An adaptive window spectral peak searching device

Abstract

The application discloses a spectrum peak searching device with self-adaptive window, comprising: an initialization module, which is used to generate initialization parameters of an observation frequency band window; wherein the initialization parameters comprise: initial upper limit, initial lower limit, initial window adjustment length and initial window length coefficient of the frequency band window; a spectrum peak searching module, which is used to perform iterative calculation of upper limit and lower limit of the observation frequency band window according to the initialization parameters, and the upper limit and lower limit of each frequency band window are obtained in each iteration; wherein the window length coefficient and the window adjustment length are updated in each iteration, and the upper limit and lower limit of the observation frequency band window obtained in the iteration are used to search the spectrum peak at the next moment; and an observation heart rate value calculation module, which is used to calculate the observation heart rate value according to the spectrum peak corresponding to the observation frequency band window. The application can adaptively adjust the window of spectrum peak searching according to different motion scenes.

Description

Technical Field

[0001] This invention relates to the field of digital signal processing, and more particularly to an adaptive window peak search device. Background Technology

[0002] Currently, existing technologies typically acquire pulse wave signals by using sensors to collect signals from the superficial temporal artery and radial artery. These signals are then processed using bandpass filtering and Fast Fourier Transform (FFT), and fixed upper and lower frequency limits are set to search for peaks in the time-spectrum graph to calculate the heart rate. (See also...) Figure 1 This is a schematic diagram of the existing technology that uses the upper and lower limits of a fixed frequency band to search for spectral peaks and calculate heart rate values.

[0003] However, fixed upper and lower limits of the frequency band have the following limitations: First, they cannot meet the accuracy requirements for calculating heart rate during exercise. Because the range of heart rate changes during exercise is too large, fixed upper and lower limits of the spectrum observation cannot satisfy the requirements, resulting in a large error between the calculated and actual heart rate values. Second, they cannot adapt to spectral analysis methods for various types of exercise. This is because manually determining the type of exercise and then adding corresponding spectral observation intervals is a very complex and difficult process. Therefore, there is an urgent need for an adaptive window peak-searching device that can adapt to different types and intensities of exercise, accurately acquire changes in human heart rate, and reduce errors in heart rate calculation. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of the prior art by proposing an adaptive window peak search device that can adaptively adjust the peak search window according to different motion scenarios.

[0005] This invention provides an adaptive window peak searching device, comprising:

[0006] An initialization module is used to generate initialization parameters for the observation band window; wherein, the initialization parameters include: the initial upper limit, the initial lower limit, and the initial window adjustment length of the band window;

[0007] The peak search module is used to iteratively calculate the upper and lower limits of the observation frequency band window according to the initialization parameters. Each iteration yields the upper and lower limits of each frequency band window. In each iteration, the window length coefficient and window adjustment length are updated, and the upper and lower limits of the observation frequency band window obtained by the iteration are used to search for the spectral peaks at the next moment.

[0008] The heart rate calculation module is used to calculate the heart rate value based on the spectral peaks corresponding to the observation frequency band window.

[0009] This invention employs iterative calculation of the upper and lower limits of the observation frequency band window, enabling adaptive adjustment of the window size at different times. This allows for adaptation of the observation frequency band window to different exercise scenarios, intensities, and populations, avoiding significant errors in heart rate readings caused by manual window size adjustments or fixed window sizes. Consequently, it enables accurate acquisition of human heart rate changes.

[0010] Furthermore, the iterative calculation of the upper and lower limits of the observed frequency band window, with each iteration yielding the upper and lower limits for each frequency band window, includes:

[0011] When the first observed heart rate value corresponding to the first observed frequency band window at the current moment is within the normal range of human heart rate, the corresponding first window length coefficient is calculated based on the sequence length of the first observed heart rate value.

[0012] Based on the first observed heart rate value, the first window length coefficient, and the first window adjustment length, the first upper limit and the first lower limit of the first observed frequency band window are jointly corrected to obtain the second upper limit and the second lower limit of the second observed frequency band window at the next moment.

[0013] Furthermore, the adaptive window peak searching device, after calculating the second upper limit and the second lower limit, also includes:

[0014] Based on the first observed heart rate value and the initial observed heart rate value, the adjustment length of the first window at the current moment is adjusted to obtain the adjustment length of the second window at the next moment.

[0015] This invention adjusts the length of the first window at the current moment based on the first observed heart rate value and the initial observed heart rate value to obtain the length of the second window at the next moment. It can adjust the window length according to the correlation between observed heart rate and time, and can further accurately obtain the changes in human heart rate.

[0016] Furthermore, after calculating the second upper and lower limits, the following also includes:

[0017] If the second upper limit is less than the upper limit threshold, then the second upper limit is adjusted by an upper limit correction value; or...

[0018] If the second lower limit is greater than the lower limit threshold, then the second lower limit is corrected by the lower limit threshold.

[0019] This invention uses threshold judgment on the upper and lower limits of the observation frequency band window to remove interference from outliers and constrain the upper and lower limits of the adjusted observation frequency band window within a reasonable range that conforms to the actual scenario, thereby enabling more accurate acquisition of human heart rate changes.

[0020] Furthermore, the initialization module is used to generate initialization parameters for the observation frequency band window, including:

[0021] Based on the initial upper limit and the initial upper limit, search for the first spectral peak corresponding to the initial frequency band window, and calculate the initial observed heart rate value at the initial time based on the first spectral peak;

[0022] The initial window length coefficient is obtained based on the sequence length of the initial observed heart rate values.

[0023] Furthermore, the second upper limit and the second lower limit of the second observation frequency band window at the next moment can be expressed as follows:

[0024]

[0025]

[0026] Among them, h i p i and d i These represent the observed heart rate value at time i, the window length coefficient, and the window adjustment length, respectively. i+1 and Kd i+1 These are the upper and lower limits of the observation frequency band window at the (i+1)th time, respectively.

[0027] Furthermore, the first window length coefficient can be expressed as:

[0028]

[0029] Where h is the sequence length of the observed heart rate values.

[0030] Furthermore, the second window adjustment length at the next moment can be expressed as:

[0031]

[0032] Where i≥2 indicates that the window length adjustment begins at the second time step, and h ０ The initial observed heart rate value, d 0, Adjust the window length for the initial time step and the first time step.

[0033] Preferably, the upper limit threshold is 3.5, the lower limit threshold is 0.9, and the lower limit correction value is 0.83Hz.

[0034] Preferably, the initial upper limit is 0.8Hz, the initial lower limit is 2.0Hz, and the initial window adjustment length is 0.68Hz. Attached Figure Description

[0035] Figure 1This is a schematic diagram of the structure of the adaptive window peak search device provided in an embodiment of the present invention;

[0036] Figure 2 is a comparative schematic diagram of peak searching provided in an embodiment of the present invention. Detailed Implementation

[0037] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0038] See Figure 1 This is a schematic diagram of the adaptive window peak search device provided in an embodiment of the present invention, including: an initialization module 11, a peak search module 12, and an observed heart rate value calculation module 13.

[0039] Initialization module 11 is used to generate initialization parameters for the observation band window; wherein, the initialization parameters include: initial upper limit, initial lower limit, initial window adjustment length, and initial window length coefficient of the band window.

[0040] Specifically, based on the initial upper limit and the initial upper limit, the first spectral peak corresponding to the initial frequency band window is searched, and based on the first spectral peak, the initial observed heart rate value at the initial time is calculated; based on the sequence length of the initial observed heart rate value, the initial window length coefficient is obtained.

[0041] Preferably, the initial upper limit is 0.8Hz, the initial lower limit is 2.0Hz, and the initial window adjustment length is 0.68Hz.

[0042] It is worth noting that the initialization module 11 mainly initializes the upper and lower limits of the frequency band window, as well as the window adjustment length and window length coefficient. Specifically, at the beginning of program execution, based on the actual heart rate value of the human body at rest, i.e., 48-120 bpm, the acquired pulse wave signal is bandpass filtered by 0.5-3.5 Hz to remove low-frequency noise (power frequency, baseline drift, electromyography) and high-frequency noise in the pulse signal. The pulse signal is then subjected to a fast Fourier transform to obtain the time spectrum of the pulse signal. The initial lower and upper limits are used as the range of the observation frequency band window at the initial moment. The first spectral peak is searched within this range, and the initial observed heart rate value is calculated based on the first spectral peak. The initial window length coefficient can be obtained based on the sequence length of the initial observed heart rate value.

[0043] The spectral peak search module 12 is used to iteratively calculate the upper and lower limits of the observation frequency band window according to the initialization parameters. Each iteration yields the upper and lower limits of each frequency band window. In each iteration, the window length coefficient and window adjustment length are updated, and the upper and lower limits of the observation frequency band window obtained by the iteration are used to search for the spectral peaks at the next moment.

[0044] The process involves iteratively calculating the upper and lower limits of the observed frequency band window. Each iteration yields the upper and lower limits for each frequency band window. This includes: when the first observed heart rate value corresponding to the first observed frequency band window at the current moment is within the normal range of human heart rate, calculating the corresponding first window length coefficient based on the sequence length of the first observed heart rate value; and adjusting the first upper and lower limits of the first observed frequency band window based on the first observed heart rate value, the first window length coefficient, and the first window adjustment length to obtain the second upper and lower limits of the second observed frequency band window at the next moment. The first window adjustment length at the current moment is adjusted based on the first observed heart rate value and the initial observed heart rate value to obtain the second window adjustment length at the next moment. If the second upper limit is less than the upper limit threshold, an upper limit correction value is assigned to the second upper limit for correction; or, if the second lower limit is greater than the lower limit threshold, the lower limit threshold is assigned to the second lower limit for correction.

[0045] It is worth noting that the peak search module 12 mainly performs a weighted calculation on the upper and lower limits of the observation frequency band obtained in each iteration, based on the observed heart rate value at the previous moment, the window adjustment length at the previous moment, and the sequence length of the observed heart rate value at the previous moment, to obtain the upper and lower limits of the observation frequency band window for the peak search at the next moment. That is, the spectrum search range of the pulse wave at the next moment is constrained within the upper and lower limits of the observation frequency band window calculated at the previous moment, and the observed heart rate value at the next moment is constrained by the lower and upper limits of the frequency of the observation frequency band window calculated at the previous moment. The window adjustment length and the sequence length of the observed heart rate value are updated in each iteration. Therefore, the observation frequency band window for calculating the observed heart rate value is updated in real time based on the heart rate value sequence at the previous time point. That is, the range of the observation frequency band window is different for each iteration, so that the range of the observation frequency band can be adaptively adjusted for different scenarios. In addition, the upper and lower limits of the calculated observation frequency band window are judged based on the upper and lower limit thresholds. If the calculated upper and lower limits deviate from the set upper and lower limit thresholds, the upper and lower limits of the window need to be corrected to ensure that the upper and lower limits of the adaptive observation frequency band window do not exceed the range of the observed bandpass filter, so as to avoid deviating from the actual range and thus ensure the reliability and authenticity of the measured heart rate value.

[0046] Preferably, the upper limit threshold is 3.5Hz, the lower limit threshold is 0.9Hz, and the lower limit correction value is 0.83Hz.

[0047] Preferably, the lower limit correction value is 0.83Hz.

[0048] Preferably, within the observation frequency window constrained by the upper and lower limits, the highest point of the spectral peak is searched as the frequency of the optimal peak, and the observed heart rate value is calculated based on the frequency of the optimal peak.

[0049] This invention uses threshold judgment on the upper and lower limits of the observation frequency band window to remove interference from outliers and constrain the upper and lower limits of the adjusted observation frequency band window within a reasonable range that conforms to the actual scenario, thereby enabling more accurate acquisition of human heart rate changes.

[0050] Preferably, the second upper limit and the second lower limit of the second observation frequency band window at the next moment can be expressed as follows:

[0051]

[0052]

[0053] Among them, h i p ｉ and d ｉ Ku represents the observed heart rate value at time i, the window length coefficient, and the window adjustment length, respectively. ｉ+1 and Kd i+1 These are the upper and lower limits of the observation frequency band window at the (i+1)th time, respectively.

[0054] Preferably, the first window length coefficient can be expressed as:

[0055]

[0056] Where h is the sequence length of the observed heart rate values.

[0057] The heart rate calculation module 13 is used to calculate the heart rate value based on the spectral peaks corresponding to the observation frequency band window.

[0058] Preferably, the second window adjustment length at the next moment can be expressed as:

[0059]

[0060] Where i≥2 indicates that the window adjustment length begins to be adjusted at the second time point, h0 is the initial observed heart rate value, and d 0, Adjust the window length for the initial time step and the first time step.

[0061] This invention adjusts the length of the first window at the current moment based on the first observed heart rate value and the initial observed heart rate value to obtain the length of the second window at the next moment. It can adjust the window length according to the correlation between observed heart rate and time, and can further accurately obtain the changes in human heart rate.

[0062] This invention employs iterative calculation of the upper and lower limits of the observation frequency band window, enabling adaptive adjustment of the window size at different times. This allows for adaptation of the observation frequency band window to different exercise scenarios, intensities, and populations, avoiding significant errors in heart rate readings caused by manual window size adjustments or fixed window sizes. Consequently, it enables accurate acquisition of human heart rate changes.

[0063] Referring to Figure 2, which is a comparative schematic diagram of the peak search provided in the embodiments of the present invention, Figure 2(a) is a flowchart of the peak search process used in the prior art. In Figure 2(a), the prior art often uses the upper and lower limits of a fixed frequency band window for sampling, which cannot adapt to motion scenarios. Adaptively adjusting the upper and lower limits of the frequency band window has limitations and low applicability to motion scenarios. Figure 2(b) is a flowchart of the peak search process with an adaptive window provided in the embodiments of the present invention. The present invention obtains an observation window of pulse wave frequency by performing weighted calculation on the heart rate value sequence at past time points, which is used to constrain the calculation range of the current heart rate value and improve the calculation accuracy of the heart rate value. Specifically, in Figure 2(b), the superficial and radial artery signals are acquired through a sensor in the flexible tactile eye. Based on these signals, a pulse wave signal is obtained. The pulse wave signal is processed by bandpass filtering from 0.5 to 3.5 Hz using photoplethysmography (PPG), and a spectral transformation is performed to obtain the time-spectrum diagram of the pulse wave signal. The spectral transformation includes MUSIC transformation, FFT transformation, or HTT transformation. The range of the out-of-control spectrum diagram is constrained by the upper and lower limits of the observation spectrum window calculated in the previous time step. Spectral peaks are searched within the adaptive observation spectrum window. The observed heart rate value at the current time step is calculated based on the searched spectral peaks. The window length coefficient and window adjustment length are updated to adjust the upper and lower limits of the observation spectrum window used to calculate the next time step. If the upper limit of the next time step is less than the lower limit threshold, the lower limit threshold is applied. If the upper limit of the next time step is greater than the upper limit threshold, the upper limit correction value is applied. At the initial time step, the initial values ​​of the observation spectrum window are set, including the initial upper limit and the initial lower limit. Compared with Figure 2(a), the present invention adds adaptive upper and lower limits calculation of the range of the observation spectrum window based on the parameters corresponding to the observation spectrum window at past time points, thereby calculating the observed heart rate value.

[0064] The embodiments of the present invention employ adaptive window peak search, which can adapt the sampling window for different motion scenarios. During the motion, the observation frequency band window for the next moment is adjusted based on the past heart rate value sequence, the current heart rate value observation window interval is updated in real time, and reasonable window update weight parameters are set, which is closer to the real scene and has higher applicability.

[0065] Those skilled in the art will understand that embodiments of this application may also include computer program products. Therefore, this application may take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application may take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0066] This application is described with reference to flowchart illustrations and / or block diagrams of apparatuses, devices (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create a machine for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0067] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0068] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0069] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A peak-searching device with an adaptive window, characterized in that, include: An initialization module is used to generate initialization parameters for the observation band window; wherein, the initialization parameters include: initial upper limit, initial lower limit, initial window adjustment length, and initial window length coefficient of the band window; The peak search module is used to calculate the corresponding first window length coefficient based on the sequence length of the first observed heart rate value when the first observed heart rate value corresponding to the first observed frequency band window at the current time is within the normal range of human heart rate values; to correct the first upper limit and the first lower limit of the first observed frequency band window based on the first observed heart rate value, the first window length coefficient, and the first window adjustment length, thereby obtaining the second upper limit and the second lower limit of the second observed frequency band window at the next time; to adjust the first window adjustment length at the current time based on the first observed heart rate value and the initial observed heart rate value, thereby obtaining the second window adjustment length at the next time; if the second upper limit is less than the upper limit threshold, then the second upper limit is corrected by the upper limit correction value; or, if the second lower limit is greater than the lower limit threshold, then the second lower limit is corrected by the lower limit threshold; wherein, the window length coefficient and the window adjustment length are updated in each iteration, and the upper and lower limits of the observed frequency band window obtained in the iteration are used to search for the peak at the next time; the second upper limit and the second lower limit of the second observed frequency band window at the next time can be expressed as follows: ， ， in, , and These represent the observed heart rate value at time i, the window length coefficient, and the window adjustment length, respectively. and These are the upper and lower limits of the observation frequency band window at time i+1, respectively; the adjustment length of the second window at the next time step can be expressed as: ， in, This indicates that the window resizing process only begins at the second time point. The initial observed heart rate value. Adjust the window length for the initial time step and the first time step; The heart rate calculation module is used to calculate the heart rate value based on the spectral peaks corresponding to the observation frequency band window.

2. The adaptive window peak searching device as described in claim 1, characterized in that, The initialization module is used to generate initialization parameters for the observation frequency band window, including: Based on the initial upper limit and the initial lower limit, search for the first spectral peak corresponding to the initial frequency band window, and calculate the initial observed heart rate value at the initial time based on the first spectral peak. The initial window length coefficient is obtained based on the sequence length of the initial observed heart rate values.

3. The adaptive window peak searching device as described in claim 1, characterized in that, The first window length coefficient can be expressed as: ， in, The length of the sequence of observed heart rate values.

4. The adaptive window peak searching device as described in claim 1, characterized in that, The upper limit threshold is 3.5 Hz, the lower limit threshold is 0.9 Hz, and the lower limit correction value is 0.83 Hz.

5. The adaptive window peak searching device as described in claim 1, characterized in that, The initial upper limit is 2.0Hz, the initial lower limit is 0.8Hz, and the initial window adjustment length is 0.68Hz.

Images