Heartbeat signal processing method and device

Abstract

The invention provides a heartbeat signal processing method and device. The method comprises the steps that echo I / Q signals of a target object are collected through a millimeter wave radar; static clutters and human body motion noise are removed from the echo I / Q signals, and target signals are obtained; filtering the target signal to obtain a heartbeat signal sequence; determining a peak point sequence of the heartbeat signal sequence, and calculating a mean square error corresponding to the peak point sequence; and when the mean square error is smaller than or equal to a set threshold value, outputting the heartbeat signal sequence. Through mean square error threshold judgment, a breathing harmonic interference scene is accurately recognized, the recognition accuracy is improved, and redundant processing when no interference exists is avoided; the phase second-order difference only suppresses breathing harmonics, heartbeat signals are not affected, and the heartbeat signal retention rate is increased; the method is high in compatibility, does not need extra hardware, can be directly integrated into an existing 77GHz millimeter wave radar heart rate detection system, and is long in adaptive detection distance and wide in detection angle; and the average accuracy and precision of heart rate estimation are obviously improved.

Description

Technical Field

[0001] This application relates to the field of millimeter-wave radar signal processing and vital sign detection technology, and in particular to a method and apparatus for processing heartbeat signals. Background Technology

[0002] In millimeter-wave radar non-contact heart rate detection, the harmonic signals generated by human respiratory signals (frequency 0.1Hz-0.5Hz) (such as the second harmonic 0.2Hz-1.0Hz and the third harmonic 0.3Hz-1.5Hz) are prone to overlap with the heartbeat signal (0.8Hz-2Hz), forming respiratory harmonic interference.

[0003] Existing technologies typically employ the following methods for processing: traditional low-pass filtering methods use a preset fixed cutoff frequency to filter out signals; adaptive filtering methods process signals by acquiring respiratory reference signals; and wavelet decomposition methods achieve multi-resolution analysis by progressively splitting the signal into low-frequency and high-frequency signal bands.

[0004] However, the aforementioned existing technologies have the following drawbacks: Traditional low-pass filtering methods require a fixed cutoff frequency to be preset. When the cutoff frequency is not within this range, problems may arise such as filtering out low-frequency heartbeat signals or failing to completely suppress high-order respiratory harmonics; adaptive filtering methods require additional acquisition of respiratory reference signals, and in multi-target scenarios, the reference signals are prone to crosstalk, leading to interference suppression failure; wavelet decomposition methods have low processing accuracy in the overlapping frequency band (1.0Hz-1.5Hz) of respiratory harmonics and heartbeat signals, which easily introduces signal distortion, resulting in a heart rate estimation deviation of more than 5%. Summary of the Invention

[0005] This application provides a method and apparatus for processing heartbeat signals to solve the problems in the prior art. Traditional low-pass filtering methods require a preset fixed cutoff frequency, and when the cutoff frequency is not within this range, low-frequency heartbeat signals may be filtered out or high-order respiratory harmonics may not be completely suppressed. Adaptive filtering methods require additional acquisition of respiratory reference signals, and in multi-target scenarios, the reference signals are prone to crosstalk, leading to interference suppression failure. Wavelet decomposition methods have low processing accuracy in the overlapping frequency band (1.0Hz-1.5Hz) of respiratory harmonics and heartbeat signals, which easily introduces signal distortion and causes heart rate estimation deviation to exceed 5%.

[0006] In a first aspect, this application provides a method for processing heartbeat signals, comprising: Acquire echo I / Q signals of the target object using millimeter-wave radar; The target signal is obtained by removing static clutter and human motion noise from the echo I / Q signal; The target signal is filtered to obtain a heartbeat signal sequence; Determine the peak point sequence of the heartbeat signal sequence, and calculate the mean square error corresponding to the peak point sequence; When the mean square error is less than or equal to a set threshold, the heartbeat signal sequence is output.

[0007] In one possible implementation, the method further includes: When the mean square error is greater than a set threshold, phase extraction is performed on the heartbeat signal sequence to obtain a phase sequence; The phase sequence is subjected to second-order difference processing to obtain a second-order phase difference sequence; The second-order phase difference sequence is smoothed by moving average to obtain the heartbeat signal sequence; Output the heartbeat signal sequence.

[0008] In one possible implementation, performing second-order difference processing on the phase sequence to obtain a second-order phase difference sequence includes: Perform a first-order difference operation on the phase sequence to obtain a first-order phase difference sequence; A second-order difference operation is performed on the first-order phase difference sequence to remove the breathing harmonics in the phase sequence, resulting in a second-order phase difference sequence.

[0009] In one possible implementation, the moving average smoothing process is performed using the following formula:

[0010] Where t represents the current time point, and k represents the index of the sampling point within the moving average window.

[0011] In one possible implementation, determining the peak sequence of the heartbeat signal sequence and calculating the mean square error corresponding to the peak sequence includes: Peak search is performed on the heartbeat signal sequence to extract multiple signal peak points, resulting in a signal peak point sequence; All signal peaks in the signal peak point sequence are normalized, and the mean square error of the normalized signal peak point sequence is calculated to obtain the mean square error of the normalized signal peak point sequence.

[0012] In one possible implementation, the acquisition of the echo I / Q signal of the target object via millimeter-wave radar includes: At different locations of the target object from the millimeter-wave radar, the echo I / Q signals of the target object under static conditions at multiple angles are obtained; The echo I / Q signal can be sustained for a first preset duration, and a sliding time window with a step size of a second pre-review time is set.

[0013] In one possible implementation, the method further includes: The MUSIC algorithm is used to calculate the heart rate of the heartbeat signal sequence to obtain the target heart rate of the target object.

[0014] Secondly, this application provides a heartbeat signal processing apparatus, comprising: The acquisition module is used to acquire the echo I / Q signals of the target object through millimeter-wave radar; The noise reduction module is used to remove static noise and human motion noise from the echo I / Q signal to obtain the target signal; A filtering module is used to filter the target signal to obtain a heartbeat signal sequence; The determination module is used to determine the peak point sequence of the heartbeat signal sequence and calculate the mean square error corresponding to the peak point sequence; The output module is used to output the heartbeat signal sequence when the mean square error is less than or equal to a set threshold.

[0015] In one possible implementation, the determining module is specifically used to perform peak search on the heartbeat signal sequence, extract multiple signal peak points to obtain a signal peak point sequence; normalize all signal peak points in the signal peak point sequence, and calculate the mean square error of the normalized signal peak point sequence to obtain the mean square error of the normalized signal peak point sequence.

[0016] In one possible implementation, the acquisition module is specifically used to acquire the echo I / Q signals of the target object at different locations at the distance from the millimeter-wave radar, using multiple angles under which the target object is stationary.

[0017] Compared with the prior art, the above-mentioned technical solution provided in this application embodiment has the following advantages: The method provided in this application embodiment acquires the echo I / Q signal of the target object through millimeter-wave radar; removes static clutter and human body motion noise from the echo I / Q signal to obtain the target signal; filters the target signal to obtain a heartbeat signal sequence; determines the peak point sequence of the heartbeat signal sequence and calculates the mean square error corresponding to the peak point sequence; and outputs the heartbeat signal sequence when the mean square error is less than or equal to a set threshold. By using a mean squared error threshold, it accurately identifies respiratory harmonic interference scenarios with an accuracy rate of 98.2%, avoiding redundant processing in the absence of interference. The second-order phase difference only suppresses respiratory harmonics (low-frequency slowly changing components) without affecting the heartbeat signal (high-frequency fluctuating components), achieving a lossless heartbeat signal retention rate of 99.5%. It has strong compatibility, requiring no additional hardware and can be directly integrated into existing 77GHz millimeter-wave radar heart rate detection systems, adapting to detection distances of 0.5m-1.5m and detection angles of 0°-60°. The estimation accuracy is significantly improved, with an average heart rate estimation accuracy increase of 4.3%-6.7%, and an accuracy rate of 97.8% in the 1.0Hz-1.5Hz signal overlap frequency band. Attached Figure Description

[0018] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0019] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] One or more embodiments are illustrated by way of example with reference numerals in the accompanying drawings. These illustrations do not constitute a limitation on the embodiments. Elements with the same reference numerals in the drawings are denoted as similar elements. Unless otherwise stated, the figures in the drawings are not to be limited by scale.

[0021] Figure 1 A flowchart illustrating an embodiment of a heartbeat signal processing method provided in this application; Figure 2 A flowchart illustrating an embodiment of another heartbeat signal processing method provided in this application; Figure 3 A flowchart illustrating an embodiment of a breathing harmonic interference suppression method provided in this application; Figure 4This is a block diagram illustrating an embodiment of a heartbeat signal processing device provided in this application. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0023] The following disclosure provides numerous different embodiments or examples for implementing various structures of this application. To simplify the disclosure, specific examples of components and arrangements are described below. These are merely examples and are not intended to limit the scope of this application. Furthermore, reference numerals and / or letters may be repeated in different examples. Such repetition is for simplification and clarity and does not in itself indicate a relationship between the various embodiments and / or arrangements discussed.

[0024] To address the problems in existing technologies, such as the need for a fixed cutoff frequency in traditional low-pass filtering methods (which may filter out low-frequency heartbeat signals or fail to completely suppress high-order respiratory harmonics when the cutoff frequency is outside this range), the requirement for additional respiratory reference signal acquisition in adaptive filtering methods (which are prone to crosstalk in multi-target scenarios, leading to interference suppression failure), and the low accuracy of wavelet decomposition in processing the overlapping frequency band (1.0Hz-1.5Hz) of respiratory harmonics and heartbeat signals (which easily introduces signal distortion and causes heart rate estimation deviations exceeding 5%), this application provides a heartbeat signal processing method that can accurately identify respiratory harmonic interference scenarios by judging through a mean square error threshold, improving the identification accuracy and avoiding redundant processing in the absence of interference. The second-order phase difference method only suppresses respiratory harmonics without affecting the heartbeat signal, thus improving the heartbeat signal retention rate. It boasts strong compatibility, requires no additional hardware, and can be directly integrated into existing 77GHz millimeter-wave radar heart rate detection systems, adapting to long detection distances and wide detection angles. The average accuracy and precision of heart rate estimation are significantly improved.

[0025] Figure 1 This is a flowchart illustrating an embodiment of a heartbeat signal processing method provided in this application. Figure 1 As shown, the method includes the following steps: S101. Acquire the echo I / Q signal of the target object through millimeter-wave radar.

[0026] This application is applied to millimeter-wave radar heart rate detection scenarios. By combining mean square error threshold judgment with second-order phase difference, it achieves respiratory harmonic interference suppression without additional hardware, can accurately identify interference scenarios, and does not lose heartbeat signals.

[0027] In this embodiment, a millimeter-wave radar is used to acquire the echo I / Q signal of a target object. This millimeter-wave radar may include components such as a transmitting antenna and a receiving antenna. The transmitting antenna transmits a millimeter-wave signal, which propagates in space and is reflected when it encounters a target object. The receiving antenna receives the signal reflected from the target object; this reflected signal contains information about the human target. The reflected signal is processed to obtain the echo I / Q signal of the target object, which contains amplitude and phase information of the signal.

[0028] In one example, the echo I / Q signal of the target object can be acquired using a 77 GHz millimeter-wave radar, thereby obtaining the echo I / Q signal of the person to be detected.

[0029] S102. Remove static noise and human motion noise from the echo I / Q signal to obtain the target signal.

[0030] In this embodiment, after obtaining the echo I / Q signal of the target object, the echo I / Q signal is processed to remove static clutter and human motion noise. Specifically, a distance-dimensional Fast Fourier Transform (FFT) can be performed on the acquired echo I / Q signal to convert the signal from the time domain to the frequency domain, thereby obtaining the signal amplitude distribution with frequency within each range cell to identify the target signal. Then, the CA-CFAR adaptive threshold filtering algorithm is used to distinguish the target signal from background noise. Finally, an improved moving average filtering method is used to further process the above-processed echo I / Q signal to remove static clutter and human motion noise, improve signal purity, and obtain the target signal.

[0031] S103. Filter the target signal to obtain the heartbeat signal sequence.

[0032] In this embodiment, the obtained target signal is filtered. Specifically, a Finite Impulse Response (FIR) filter can be used to process the denoised signal. The passband range of this FIR filter can be set to 0.8Hz to 2Hz (i.e., the heartbeat signal frequency range), allowing signals within this frequency range to pass smoothly while attenuating signals outside the passband. Simultaneously, the amplitude of the signal within the stopband is reduced, thereby effectively suppressing unwanted frequency components. Then, the filtered signal is sampled, and a heartbeat signal sequence is obtained by setting the sampling rate and time window length.

[0033] In one example, the sampling rate can be set to 25Hz, with an 18-second time window, to segment the signal and obtain a series of 18-second signal segments. Each signal segment contains the signal data collected within that time period. By analyzing these signal segments, the preprocessed heartbeat signal sequence S(t) can be obtained.

[0034] S104. Determine the peak point sequence of the heartbeat signal sequence and calculate the mean square error corresponding to the peak point sequence.

[0035] In this embodiment, after obtaining the heartbeat signal sequence, a peak search is performed on the heartbeat signal sequence to extract signal peak points, and a peak point sequence is obtained based on these peak points. The maximum amplitude value in the peak point sequence is obtained and normalized. The mean of the peak point sequence is calculated using the normalized peak point sequence, and then the root mean square error of the peak point sequence is obtained. The formula for calculating the root mean square error of the normalized peak point sequence can be:

[0036] Where M represents the mean squared error, and n represents the number of peak points. This represents the peak point.

[0037] In one example, a peak search is performed on the obtained heartbeat signal sequence. When extracting signal peak points, the number of peak points can be set to no less than 30, and the time interval between adjacent peak points can not exceed 1.25 seconds, thereby obtaining a peak point sequence.

[0038] S105. When the mean square error is less than or equal to the set threshold, output the heartbeat signal sequence.

[0039] In this embodiment of the application, after obtaining the mean squared error of the peak sequence, a preset mean squared error threshold is obtained. The mean squared error of the peak sequence is compared with the preset mean squared error threshold. When the mean squared error of the peak sequence is less than or equal to the preset mean squared error threshold, a heartbeat signal sequence is output. The output heartbeat signal sequence can be used for subsequent MUSIC algorithm heart rate estimation to obtain the target heart rate of the target object.

[0040] In an optional embodiment of this application, when the mean square error of the peak sequence is greater than a preset mean square error threshold, it indicates the presence of respiratory harmonic interference, i.e., the respiratory harmonics cause the peak fluctuation amplitude to exceed 30%. Respiratory harmonic interference suppression processing is performed, and the heartbeat signal sequence after interference processing is output. The output heartbeat signal sequence can be used for subsequent MUSIC algorithm heart rate estimation.

[0041] The technical solution provided in this application involves acquiring echo I / Q signals of a target object using millimeter-wave radar; removing static clutter and human motion noise from the echo I / Q signals to obtain the target signal; filtering the target signal to obtain a heartbeat signal sequence; determining the peak point sequence of the heartbeat signal sequence and calculating the mean square error corresponding to the peak point sequence; and outputting the heartbeat signal sequence when the mean square error is less than or equal to a set threshold. By using the mean square error threshold, respiratory harmonic interference scenarios can be accurately identified, improving the identification accuracy and avoiding redundant processing in the absence of interference. The second-order phase difference only suppresses respiratory harmonics without affecting the heartbeat signal, thus improving the heartbeat signal retention rate. It has strong compatibility, requires no additional hardware, and can be directly integrated into existing 77GHz millimeter-wave radar heart rate detection systems, adapting to long detection distances and wide detection angles. The average accuracy and precision of heart rate estimation are significantly improved.

[0042] Figure 2 A flowchart illustrating an embodiment of another heartbeat signal processing method provided in this application. Figure 2 Includes the following steps: S201. At different locations of the target object from the millimeter-wave radar, the echo I / Q signals of the target object at multiple angles under static conditions are obtained.

[0043] This application is applied to millimeter-wave radar heart rate detection scenarios. By combining mean square error threshold judgment with second-order phase difference, it achieves respiratory harmonic interference suppression without additional hardware, can accurately identify interference scenarios, and does not lose heartbeat signals.

[0044] In this embodiment, the echo I / Q signal of a target object is acquired using a millimeter-wave radar, which may include components such as a transmitting antenna and a receiving antenna. The transmitting antenna transmits a millimeter-wave signal, which is reflected when it encounters the target object being tested. The receiving antenna receives the signal reflected from the target object, and this reflected signal contains information about the target object. This reflected signal is processed to obtain the echo I / Q signal of the target object, which contains amplitude and phase information. Furthermore, the echo I / Q signal of the target object at different locations relative to the millimeter-wave radar can be obtained from multiple angles while the target object is stationary. The echo I / Q signal can be sustained for a first preset duration, with a sliding time window set to the first preset time, and the step size being a second pre-examination time.

[0045] In one example, a 77 GHz millimeter-wave radar can be used to acquire the echo I / Q signals of the target object. This 77 GHz millimeter-wave radar includes components such as a transmitting antenna and a receiving antenna. The radar parameters can be set to a starting frequency of 77 GHz, a frequency modulation slope of 30 MHz / μs, a frame period of 40 ms, and a slow-time sampling rate of 25 Hz. The target object to be detected is positioned at distances of 0.5 m, 1.0 m, and 1.5 m, and at angles of 0°, 30°, and 60°, respectively, and its echo I / Q signals are acquired over 120 seconds using the 77 GHz millimeter-wave radar, with an 18-second sliding time window and a step size of 1 second.

[0046] S202. Remove static noise and human motion noise from the echo I / Q signal to obtain the target signal.

[0047] In this embodiment, after acquiring the echo I / Q signal of the target object, denoising processing is performed on the echo I / Q signal. Specifically, a distance-dimensional Fast Fourier Transform (FFT) can be performed on the acquired echo I / Q signal to convert the signal from the time domain to the frequency domain, thereby obtaining signal characteristics at different distances to identify the target signal. Next, the CA-CFAR adaptive threshold filtering algorithm is applied. This algorithm uses the historical amplitude of the same distance cell as a reference to calculate a dynamic threshold, and then compares the amplitude of the cell to be detected with the threshold to distinguish the target signal from background noise. Finally, an improved moving average filtering method is used to further process the signal to remove static clutter and human motion noise, improving signal purity and obtaining the target signal.

[0048] Continuing from the previous example, after acquiring the 120s echo I / Q signal of the target object, the signal characteristics at different distances are obtained by using distance-dimensional FFT. Then, CA-CFAR is used to filter and distinguish the target signal from the background noise. The threshold coefficient can be set to 3.5. Finally, an improved moving average filter is used with a phase difference threshold of 1.0 to process and obtain the target signal.

[0049] S203. Filter the target signal to obtain the heartbeat signal sequence.

[0050] In this embodiment, the obtained target signal is filtered. Specifically, a Finite Impulse Response (FIR) filter can be used to process the denoised signal. The passband range of the FIR filter can be set to 0.8Hz to 2Hz, allowing signals within this frequency range to pass smoothly while attenuating signals outside the passband. Simultaneously, the stopband attenuation of the filter can be set to no less than 40dB. Within the stopband range, the signal amplitude is significantly reduced, effectively suppressing unwanted frequency components. The filter order can be set to 40, making the filter's frequency response closer to the preset state and better distinguishing between the passband and stopband. Then, the target signal processed by the FIR filter is sampled at a sampling rate of 25Hz, with an 18-second time window. The signal is segmented to obtain a series of 18-second signal segments. By analyzing these signal segments, the preprocessed heartbeat signal sequence S(t) can be obtained.

[0051] S204. Perform peak search on the heartbeat signal sequence, extract multiple signal peak points, and obtain the signal peak point sequence.

[0052] In this embodiment, after obtaining the heartbeat signal sequence, a peak search is performed on the heartbeat signal sequence to extract multiple signal peak points, and a peak point sequence is obtained based on these signal peak points. Specifically, the obtained heartbeat signal sequence is scanned point by point to determine whether each point is a peak point, and signal peak points that meet the conditions are extracted. The number of peak points can be set to be no less than 30, and the time interval between adjacent peak points can not exceed 1.25 seconds, thereby obtaining the peak point sequence.

[0053] S205. Normalize all signal peaks in the signal peak sequence, and calculate the mean square error of the normalized signal peak sequence to obtain the mean square error of the normalized signal peak sequence.

[0054] In this embodiment, after obtaining the signal peak point sequence, the maximum amplitude value in the peak point sequence is obtained. Normalization is then performed on all signal peak points in the signal peak point sequence. The mean of the normalized peak point sequence is calculated using the normalized peak point sequence, and thus the root mean square error of the normalized peak point sequence is obtained. The formula for calculating the root mean square error of the normalized peak point sequence can be:

[0055] Where M represents the mean squared error, and n represents the number of peak points. Represents the peak point. This represents the mean of the peak point sequence.

[0056] S206. When the mean square error is less than or equal to the set threshold, output the heartbeat signal sequence.

[0057] In this embodiment, after obtaining the mean squared error of the peak sequence, a preset mean squared error threshold is acquired. The mean squared error of the peak sequence is compared with the preset mean squared error threshold. When the mean squared error of the peak sequence is less than or equal to the preset mean squared error threshold, a heartbeat signal sequence is output. The MUSIC algorithm is then called to calculate the heart rate of the heartbeat signal sequence to obtain the target heart rate of the target object.

[0058] In one example, through multiple sets of experiments, it was verified that the preset mean squared error threshold can be C=0.3. This threshold can achieve an interference identification accuracy of 98.2%. When the mean squared error of the peak sequence is less than or equal to the preset mean squared error threshold, i.e., M... When C≤0, the peak fluctuation amplitude caused by respiratory harmonics does not exceed 30%, and the output heartbeat signal sequence is generated.

[0059] In an optional embodiment of this application, when the mean square error of the peak sequence is greater than a preset mean square error threshold, respiratory harmonic interference suppression processing is performed. The processing flow is detailed below. Figure 3 .

[0060] Figure 2 The illustrated process provides a method for processing heartbeat signals. It involves collecting echo I / Q signals from a seated target object at multiple angles at different locations relative to the millimeter-wave radar. Static clutter and human motion noise are removed from the echo I / Q signals to obtain the target signal. The target signal is then filtered to obtain a heartbeat signal sequence. Peak search is performed on the heartbeat signal sequence to extract multiple signal peak points, resulting in a signal peak point sequence. All signal peak points in the signal peak point sequence are normalized, and the mean square error (MSE) is calculated on the normalized signal peak point sequence to obtain the MSE. The heartbeat signal sequence is output when the MSE is less than or equal to a set threshold. It accurately identifies respiratory harmonic interference scenarios, improving recognition accuracy and avoiding redundant processing in the absence of interference; the second-order phase difference only suppresses respiratory harmonics without affecting the heartbeat signal, thus improving the heartbeat signal retention rate; it has strong compatibility, requires no additional hardware, and can be directly integrated into existing 77GHz millimeter-wave radar heart rate detection systems, adapting to long detection distances and wide detection angles; the average accuracy and precision of heart rate estimation are significantly improved.

[0061] Figure 3 This is a flowchart illustrating an embodiment of a breathing harmonic interference suppression method provided in this application. Figure 3 Includes the following steps: S301. When the mean square error is greater than a set threshold, the phase of the heartbeat signal sequence is extracted to obtain the phase sequence.

[0062] In this embodiment, when the mean square error of the peak sequence is greater than a preset mean square error threshold, it indicates the presence of respiratory harmonic interference, i.e., respiratory harmonics cause the peak fluctuation amplitude to exceed 30%, and respiratory harmonic interference suppression processing is performed. Specifically, phase extraction is performed on the heartbeat signal sequence to obtain the I-channel and Q-channel signals of the heartbeat signal sequence. For each time point, the phase is calculated using the arctangent function to obtain a series of phase values, thereby obtaining the phase sequence.

[0063] In one example, the I-channel and Q-channel signals of the heartbeat signal sequence are obtained. For each time point t, the phase is calculated using the arctangent function, which can be:

[0064] A series of phase values ​​are obtained, which in turn yields a phase sequence. .

[0065] S302. Perform a first-order difference operation on the phase sequence to obtain a first-order phase difference sequence.

[0066] In this embodiment, a first-order difference operation is performed on the obtained phase sequence to obtain a first-order phase difference sequence. Specifically, for each time point t, the phase difference between adjacent time points can be calculated to obtain a first-order phase difference sequence, where each element of the first-order phase difference sequence is the difference between adjacent elements in the phase sequence. This first-order phase difference sequence can be:

[0067] By using first-order differential operations, low-frequency respiratory harmonics are separated from high-frequency heartbeat signals, and the fundamental influence of respiratory harmonics is weakened, creating conditions for precise interference suppression in subsequent second-order differential operations, and indirectly ensuring the integrity of the heartbeat signal.

[0068] S303. Perform a second-order difference operation on the first-order phase difference sequence to remove breathing harmonics in the phase sequence and obtain a second-order phase difference sequence.

[0069] In this embodiment, after obtaining the first-order phase difference sequence, a second-order difference operation is performed on the first-order phase difference sequence. For each time point t in the first-order phase difference sequence, the difference between the difference value at the current time point and the difference value at the previous time point is calculated to obtain a new difference value. Based on the new difference value, a second-order phase difference sequence is obtained to remove breathing harmonics from the phase sequence. This second-order phase difference sequence can be:

[0070] By using second-order difference, respiratory harmonic interference (including higher harmonics) is precisely suppressed, while the high-frequency fluctuation characteristics of the heartbeat signal are fully preserved, thus inheriting the signal separation effect of first-order difference.

[0071] S304. Perform a moving average smoothing process on the second-order phase difference sequence to obtain a heartbeat signal sequence.

[0072] S305, outputs a heartbeat signal sequence.

[0073] The following is a unified discussion of S304-S305.

[0074] In this embodiment, after obtaining the second-order phase difference sequence, a moving average smoothing process is performed on the second-order phase difference sequence. This smooths out irregular high-frequency spikes in the signal after second-order difference processing without damaging the core features of the heartbeat signal, stabilizing the signal fluctuation trend and ensuring the integrity and stability of the reconstructed signal S′(t), thus laying the foundation for high-precision heart rate estimation. A window length of 5 sampling points can be set. For each time point t, the second-order phase difference values ​​of the current time point and the four time points preceding it (a total of 5 time points) are calculated to reconstruct the interference-free heartbeat signal sequence.

[0075] Moving average smoothing can be performed using the following formula:

[0076] Where t represents the current time point, and k represents the index of the sampling point within the moving average window, with k ranging from 0 to 4.

[0077] The de-interference heartbeat signal sequence is output for subsequent MUSIC algorithm heart rate estimation to obtain the target heart rate of the target object.

[0078] Figure 3 The illustrated process provides a method for suppressing respiratory harmonic interference. When the mean square error exceeds a set threshold, phase extraction is performed on the heartbeat signal sequence to obtain a phase sequence. A first-order difference operation is performed on the phase sequence to obtain a first-order phase difference sequence. A second-order difference operation is then performed on the first-order phase difference sequence to remove respiratory harmonics, resulting in a second-order phase difference sequence. Finally, the heartbeat signal sequence is output. Firstly, the second-order phase difference operation only suppresses respiratory harmonics without affecting the heartbeat signal, thus improving the heartbeat signal retention rate. It boasts strong compatibility, requiring no additional hardware and can be directly integrated into existing 77GHz millimeter-wave radar heart rate detection systems, adapting to long detection distances and wide detection angles. The average accuracy and precision of heart rate estimation are significantly improved.

[0079] Figure 4 This is a block diagram illustrating an embodiment of a heartbeat signal processing apparatus provided in this application. Figure 4 As shown, the device includes: Acquisition module 401 is used to acquire echo I / Q signals of target objects via millimeter-wave radar; The noise reduction module 402 is used to remove static noise and human motion noise from the echo I / Q signal to obtain the target signal; Filtering module 403 is used to filter the target signal to obtain a heartbeat signal sequence; The determination module 404 is used to determine the peak point sequence of the heartbeat signal sequence and calculate the mean square error corresponding to the peak point sequence; The output module 405 is used to output the heartbeat signal sequence when the mean square error is less than or equal to a set threshold.

[0080] In one possible implementation, the device further includes: The second-order phase difference module 406 is specifically used to extract the phase of the heartbeat signal sequence when the mean square error is greater than a set threshold, to obtain a phase sequence; to perform second-order difference processing on the phase sequence to obtain a second-order phase difference sequence; to perform moving average smoothing processing on the second-order phase difference sequence to obtain a heartbeat signal sequence; and to output the heartbeat signal sequence.

[0081] In one possible implementation, the second-order phase difference module 406 is specifically used to perform a first-order difference operation on the phase sequence to obtain a first-order phase difference sequence; and to perform a second-order difference operation on the first-order phase difference sequence to remove breathing harmonics in the phase sequence to obtain a second-order phase difference sequence.

[0082] In one possible implementation, the determining module 404 is specifically used to perform peak search on the heartbeat signal sequence, extract multiple signal peak points to obtain a signal peak point sequence; normalize all signal peak points in the signal peak point sequence, and calculate the mean square error of the normalized signal peak point sequence to obtain the mean square error of the normalized signal peak point sequence.

[0083] In one possible implementation, the acquisition module 401 is specifically used to acquire echo I / Q signals of the target object at different positions of the target object and the millimeter-wave radar at multiple angles when the target object is seated; wherein the echo I / Q signals can be sustained for a first preset duration, and a sliding time window of the first preset time is set with a step size of a second pre-screening time.

[0084] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0085] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented using software plus a general-purpose hardware platform, or of course, using hardware. Based on this understanding, the above technical solutions, in essence or the parts that contribute to the related technology, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0086] It should be understood that the terminology used herein is for the purpose of describing particular exemplary embodiments only and is not intended to be limiting. Unless the context clearly indicates otherwise, the singular forms “a,” “an,” and “described” as used herein may also include the plural forms. The terms “comprising,” “including,” “containing,” and “having” are inclusive and therefore indicate the presence of the stated features, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, steps, operations, elements, components, and / or combinations thereof. The method steps, processes, and operations described herein are not construed as requiring them to be performed in a particular order described or illustrated unless the order of performance is explicitly indicated. It should also be understood that additional or alternative steps may be used.

[0087] The above description is merely a specific embodiment of this application, enabling those skilled in the art to understand or implement this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A method for processing heartbeat signals, characterized in that, The method includes: Acquire echo I / Q signals of the target object using millimeter-wave radar; The target signal is obtained by removing static clutter and human motion noise from the echo I / Q signal; The target signal is filtered to obtain a heartbeat signal sequence; Determine the peak point sequence of the heartbeat signal sequence, and calculate the mean square error corresponding to the peak point sequence; When the mean square error is less than or equal to a set threshold, the heartbeat signal sequence is output.

2. The method according to claim 1, characterized in that, The method further includes: When the mean square error is greater than a set threshold, phase extraction is performed on the heartbeat signal sequence to obtain a phase sequence; The phase sequence is subjected to second-order difference processing to obtain a second-order phase difference sequence; The second-order phase difference sequence is smoothed by moving average to obtain the heartbeat signal sequence; Output the heartbeat signal sequence.

3. The method according to claim 2, characterized in that, The step of performing second-order difference processing on the phase sequence to obtain a second-order phase difference sequence includes: Perform a first-order difference operation on the phase sequence to obtain a first-order phase difference sequence; A second-order difference operation is performed on the first-order phase difference sequence to remove the breathing harmonics in the phase sequence, resulting in a second-order phase difference sequence.

4. The method according to claim 2, characterized in that, The moving average smoothing process is performed using the following formula: Where t represents the current time point, and k represents the index of the sampling point within the moving average window.

5. The method according to claim 1, characterized in that, Determining the peak point sequence of the heartbeat signal sequence and calculating the mean square error corresponding to the peak point sequence includes: Peak search is performed on the heartbeat signal sequence to extract multiple signal peak points, resulting in a signal peak point sequence; All signal peaks in the signal peak point sequence are normalized, and the mean square error of the normalized signal peak point sequence is calculated to obtain the mean square error of the normalized signal peak point sequence.

6. The method according to claim 1, characterized in that, The acquisition of echo I / Q signals of the target object via millimeter-wave radar includes: At different locations of the target object from the millimeter-wave radar, the echo I / Q signals of the target object under static conditions at multiple angles are obtained; The echo I / Q signal can be sustained for a first preset duration, and a sliding time window with a step size of a second pre-review time is set.

7. The method according to any one of claims 1-6, characterized in that, The method further includes: The MUSIC algorithm is used to calculate the heart rate of the heartbeat signal sequence to obtain the target heart rate of the target object.

8. A heartbeat signal processing device, characterized in that, The device includes: The acquisition module is used to acquire the echo I / Q signals of the target object through millimeter-wave radar; The noise reduction module is used to remove static noise and human motion noise from the echo I / Q signal to obtain the target signal; A filtering module is used to filter the target signal to obtain a heartbeat signal sequence; The determination module is used to determine the peak point sequence of the heartbeat signal sequence and calculate the mean square error corresponding to the peak point sequence; The output module is used to output the heartbeat signal sequence when the mean square error is less than or equal to a set threshold.

9. The apparatus according to claim 8, characterized in that, The determining module is specifically used to perform peak search on the heartbeat signal sequence, extract multiple signal peak points to obtain a signal peak point sequence; normalize all signal peak points in the signal peak point sequence, and calculate the mean square error of the normalized signal peak point sequence to obtain the mean square error of the normalized signal peak point sequence.

10. The apparatus according to claim 8, characterized in that, The acquisition module is specifically used to collect the echo I / Q signals of the target object at different locations at the distance from the millimeter-wave radar, using multiple angles under which the target object is stationary.

Images