Doppler signals corresponding to vital signs may include: performing range-FFT processing on the echo data to generate a range bin-frame heatmap; and using the range bin with the largest amount of information in the range bin-frame heatmap as the micro-Doppler signal generated from the vital sign signal.

[0079] To obtain phase information, FFT processing is required for the received data of each Tx-Rx channel pair. This involves performing FFT column-wise on the acquired data blocks, known as range-FFT. The purpose of range-FFT is to distinguish targets in single bins at different distances. Performing range-FFT on each frame of signal over continuous time yields a range bin-frame heatmap. This heatmap allows observation of multiple targets separated along the range dimension. However, multiple targets within the same distance cannot be distinguished, requiring signal processing in another dimension. In the case of a single target, the range bin with the highest information content is selected as the source for Doppler information generated from vital sign signals. Figure 2 As shown, the highlighted areas represent the range cells containing Doppler information. Data from the range cells across consecutive time intervals was extracted, and the arctangent algorithm was used to extract the historical phase change process.

[0080] In some embodiments of this specification, phase extraction of the micro-Doppler signal to obtain phase information may include: processing the micro-Doppler signal using an arctangent algorithm to obtain phase information; and unwinding the phase information to obtain unwound phase information.

[0081] After radar detects a target, the peak value of the target range bin is represented by a complex signal, with its real and imaginary parts constituting the true phase information of the signal. Mathematically, the phase value of the echo data can be obtained using the arctangent function. However, in computer processing, there is a phase dewinding problem. The phase interval obtained using the arctangent function is [-π, π] with a period of 2π. There are 2π jumps at the two boundaries, causing the phase change curve to break, which is the phase dewinding problem. In radar signal processing, the phase value of each pulse after range-FFT transformation is obtained using the arctangent function. To solve the phase dewinding problem, phase dewinding is performed to correct the processed phase data. For example, in Matlab, the unwrap(w) function can be called to dewind and obtain the actual phase value.

[0082] Step S103: The phase information is filtered using a first bandpass filter and a second bandpass filter respectively to obtain first phase data and second phase data.

[0083] Different vital signs can be separated using bandpass filters. For example, vital signs can include primary and secondary vital signs. The primary vital sign is the respiratory signal, and the secondary vital sign is the heartbeat signal. To accurately separate and extract the respiratory and heartbeat signals while filtering out high-frequency noise, two bandpass filters need to be constructed. The respiratory frequency is typically 0.1-0.8 Hz, and the heartbeat frequency is typically 1-2 Hz; these frequency bands are the passband cutoff frequency and stopband cutoff frequency of the two filters, respectively. A second-order IIR filter meeting these conditions can be designed using Matlab's toolbox.

[0084] Step S104: Based on the M-Rife interpolation algorithm, frequency tracking is performed on the first phase data and the second phase data respectively to obtain a first frequency value and a second frequency value; the first frequency value is used to characterize the first vital sign; the second frequency value is used to characterize the second vital sign.

[0085] M-Rife interpolation is performed on each frame of data. M-Rife is an improvement on the Rife algorithm. The Rife algorithm estimates frequency by interpolating the maximum spectral line of the amplitude spectrum and the adjacent second-largest spectral line.

[0086] In some embodiments of this specification, frequency tracking is performed on the first phase data based on the M-Rife interpolation algorithm to obtain a first frequency value, including: frequency estimation by interpolating the maximum amplitude spectrum line and the adjacent second maximum spectrum line in the first phase data to obtain a first estimated frequency value; when the first estimated frequency value is in the central region of two adjacent quantization frequency points, frequency tracking is performed on the first estimated frequency value to obtain a first frequency value; when the first estimated frequency value is close to the quantization frequency point, the signal sequence is shifted, interpolation is performed again, and then frequency tracking is performed to obtain the first frequency value.

[0087] The M-Rife interpolation principle is as follows. Assuming the signal, after FFT, has a maximum spectral line at k0, the frequency estimate obtained through the Rife algorithm is:

[0088]

[0089] Among them, f s Let be the sampling frequency, N be the number of sampling points, |X(k0)| be the maximum spectral value, and |X(k0+r)| be the second largest spectral value. When |X(k0+1)| < |X(k0-1)|, r = -1. When |X(k0+1)| ≥ |X(k0-1)|, r = 1.

[0090] When the signal frequency is located at the center of two adjacent quantization frequency points, the Rife algorithm has high accuracy, with a mean square error close to CRLB. However, its accuracy decreases when the signal is near the maximum or second-largest frequency point. The M-Rife algorithm works as follows: First, perform one Rife algorithm iteration. If the estimated frequency value is within the center region of two adjacent quantization frequency points, the algorithm stops. If the estimated frequency value is close to the quantization frequency point, the signal sequence is shifted by Δf / 3 (Δf = f). s / N), f s Where is the sampling frequency, and N is the number of FFT points. Then, the Rife algorithm is re-executed to obtain the frequency estimate. For example... Figure 3 As shown, the relationship between the interpolated frequency points and the actual frequency values ​​is illustrated.

[0091] In the above embodiments, considering the influence of human body micro-movements, clutter and harmonics in complex environments, when performing FFT frequency estimation on phase changes, the point with the maximum frequency amplitude may not necessarily be the history of phase changes caused by vibrations caused by breathing or heartbeat. Therefore, in this scheme, the M-Rife interpolation algorithm is used to sequentially perform M-Rife interpolation on the maximum preset number of frequency points, which can well include the actual respiratory frequency value or heartbeat frequency value, thereby improving the accuracy of vital sign detection.

[0092] In some embodiments of this specification, frequency tracking of the first estimated frequency value to obtain a first frequency value may include: frequency tracking of the first estimated frequency value to obtain at least one frequency change link; acquiring echo data within the current unit time up to the current moment; the echo data within the current unit time includes multiple frames of echo data; determining the frequency matching point between two adjacent frames of echo data in the multiple frames of echo data, and calculating the link length value formed by the frequency matching point; calculating multiple link lengths within the current unit time; selecting the longest link among the multiple links whose initial link frequency value is within the error range of the frequency value at the previous moment as the historical curve of frequency change at the previous moment within the current unit time, and determining the tail frequency of the longest link as the first frequency value at the current moment.

[0093] In this embodiment, the frequency tracking method can use the respiratory and heart rate from the previous moment to calculate the current frequency. Frequency matching points between adjacent frames can be found within the current unit of time (e.g., one second), and the ratio of the link at each frequency matching point can be calculated. The longest link where the initial frequency in the current unit of time has an acceptable error with the frequency from the previous moment is considered a continuation of the frequency from the previous second in the current second. Therefore, the frequency at the end of the link is considered the frequency of the detected signal at the current moment. By tracking the frequency after interpolation using the M-Rife algorithm, the history of changes in respiratory rate and heart rate can be determined, further improving the accuracy and real-time performance of vital sign monitoring.

[0094] In some embodiments of this specification, determining the frequency matching point between two adjacent echo data frames in the multi-frame echo data and calculating the link length value formed by the frequency matching point may include at least one of the following: for a specified frequency value in the current frame, if there is no frequency value matching the specified frequency value in the next frame, the link corresponding to the specified frequency value is terminated; for a specified frequency value in the current frame, if there is only one frequency value matching the specified frequency value in the next frame, a connection is established between the specified frequency value and the matching frequency value in the next frame, and the link length of the connection is assigned; for a specified frequency value in the current frame, if there are at least two frequency values ​​matching the specified frequency value in the next frame, at least two connections are formed according to the degree of matching between the at least two frequency values ​​and the specified frequency value, and the link length corresponding to each of the at least two connections is assigned; for a certain frequency value in the next frame, if there is no frequency value matching it in the current frame, the frequency value is determined as a new frequency value, and the new frequency value is tracked.

[0095] In this embodiment, a novel frequency tracking algorithm can be used to track respiratory rate and heart rate. The frequency tracking algorithm may include the following four steps:

[0096] Step 1: Assume there are 19 frequency points in frame k, namely... Assume that the first n-1 points have already found corresponding frequency matching points in the next frame k+1, and each frequency point forms a link with its corresponding frequency matching point. For the frequency points in frame k... If all points at frame k+1 satisfy equation (2), then it indicates that... There is no corresponding frequency matching point in the next frame. In this case, the frequency point can be determined. It "dies" at frame k+1, at which point... For the link tail termination, such as Figure 4 As shown.

[0097]

[0098] In the above formula, m represents the frequency point at frame k+1, and Δ is the matching interval.

[0099] If a frequency point exists If equation (2) is not satisfied, such as Figure 5 As shown, further judgment is then made, and the judgment conditions are as follows:

[0100]

[0101] Where i≠m represents any frequency point in frame k+1 that is not equal to m. In this case, only the frequency point excluding m is considered. External and frequency points The second-best matching point with the smallest error is considered as the frequency of the best and second-best matching points. Candidate matching points. δ1 is and The distance between them. δ2 is and The distance between them.

[0102] Step 2: Further evaluate the candidate matching points from the previous step. Assume the candidate matching points at frame k+1... If there is no suitable frequency point at frame k, then... for At the unique matching point in the next frame, the connection length between these two points is assigned a value of 1, such as... Figure 6 As shown. At this point, the following conditions are met:

[0103]

[0104] Suppose the candidate matching point at frame k+1 It can be a candidate of multiple frequency points at frame k, satisfying formula (3), at this time The frequency point at frame k and Multiple matching points. In this case, a link is created only between the two matching points with the smallest error between frame k and its corresponding frame k+1. The link length is then assigned based on the ratio of the frequency difference between frame k and the frequency of the matching point, such as... Figure 7 As shown. Figure 7 In the middle, L1 k,k+1 for and The link length value between them. L2 k,k+1 for and The link length between them. Note that the frequency ratio needs to be replaced by the frequency difference between the two links, because the closer the frequencies are, the more likely they are to be a continuation of the same signal.

[0105] Step 3: The proposed frequency tracking algorithm includes a concept called "birth" to detect new frequencies appearing in consecutive frames of vital sign data. Specifically, if the frequency value in the (k+1)th frame... If no frequency is found to match any frequency in the previous frame k, it is classified as the "birth" of a new frequency in the (k+1)th frame. This situation may occur in... A new correlation is established between the frequency values ​​of the first and second frames (k+2), indicating the potential presence of a new frequency component in the vital signs signal. It is noteworthy that this newly identified frequency does not necessarily correspond to a true respiratory or heartbeat signal, but may be attributed to unwanted interference or noise, commonly referred to as clutter. These signals may have been masked in the previous frame due to low amplitude or other factors, manifesting as new frequency components in the current frame. Frequency tracking steps 1 and 2 are performed for each new "birth" frequency, as follows: Figure 8 As shown.

[0106] Step 4: Perform steps one through three sequentially for all frames within time (t, t+1) to calculate the length of all links. If the respiratory or heart rate calculated at time t is F... t Find all links that are related to F t The longest link with a phase difference within the maximum permissible error, the frequency F at the end of the link. t+1 That is, the respiratory or heart rate at time t+1, such as Figure 9 As shown.

[0107] In this embodiment, the proposed frequency tracking method can track the first and second frequencies corresponding to vital signs. During the tracking process, the frequency at the current moment can be determined based on the frequency at the previous moment, which can further improve the accuracy of frequency determination and enable real-time monitoring of vital signs.

[0108] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to interchangeably. Each embodiment focuses on describing the differences from other embodiments. For details, please refer to the foregoing descriptions of the relevant processing embodiments; they will not be repeated here.

[0109] The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0110] The above method will be described below with reference to a specific embodiment. However, it is worth noting that this specific embodiment is only for better illustration of this specification and does not constitute an improper limitation of this specification.

[0111] This specific embodiment provides a contactless vital sign monitoring algorithm based on the M-Rife algorithm for frequency tracking. This algorithm is applicable not only to respiratory rate and heart rate calculations per minute but also maintains high accuracy in calculations per second. The method performs M-Rife frequency interpolation on the frequency after the Fast Fourier Transform of each frame of data, and uses the proposed frequency tracking method on all interpolated frequency points to generate a historical frequency change curve per second. Therefore, the monitoring method in this embodiment exhibits strong performance in instantaneous frequency calculation. In short, the method in this embodiment includes the following steps.

[0112] 1) Fix the radar so that its antenna is pointed at the chest of the human body;

[0113] 2) Analyze the echo signal based on the radar parameters and organize the data into an easy-to-process data block format;

[0114] 3) Use differential algorithms to remove the influence of clutter signals;

[0115] 4) Based on 1D FFT, select the distance cell with the largest mean square error as the signal for chest displacement;

[0116] 5) Use a phase extraction algorithm on the selected distance cell to extract the phase change history;

[0117] 6) Design a second-order IIR bandpass filter to extract respiratory and heartbeat signals located in different frequency ranges;

[0118] 7) Perform M-Rife frequency interpolation after FFT of the corresponding respiratory and heartbeat signals;

[0119] 8) Use a frequency tracking algorithm to track frequency changes and obtain the history of changes in respiratory and heart rate.

[0120] Specifically, the embodiments in this specification are mainly based on the M-Rife algorithm for frequency tracking, providing a method for monitoring vital signs that ensures calculation accuracy while achieving high efficiency.

[0121] Please refer to Figure 10 The diagram shows a signal processing flowchart of the vital signs monitoring method in the embodiments of this specification. Figure 10 In the diagram, M represents the number of fast-time sampling points for the echo data, N represents the number of slow-time sampling points for the echo data, and Ant represents the number of radar receiving channels. After obtaining the echo data, a range FFT can be performed to obtain target range information. After obtaining the target range information, the range cell with the largest information is selected as the range cell containing the micro-Doppler signal. Phase extraction is then performed on the micro-Doppler signal to obtain the antenna's phase history change information. Two bandpass filters can then be constructed based on the breathing frequency and heart rate to separate the breathing signal and heart rate signal. After separation, FFT frequency estimation is performed for each frame of data. The M-Rife algorithm is used for interpolation to improve the accuracy of frequency estimation. Finally, frequency tracking is performed for each frequency value in each frame at the current time, and a history link of changes for each frequency value is established.

[0122] Please refer to Figure 11 The diagram shows a detailed flowchart of the vital signs monitoring method in this embodiment. Figure 11 As shown, this method specifically includes the following steps.

[0123] Step 1: Fix the radar so that its antenna is directly facing the chest of the human body, and ensure that there are no other vibration signals in the surrounding environment;

[0124] Step 2: Based on a review of existing literature and an understanding of the basic operating mode of FMCW millimeter-wave radar, the proposed radar front-end parameter design is shown in Table 1 below:

[0125] Table 1

[0126]

[0127] The designed radar parameters are burned into the radar development board. Then, the radar development board is secured and connected to the data acquisition board (DCK-ALPS-ETHERNET) to obtain the ADC-generated data, facilitating subsequent testing of our method in Matlab. The connection and data acquisition steps between the radar development board and the data acquisition board are as follows:

[0128] 1) Check the status of the jumper caps on the DCK board to ensure they are in the correct position;

[0129] 2) Connect the Calterah RDP board to the DCK board via the J1 and J2 connectors on each board;

[0130] 3) Turn on the power switch on DCK;

[0131] 4) Connect the Micro USB port on the Calterah RDP board to your computer using a USB cable;

[0132] 5) Connect the Ethernet port on the DCK board to the computer using a network cable;

[0133] 6) Select to use the following IP address. Set the IP address to 192.168.1.100, the subnet mask to 255.255.255.0, and the default gateway to 192.168.1.1;

[0134] 7) Use the Calterah's built-in GUI application to collect raw ADC / FFT data.

[0135] The collected data will be stored in the form of .dat files. The contents of the files will be read by Matlab and stored as data blocks in the form of Chirp sampling points × frame number × antenna number for easy processing later.

[0136] Step 3: Clutter Cancellation. Radar echoes contain many fixed clutter components, including static target echoes and noise components. Generally, the target echo power is greater than the interference clutter components. Although CFAR can use thresholds to determine target presence, in practice, CFAR cannot completely identify the target. In engineering, two methods are generally used: MTI and pulse Doppler processing. MTI filters out clutter signals, while pulse Doppler processing calculates the Doppler spectrum to obtain the result.

[0137] The MTI filter primarily employs pulse cancellation. For each pulse in the radar echo signal, the clutter component is the same; only moving targets experience phase changes due to their distance. When two consecutive pulses cancel each other out, the stationary target component is completely eliminated, leaving only the moving target component. In this study, only the micro-Doppler phase information of vital signs needs to be detected. Therefore, MTI is proposed to identify moving target information. The moving target information is then subtracted from the original signal, which can reduce the interference of body micro-movements on the micro-Doppler signal analysis of vital signs to a certain extent.

[0138] Step 4: Perform 3D-FFT processing on the above data to generate a distance-angle heatmap. Determine whether a target exists inside the vehicle based on the energy level of the highlighted areas in the heatmap. If a target exists, proceed to the next step. If not, simply display "No target found".

[0139] Step 5: Doppler Information Extraction. Phase extraction is performed on the Doppler signal obtained in Step 3. To obtain the target's phase information, FFT processing is required for each pair of Tx-Rx channel received data, i.e., FFT is performed column-wise on the acquired data blocks, called range-FFT. The purpose of range-FFT is to distinguish targets in single bins at different distances. Range-FFT is performed on each frame of signal in continuous time to obtain a range bin-frame heatmap. The generated range bin-frame heatmap allows observation of the separation of multiple targets in the range dimension. However, multiple targets within the same distance cannot be distinguished, requiring signal processing in another dimension. In the case of a single target, the range bin with the most information is selected as the Doppler information for the vital sign signal. Figure 2 As shown, the highlighted areas represent the range cells containing Doppler information. Data from the range cells across consecutive time intervals was extracted, and the arctangent algorithm was used to extract the historical phase change process.

[0140] Step 6: Phase Dewinding. After radar detects a target, the peak value of the target range bin is represented by a complex signal, with its real and imaginary parts constituting the true phase information of the signal. Mathematically, the phase value of the echo signal can be obtained using the arctangent function. However, in computer processing, there is a phase dewinding problem. The phase interval obtained using the arctangent function is [-π, π], with a period of 2π. There are 2π jumps at the two boundaries, causing the phase change curve to break, which is the phase dewinding problem. In radar signal processing, the phase value of each pulse after range-FFT transformation is obtained through the arctangent function. To solve the phase dewinding problem, phase dewinding is performed to correct the processed phase data. In Matlab, the unwrap(w) function can be called to perform dewinding and obtain the actual phase value.

[0141] Step 7: Separate respiratory and heartbeat signals using bandpass filters. To accurately separate and extract respiratory and heartbeat signals while filtering out high-frequency noise, two bandpass filters need to be constructed. Preliminary research indicates that the respiratory frequency is 0.1-0.8 Hz and the heartbeat frequency is 1-2 Hz. These frequency bands are the passband cutoff frequency and stopband cutoff frequency of the two filters, respectively. A second-order IIR filter meeting these conditions is designed using Matlab's toolbox.

[0142] Step 8: Perform M-Rife interpolation on each frame of data. M-RIFE is an improved version of the RIFE algorithm. The Rife algorithm estimates frequency by interpolating the maximum spectral line of the amplitude spectrum and the adjacent second-largest spectral line.

[0143] Step 9: Use the proposed frequency tracking algorithm to track respiratory rate and heart rate. Considering the influence of human micro-movements, clutter, and harmonics in a complex environment, when performing FFT frequency estimation on phase changes, the point of maximum frequency amplitude may not necessarily be the history of phase changes caused by vibrations from breathing or heartbeats. In reality, the 10 largest frequency points account for more than 99% of the energy in this time period. Therefore, performing M-Rife interpolation on the 10 largest frequency points sequentially can well include the actual respiratory rate or heart rate values.

[0144] The proposed frequency tracking method uses the breathing and heart rate from the previous moment to calculate the current rate. This method finds frequency matching points between adjacent frames within one second and calculates the link ratio for each matching point. The longest link whose initial frequency error from the previous moment is within an acceptable range is considered a continuation of the previous second's frequency in the current second. Therefore, the frequency at the end of the link is considered the frequency of the detected signal at the current moment.

[0145] Step 10: Continue executing steps 8 and 9 on the subsequently received data to complete the tracking of respiratory rate and heart rate over the remaining time, and display the real-time respiratory and heart rate, obtaining the relationship between respiratory frequency and true frequency within 100 seconds as follows: Figure 12 The diagram shows the relationship between the frequency calculated by the frequency tracking algorithm and the actual frequency.

[0146] Based on the same inventive concept, this specification also provides a vital signs monitoring device, as described in the following embodiments. Since the principle by which the vital signs monitoring device solves the problem is similar to that of the vital signs monitoring method, the implementation of the vital signs monitoring device can refer to the implementation of the vital signs monitoring method, and repeated details will not be elaborated further. As used below, the terms "unit" or "module" can refer to a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated. Figure 13 This is a structural block diagram of a vital signs monitoring device according to an embodiment of this specification, such as... Figure 13 As shown, it includes: a determination module 1301, a detection module 1302, a filtering module 1303, and an interpolation module 1304. The structure is described below.

[0147] The determination module 1301 is used to acquire echo data; and to perform 3D-FFT processing on the echo data to determine whether the echo data contains a target.

[0148] The detection module 1302 is used to perform target detection when it is determined that the echo data contains a target, and obtain the micro-Doppler signal corresponding to vital signs; and to extract the phase of the micro-Doppler signal to obtain phase information.

[0149] The filtering module 1303 is used to filter the phase information using a first bandpass filter and a second bandpass filter respectively to obtain first phase data and second phase data.

[0150] The interpolation module 1304 is used to perform frequency tracking on the first phase data and the second phase data based on the M-Rife interpolation algorithm to obtain a first frequency value and a second frequency value; the first frequency value is used to characterize a first vital sign; and the second frequency value is used to characterize a second vital sign.

[0151] In some embodiments of this specification, the determining module is specifically used for: acquiring echo signals; analyzing the echo signals using radar parameters to obtain echo data; and performing clutter cancellation processing on the echo data using an MTI filter to obtain processed echo data.

[0152] In some embodiments of this specification, the determining module may also be specifically used to: perform 3D-FFT processing on the echo data to generate a distance-angle heatmap; and determine whether a target exists based on the energy magnitude of the highlighted portion in the distance-angle heatmap.

[0153] In some embodiments of this specification, the detection module may be specifically used to: perform distance-FFT processing on the echo data to generate a distance bin-frame heatmap; and use the distance bin with the largest amount of information in the distance bin-frame heatmap as the micro-Doppler signal generated by the vital signs signal.

[0154] In some embodiments of this specification, the detection module may also be specifically used to: process the micro-Doppler signal using an arctangent algorithm to obtain phase information; and perform phase unwinding on the phase information to obtain unwound phase information.

[0155] In some embodiments of this specification, the interpolation module may be specifically used to: perform frequency estimation by interpolating the maximum amplitude spectrum line and the adjacent second maximum spectrum line in the first phase data to obtain a first estimated frequency value; when the first estimated frequency value is in the central region of two adjacent quantization frequency points, perform frequency tracking on the first estimated frequency value to obtain a first frequency value; when the first estimated frequency value is close to the quantization frequency point, perform signal sequence shifting and re-interpolate for frequency tracking to obtain a first frequency value.

[0156] In some embodiments of this specification, the device may further include a tracking module, which may be specifically used for: frequency tracking of the first estimated frequency value to obtain at least one frequency change link; acquiring echo data within the current unit time up to the current moment; the echo data within the current unit time includes multiple frames of echo data; determining the frequency matching point between two adjacent frames of echo data in the multiple frames of echo data, and calculating the link length value formed by the frequency matching point; calculating multiple link lengths within the current unit time; selecting the longest link among the multiple links whose initial link frequency value is within the error range of the frequency value at the previous moment as the historical curve of frequency change at the previous moment within the current unit time, and determining the tail frequency of the longest link as the first frequency value at the current moment.

[0157] In some embodiments of this specification, determining the frequency matching point between two adjacent echo data frames in the multi-frame echo data and calculating the link length value formed by the frequency matching point may include at least one of the following: for a specified frequency value in the current frame, if there is no frequency value matching the specified frequency value in the next frame, the link corresponding to the specified frequency value is terminated; for a specified frequency value in the current frame, if there is only one frequency value matching the specified frequency value in the next frame, a connection is established between the specified frequency value and the matching frequency value in the next frame, and the link length of the connection is assigned; for a specified frequency value in the current frame, if there are at least two frequency values ​​matching the specified frequency value in the next frame, at least two connections are formed according to the degree of matching between the at least two frequency values ​​and the specified frequency value, and the link length corresponding to each of the at least two connections is assigned; for a certain frequency value in the next frame, if there is no frequency value matching it in the current frame, the frequency value is determined as a new frequency value, and the new frequency value is tracked.

[0158] This specification also provides a computer device, which can be found in the following description. Figure 14 The diagram shown illustrates the structural composition of a computer device based on the vital signs monitoring method provided in the embodiments of this specification. Specifically, the computer device may include an input device 141, a processor 142, and a memory 143. The memory 143 stores processor-executable instructions. When the processor 142 executes the instructions, it implements the steps of the vital signs monitoring method described in any of the above embodiments.

[0159] In this embodiment, the input device can specifically be one of the main devices for information exchange between the user and the computer system. The input device may include a keyboard, mouse, light pen, handwriting input tablet, voice input device, etc.; the input device is used to input raw data and programs for processing these data into the computer. The input device can also receive data transmitted from other modules, units, and devices. The processor can be implemented in any suitable manner. For example, the processor can take the form of a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro)processor, logic gates, switches, application-specific integrated circuits (ASICs), programmable logic controllers, and embedded microcontrollers, etc. The memory can specifically be a memory device used to store information in modern information technology. The memory can include multiple layers; in digital systems, anything that can store binary data can be considered memory; in integrated circuits, a circuit without physical form but with storage function is also called memory, such as RAM, FIFO, etc.; in a system, a storage device with physical form is also called memory, such as a memory stick, TF card, etc.

[0160] In this embodiment, the specific functions and effects implemented by the computer device can be explained in comparison with other embodiments, and will not be repeated here.

[0161] This specification also provides a computer storage medium based on a vital signs monitoring method, wherein the computer storage medium stores computer program instructions that, when executed, implement the steps of the vital signs monitoring method described in any of the above embodiments.

[0162] In this embodiment, the storage medium includes, but is not limited to, Random Access Memory (RAM), Read-Only Memory (ROM), cache, hard disk drive (HDD), or memory card. The memory can be used to store computer program instructions. The network communication unit can be an interface configured according to standards specified in the communication protocol for network connection communication.

[0163] In this embodiment, the specific functions and effects implemented by the program instructions stored in the computer storage medium can be explained by comparison with other embodiments, and will not be repeated here.

[0164] Obviously, those skilled in the art will understand that the modules or steps of the embodiments described above can be implemented using general-purpose computing devices. They can be centralized on a single computing device or distributed across a network of multiple computing devices. Optionally, they can be implemented using computer-executable program code, thereby storing them in a storage device for execution by a computing device. In some cases, the steps shown or described can be performed in a different order than those presented herein, or they can be fabricated as separate integrated circuit modules, or multiple modules or steps can be fabricated as a single integrated circuit module. Thus, the embodiments of this specification are not limited to any particular combination of hardware and software.

[0165] It should be understood that the above description is for illustrative purposes and not for limitation. Many embodiments and applications beyond the provided examples will be apparent to those skilled in the art upon reading the above description. Therefore, the scope of this specification should not be determined by reference to the above description, but rather by reference to the foregoing claims and the full scope of their equivalents.

[0166] The above description is merely a preferred embodiment of this specification and is not intended to limit this specification. Various modifications and variations can be made to the embodiments described herein by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this specification should be included within the scope of protection of this specification.

Claims

1. A method for monitoring vital signs, characterized in that, include: Acquire echo data; The echo data is processed using 3D-FFT to determine whether the echo data contains a target; If the echo data is confirmed to contain the target, target detection is performed to obtain the micro-Doppler signal corresponding to vital signs; Phase information is obtained by performing phase extraction on the micro-Doppler signal; The phase information is filtered using a first bandpass filter and a second bandpass filter respectively to obtain first phase data and second phase data; Based on the M-Rife interpolation algorithm, frequency tracking is performed on the first phase data and the second phase data respectively to obtain a first frequency value and a second frequency value; the first frequency value is used to characterize the first vital sign; the second frequency value is used to characterize the second vital sign. Specifically, based on the M-Rife interpolation algorithm, frequency tracking is performed on the first phase data to obtain a first frequency value, including: frequency estimation is performed by interpolating the maximum amplitude spectrum line and the adjacent second maximum spectrum line in the first phase data to obtain a first estimated frequency value; when the first estimated frequency value is in the central region of two adjacent quantization frequency points, frequency tracking is performed on the first estimated frequency value to obtain the first frequency value. The process of frequency tracking to obtain a first frequency value includes: performing frequency tracking on the first estimated frequency value to obtain at least one frequency change link; acquiring echo data within the current unit time up to the current moment; the echo data within the current unit time includes multiple frames of echo data; determining the frequency matching point between two adjacent frames of echo data in the multiple frames of echo data, and calculating the link length value formed by the frequency matching point; calculating multiple link lengths within the current unit time; selecting the longest link among the multiple links whose initial link frequency value is within the error range of the frequency value at the previous moment as the historical curve of frequency change at the previous moment within the current unit time, and determining the tail frequency of the longest link as the first frequency value at the current moment.

2. The vital signs monitoring method according to claim 1, characterized in that, Acquire echo data, including: Acquire echo signal; The echo signal is analyzed using radar parameters to obtain echo data; The echo data is processed by using an MTI filter to eliminate clutter, resulting in processed echo data.

3. The vital signs monitoring method according to claim 1, characterized in that, The echo data is processed using 3D-FFT to determine whether the echo data contains a target, including: The echo data is processed using 3D-FFT to generate a distance-angle heatmap; The presence of a target is determined by the energy level of the highlighted portion in the distance-angle heatmap.

4. The vital signs monitoring method according to claim 1, characterized in that, Target detection is performed to obtain micro-Doppler signals corresponding to vital signs, including: The echo data is processed by range-FFT to generate a range bin-frame heatmap; The distance cell with the largest amount of information in the distance cell-frame heatmap is used as the micro-Doppler signal generated by the vital signs signal.

5. The vital signs monitoring method according to claim 1, characterized in that, Phase extraction is performed on the micro-Doppler signal to obtain phase information, including: The phase information is obtained by processing the micro-Doppler signal using the arctangent algorithm; The phase information is unwound to obtain the unwound phase information.

6. The vital signs monitoring method according to claim 1, characterized in that, Based on the M-Rife interpolation algorithm, frequency tracking is performed on the first phase data to obtain the first frequency value, which also includes: If the first estimated frequency value is close to the quantization frequency point, the signal sequence is shifted and interpolated again, followed by frequency tracking to obtain the first frequency value.

7. The vital signs monitoring method according to claim 1, characterized in that, Determine the frequency matching point between two adjacent echo data frames in the multi-frame echo data, and calculate the link length value formed by the frequency matching point, including at least one of the following: If, for a specified frequency value in the current frame, there is no frequency value in the next frame that matches the specified frequency value, the link corresponding to the specified frequency value is terminated. For a specified frequency value in the current frame, if there is only one frequency value in the next frame that matches the specified frequency value, a connection is established between the specified frequency value and the matching frequency value in the next frame, and the link length of the connection is assigned. For a given frequency value in the current frame, if there are at least two frequency values ​​in the next frame that match the given frequency value, at least two connections are formed based on the degree of matching between the at least two frequency values ​​and the given frequency value, and the link length corresponding to each of the at least two connections is assigned a value. For a certain frequency value in the next frame, if there is no matching frequency value in the current frame, the frequency value is determined as a new frequency value and the new frequency value is tracked.

8. A vital signs monitoring device, characterized in that, include: The determination module is used to acquire echo data; The echo data is processed using 3D-FFT to determine whether the echo data contains a target; The detection module is used to detect the target when the echo data is determined to contain the target, and to obtain the micro-Doppler signal corresponding to the vital signs. Phase information is obtained by performing phase extraction on the micro-Doppler signal; The filtering module is used to filter the phase information using a first bandpass filter and a second bandpass filter respectively to obtain first phase data and second phase data. An interpolation module is used to perform frequency tracking on the first phase data and the second phase data based on the M-Rife interpolation algorithm to obtain a first frequency value and a second frequency value; the first frequency value is used to characterize a first vital sign; and the second frequency value is used to characterize a second vital sign. Specifically, the interpolation module is used to: estimate the frequency by interpolating the maximum amplitude spectrum line and the adjacent second maximum spectrum line in the first phase data to obtain a first estimated frequency value; when the first estimated frequency value is in the central region of two adjacent quantization frequency points, perform frequency tracking on the first estimated frequency value to obtain a first frequency value; wherein, performing frequency tracking on the first estimated frequency value to obtain the first frequency value includes: performing frequency tracking on the first estimated frequency value to obtain at least one frequency change link; acquiring echo data within the current unit time up to the current moment; the echo data within the current unit time includes multiple frames of echo data; determining the frequency matching point between two adjacent frames of echo data in the multiple frames of echo data, and calculating the link length value formed by the frequency matching point; calculating multiple link lengths within the current unit time; selecting the longest link among the multiple links whose initial link frequency value is within the error range of the frequency value at the previous moment as the historical curve of the frequency change at the previous moment within the current unit time, and determining the tail frequency of the longest link as the first frequency value at the current moment.

9. A computer-readable storage medium storing computer instructions thereon, characterized in that, When the instructions are executed by the processor, they implement the steps of the method according to any one of claims 1 to 7.

Images