Intelligent wearable device cross-terminal health data synchronization method

Abstract

The application discloses a kind of intelligent wearable equipment cross-terminal health data synchronization method, for in weak positioning or positioning lag working condition, the history activity heat abnormality of signal misjudgment caused by non-human condition is backtracked correction.Method locks physical time anchor point when exclusive evidence triggers correction, executes asymmetric bidirectional scanning to determine starting boundary and termination boundary to obtain correction time period information, and sends heat correction request information including correction time period information, physiological load data, inertial motion intensity sequence and mismatch tolerance coefficient.Terminal equipment generates basic physiological heat component based on physiological load data, constructs dynamic upper limit constraint inertial motion intensity to obtain limited activity heat component and generates corrected heat data;Boundary smoothing compensation is carried out at termination boundary, and mismatch tolerance coefficient is returned according to compensation amplitude calibration update, to realize consistent with physiological record of same time period and form cross-terminal data synchronization closed loop.

Description

Technical Field

[0001] This invention relates to the field of health data processing technology, and more specifically, to a method for synchronizing health data across terminals in smart wearable devices. Background Technology

[0002] Health data processing technology is commonly used for cross-device health data synchronization in smart wearable devices. Existing synchronization solutions typically perform activity identification and calorie estimation on the device side, write the results into historical calorie records, and then synchronize them with the terminal device. Simultaneously, physiological signals can be collected to characterize the physiological state over the same time period, supporting the calculation and display of derived indicators such as stress or recovery metrics. In practical use, motion-related signals collected by wearable devices under different operating conditions may include non-human operating condition components such as vehicle movement and equipment vibration. This is especially prone to misjudgment in activity identification and calorie estimation under weak or delayed localization, thus necessitating the need for retrospective correction of historical calorie records.

[0003] The existing technology has the following shortcomings: On the one hand, under conditions of weak or delayed positioning, non-human working condition signals such as vehicle movement and equipment vibration collected by inertial sensors are easily misidentified as human activity, leading to an abnormally high estimation of activity heat and its inclusion in historical heat records. Due to limited observability and confusing causes, the system struggles to distinguish between non-human working conditions and real human activity in a timely and reliable manner, causing the deviation to accumulate. On the other hand, when subsequent evidence indicating non-human working conditions emerges and triggers backtracking correction, simply rolling back or adjusting the activity heat may lead to inconsistencies with physiological records from the same period, causing abnormal shifts in stress-related or recovery-related derived indicators during that period. Furthermore, insufficient boundary transition processing during backtracking correction may result in discontinuities in the cumulative performance between the corrected segment and subsequent historical segments, affecting the maintenance of data consistency after cross-terminal synchronization. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art, embodiments of the present invention provide a method for synchronizing health data across terminals of smart wearable devices to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: A method for synchronizing health data across terminals in a smart wearable device includes: S101, when the smart wearable device triggers a retrospective correction of historical calorie records, the timing of the correction is locked as a physical time anchor point; an asymmetric bidirectional scan is performed based on the physical time anchor point, tracing back along the historical direction to the moment when the ratio of inertial motion intensity to physiological load intensity first exceeds the mismatch tolerance coefficient as the starting boundary; the decay trend of inertial signal energy is tracked along the future direction, and the termination boundary is determined when the inertial signal energy shows a decay state and the ratio of inertial motion intensity to physiological load intensity falls back to not exceed the mismatch tolerance coefficient; the correction time period information is determined based on the starting boundary and the termination boundary; a calorie correction request information containing the correction time period information, physiological load data within the correction time period, inertial motion intensity sequence within the correction time period, and mismatch tolerance coefficient is generated and sent to the terminal device; S102, after receiving the heat correction request information, the terminal device obtains the heat data within the correction period from the historical heat record and reconstructs it according to the basic physiological heat component and the restricted activity heat component; calculates the corresponding basic physiological heat component based on the physiological load data within the correction period; constructs a dynamic upper limit of the inertial motion intensity based on the physiological load data and the mismatch tolerance coefficient, and applies the upper limit constraint to the inertial motion intensity within the correction period to calculate the restricted activity heat component. S103, the terminal device generates corrected calorie data based on the basic physiological calorie components and the restricted activity calorie components; extracts the numerical continuity feature of the corrected calorie data at the termination boundary, wherein the numerical continuity feature is the difference between the corrected cumulative value at the end of the termination boundary and the cumulative value of the adjacent historical calorie records after the termination boundary; performs boundary smoothing compensation on the transition segment after the termination boundary, and writes the compensated final data into the historical calorie records; calibrates and updates the mismatch tolerance coefficient based on the compensation amplitude of the boundary smoothing compensation; and sends the compensated final data and the updated mismatch tolerance coefficient to the smart wearable device.

[0006] In a preferred embodiment, before performing the ratio calculation, the smart wearable device aligns the inertial motion data and physiological load data at a uniform time granularity, and performs aggregation or resampling on the data within each time granularity to generate an inertial motion intensity sequence and a physiological load intensity sequence under the same time reference, wherein the physical time anchor is the timestamp corresponding to the modified entry in the historical calorie record involved in the retrospective correction.

[0007] In a preferred embodiment, the inertial motion intensity and the inertial signal energy are calculated based on data from a triaxial accelerometer. The inertial motion intensity is the value obtained by performing mean or variance calculation on the acceleration modulus sequence within a sliding window corresponding to the time granularity, and the inertial signal energy is the value obtained by performing energy accumulation on the acceleration modulus sequence within the sliding window.

[0008] In a preferred embodiment, the physiological load intensity is calculated based on heart rate data extracted from the photoplethysmography signal. The physiological load intensity is the average heart rate value within the time granularity, or the current metabolic intensity value obtained based on the average heart rate value and combined with the user's basal metabolic rate mapping.

[0009] In a preferred embodiment, the initial value of the mismatch tolerance coefficient is derived from the user's historical steady-state profile. The smart wearable device determines the initial value of the mismatch tolerance coefficient based on the coupling dispersion of the inertial motion intensity and physiological load intensity in the uncorrected state. The coupling dispersion is the degree of fluctuation or deviation of the ratio of inertial motion intensity and physiological load intensity at a uniform time granularity. After receiving the updated mismatch tolerance coefficient returned by the terminal device, the historical steady-state profile is iteratively updated.

[0010] In a preferred embodiment, a denominator protection mechanism is introduced when calculating the ratio. A lower limit for physiological intensity is preset. When the physiological load intensity is lower than the lower limit, the lower limit is used instead of the physiological load intensity in the ratio calculation.

[0011] In a preferred embodiment, the determination of the termination boundary includes anti-jitter logic, establishing a termination confirmation window of a preset length, and determining the termination boundary only when the inertial signal energy within the termination confirmation window shows a decay state and the ratios corresponding to each time granularity fall back to no more than the mismatch tolerance coefficient.

[0012] In a preferred embodiment, 8. The process of constructing and constraining the dynamic upper limit includes: the terminal device obtains the dynamic upper limit by multiplying the physiological load intensity by the mismatch tolerance coefficient at each time granularity, and replaces the inertial motion intensity corresponding to the time granularity with the dynamic upper limit when the inertial motion intensity is greater than the dynamic upper limit.

[0013] In a preferred embodiment, when the terminal device performs boundary smoothing compensation, it calculates the difference between the cumulative value of the corrected heat data at the end of the termination boundary and the cumulative value of the adjacent historical heat records after the termination boundary, and distributes the difference within a preset transition length to form boundary smoothing compensation.

[0014] In a preferred embodiment, when calibrating and updating the mismatch tolerance coefficient, the terminal device adjusts the mismatch tolerance coefficient according to the compensation amplitude of the boundary smoothing compensation, and applies upper and lower limit constraints to the updated mismatch tolerance coefficient.

[0015] The present invention provides a method for synchronizing health data across terminals in smart wearable devices, highlighting its effectiveness and advantages. This invention locks the physical time anchor point when exclusive evidence triggers retrospective correction, and uses it as a benchmark to perform asymmetric bidirectional scanning to determine the start and end boundaries, thereby obtaining the correction time period information and generating heat correction request information to trigger terminal recalculation. The terminal device generates basic physiological heat components based on physiological load data within the correction time period, and constructs a dynamic upper limit for inertial motion intensity by multiplying the physiological load intensity by a mismatch tolerance coefficient. A dynamic upper limit constraint is applied to the inertial motion intensity sequence within the correction time period to obtain restricted activity heat components, thereby generating corrected heat data to avoid inconsistencies with physiological records from the same time period. At the end boundary, the terminal device distributes the accumulated difference within a preset transition length to form boundary smoothing compensation, reducing the cumulative discontinuity between the correction segment and subsequent historical segments. After calibrating and updating the mismatch tolerance coefficient based on the compensation amplitude, the data is transmitted back to the smart wearable device to support cross-terminal data synchronization closed loop and iterative updates of the user's historical steady-state profile. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the method flow of the present invention; Figure 2 This is a flowchart illustrating step S101 of the present invention. Detailed Implementation

[0017] 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.

[0018] This invention relates to a method for synchronizing health data across terminals in smart wearable devices, and more particularly to a process for retrospectively correcting activity-related calorie data in historical calorie records. Under conditions of weak or delayed localization, signals from non-human working conditions, such as vehicle movement and equipment vibration, collected by inertial sensors can easily be misidentified as human activity during activity recognition and calorie estimation, leading to an abnormally high level of activity-related calorie data in historical records. When subsequent, exclusive evidence indicating non-human working conditions emerges, retrospective correction of the activity-related calorie data for that period is necessary. Failure to correct will result in long-term accumulated biases, while direct rollback may lead to inconsistencies with physiological records from the same period, causing abnormal shifts in stress-related or recovery-related derived indicators during that time period.

[0019] Therefore, the technical problem to be solved by the present invention is to determine the interpretable correction period through asymmetric bidirectional scanning driven by physical time anchors in scenarios where non-human working conditions lead to miscalculation of activity heat and the emergence of exclusive evidence triggers retrospective correction. On the terminal device side, the activity heat is recalculated with physiological load constraints and boundary smoothing compensation to achieve consistency with physiological records in the same time period. At the same time, the mismatch tolerance coefficient is calibrated and transmitted back to support cross-terminal data synchronization closed loop.

[0020] Based on the above design, this invention constructs a complete process for a cross-terminal health data synchronization method for smart wearable devices, consisting of steps S101 to S103 sequentially. (Refer to...) Figure 1 , Figure 1 This is a schematic diagram of the method flow of the present invention.

[0021] Step S101, "Determining and Generating the Correction Time Period," is used in scenarios where exclusive evidence triggers retrospective correction. It locks the physical time anchor point and performs an asymmetric bidirectional scan based on it to determine the start and end boundaries of the correction time period. This step reads inertial motion data X101, physiological load data X102, and historical calorie records X103, generates and sends calorie correction request information R101 to the terminal device, allowing step S102 to directly perform calorie component reconstruction and dynamic upper limit constraints. Exclusive evidence indicates the presence of non-human working condition interference and triggers retrospective correction of historical calorie records. R101 encapsulates the correction time period boundaries, physiological load data within the period, inertial motion intensity sequence, and mismatch tolerance coefficient. In an optional implementation, R101 includes a time boundary group, a physiological load group, an inertial intensity group, and a tolerance coefficient group, respectively used to carry the start and end boundaries, the physiological load data within the correction time period, the inertial motion intensity sequence within the correction time period, and the mismatch tolerance coefficient.

[0022] Step S102, component reconstruction and dynamic upper limit constraint, is used by the terminal device to construct a dynamic upper limit based on physiological load and apply upper limit constraints to the inertial motion intensity sequence after receiving the heat correction request information. This yields the restricted activity heat component, which, together with the basic physiological heat component, forms the reconstruction result. This step reads the heat correction request information R101 and combines it with historical heat records X103 to generate a heat component reconstruction result set R102, which is then used by step S103 to generate corrected heat data and perform boundary smoothing compensation. The heat component reconstruction result set R102 encapsulates the basic physiological heat component, the restricted activity heat component, and the dynamic upper limit. This step is only responsible for component generation and constraint calculation. In an optional implementation, R102 includes a basic component group, a restricted component group, and a dynamic upper limit group, which respectively carry the basic physiological heat component, the restricted activity heat component, and the dynamic upper limit.

[0023] Step S103, "Correction Writing and Tolerance Calibration Feedback," is used to generate corrected heat data based on the heat component reconstruction result set R102, and to perform boundary smoothing compensation at the termination boundary to eliminate cumulative discontinuities. Simultaneously, it calibrates and updates the mismatch tolerance coefficient and feeds it back to form a cross-terminal closed loop. This step reads the heat component reconstruction result set R102 and combines it with historical heat records X103 to generate corrected heat data. It calculates the cumulative value difference and distributes it within a preset transition length to form boundary smoothing compensation. The compensated final data is written to the historical heat record X103, and the corrected heat data and calibration result set R103 are generated and sent to the smart wearable device. R103 encapsulates the corrected heat data, boundary smoothing compensation, and updated mismatch tolerance coefficient, enabling the smart wearable device to update the user's historical steady-state profile and improve the consistency of subsequent boundary judgments and dynamic upper limit construction. In an optional implementation, R103 includes a correction data group, a compensation group, and a tolerance update group, which are used to carry the corrected heat data, boundary smoothing compensation, and updated mismatch tolerance coefficient, respectively.

[0024] By implementing the above method, the present invention locks the physical time anchor point and performs asymmetric bidirectional scanning to determine the correction period when exclusive evidence triggers retrospective correction. This allows the terminal device to construct a dynamic upper limit based on physiological load and recalculate activity calories within this period to avoid inconsistencies with physiological records from the same time period. Furthermore, boundary smoothing compensation is performed at the termination boundary, and the mismatch tolerance coefficient is calibrated and updated back, ensuring that the correction results remain continuous in the cumulative caliber of the historical segment and forming a cross-terminal synchronous closed loop.

[0025] The implementation process and operational effects of the method of the present invention will be described in detail below with reference to specific embodiments. It should be understood that the embodiments are only used to illustrate the technical solution of the present invention, and not to limit it. The relevant steps, parameters, and module divisions can be appropriately adjusted without changing the essence of the invention.

[0026] In an optional embodiment, step S101 is performed by a smart wearable device, and its implementation process is summarized as aperture alignment and anchor point preparation, boundary determination, request encapsulation and transmission. (Refer to...) Figure 2 , Figure 2 This is a flowchart illustrating step S101 of the present invention.

[0027] In the aperture alignment and anchor point preparation, the smart wearable device acquires raw signals from inertial sensors such as the triaxial accelerometer from X101, aggregates or resamples the signals at a uniform time granularity, and generates an inertial motion intensity sequence and an inertial signal energy sequence. The inertial motion intensity can be obtained by performing mean or variance calculations on the acceleration modulus sequence within a sliding window corresponding to the time granularity, and the inertial signal energy can be obtained by accumulating the energy of the acceleration modulus sequence within the same window. The smart wearable device acquires physiological signals such as heart rate from X102, and generates a physiological load intensity sequence at the same time granularity, placing the inertial and physiological sequences under the same time reference. The smart wearable device locates the timestamp corresponding to the modified item involved in this backtracking correction from X103, locks this timestamp as the physical time anchor point, and simultaneously reads the necessary historical time context before and after the anchor point to limit the scanning range. The smart wearable device obtains the initial value of the mismatch tolerance coefficient from the user's historical steady-state profile or preset parameters, serving as a shared control parameter for starting boundary determination, ending boundary determination, and dynamic upper limit construction. Specifically, the user's historical steady-state profile is used to store the typical coupling relationship and fluctuation statistics of the user's inertial motion intensity and physiological load intensity in the uncorrected state, and provides the initial value of the mismatch tolerance coefficient for iterative updating after transmission.

[0028] In boundary determination, the smart wearable device traces back point by point along the historical direction using a physical time anchor point as the reference, calculating the ratio of inertial motion intensity to physiological load intensity at each time granularity. To avoid ratio distortion caused by excessively low physiological load, a denominator protection mechanism is introduced, with a preset lower limit for physiological load. When the physiological load intensity is below this lower limit, it is used in the ratio calculation. When the ratio first exceeds the mismatch tolerance coefficient during the tracing process, the corresponding moment is determined as the starting boundary. The smart wearable device tracks point by point along the future direction using a physical time anchor point as the reference, while simultaneously monitoring whether the inertial signal energy enters a decay state and whether the ratio falls back to below the mismatch tolerance coefficient. When the inertial signal energy is decaying and the ratio falls back to below the mismatch tolerance coefficient, the corresponding moment is determined as the termination boundary. To suppress premature termination caused by short-term oscillations, a termination confirmation window can be optionally established. The termination boundary is confirmed only when multiple consecutive time granularities within the window meet the conditions of continuous energy decay and ratio regression, and the end point of the window is taken as the termination boundary.

[0029] In the request encapsulation and transmission process, the smart wearable device forms a correction time period information based on the start and end boundaries. It extracts physiological load data and inertial motion intensity sequences within the correction time period from the original data stream and encapsulates them together with a mismatch tolerance coefficient to generate a heat correction request information R101, which is then sent to the terminal device. Optional settings for the maximum backtracking length and maximum forward tracking length limit the furthest search range. When the length is reached but the boundary conditions are not met, a degenerate caliber is used to determine the boundary and generate R101, ensuring that downstream steps complete recalculation under a unified entry caliber and refine the data only after sufficient data is available. Specifically, the maximum backtracking length limits the furthest time granularity span for backtracking along the historical direction, and the maximum forward tracking length limits the furthest time granularity span for tracking along the future direction.

[0030] To facilitate implementation and standardize boundary determination criteria, this embodiment provides an optional calculation scheme.

[0031] First define the ratio ;in, To unify time-granularity indexing, This is the time granularity index corresponding to the physical time anchor point. The inertial motion intensity is calculated from the inertial motion data X101. The physiological load intensity is calculated from the physiological load data multiplied by 10². The energy of the inertial signal is calculated from the inertial motion data X101. This is the mismatch tolerance coefficient. This is the lower limit of the physiological intensity benchmark. To terminate the confirmation window length. When traversing back along the historical direction, the first condition is met. The time granularity is denoted as the starting boundary. When tracing along a future direction, if in continuous... Simultaneously satisfying at each time granularity And satisfy Then the endpoint of the continuous interval is taken as the termination boundary. .

[0032] For example, The lower quantile of the resting-period physiological workload intensity can be used to avoid an artificially high ratio due to excessively low physiological workload. A value of 3 to 10 is recommended to prevent premature termination caused by short-term oscillations. A value of 1.0 to 5.0 can be used because this coefficient affects both the initial boundary threshold and the terminal dynamic upper limit multiple.

[0033] In an optional embodiment, step S102 is executed by the terminal device, and its implementation process is summarized as input parsing and target alignment, dynamic upper limit constraint and component generation, and result writing back.

[0034] In the input parsing and target alignment process, the terminal device parses the physiological load data inertial motion intensity sequence and mismatch tolerance coefficient from R101, and obtains the calorie data within the correction time period from X103 as the reference for reconstructing the target sequence and subsequent write-back. The terminal device calculates the basal physiological calorie component based on the physiological load data, making this component a metabolic calorie base unaffected by inertial signals from non-human working conditions. The basal physiological calorie component can be implemented by mapping metabolic intensity to average heart rate and combining it with the user's basal metabolic rate. The user's basal metabolic rate can be derived from the user's historical steady-state profile and used as the conversion benchmark for the physiological load intensity to the basal physiological calorie component.

[0035] In the dynamic upper limit constraint and component generation, the terminal device constructs a dynamic upper limit at each time granularity using physiological load intensity and mismatch tolerance coefficient, and compares the inertial motion intensity point by point with the dynamic upper limit. When the inertial motion intensity exceeds the dynamic upper limit, the inertial motion intensity at that time granularity is replaced with the dynamic upper limit to suppress outlier intensity caused by vibrations from non-human working conditions. When the inertial motion intensity does not exceed the dynamic upper limit, the original value is retained. The terminal device maps the constrained inertial motion intensity to a restricted activity heat component and aligns it with the basic physiological heat component at the same time granularity. The activity heat mapping can use linear conversion or piecewise conversion, and maintains consistency within consecutive sessions of the same user to ensure reproducibility of recalculations.

[0036] In the result write-back, the terminal device writes the basic physiological heat components, restricted activity heat components, and dynamic upper limit into the basic component group, restricted component group, and dynamic upper limit group of the heat component reconstruction result set R102, and stores R102 in association with the correction time period information, so that step S103 can directly read R102 to generate the corrected heat data and perform boundary smoothing compensation.

[0037] To facilitate implementation and standardize the construction and constraint criteria of dynamic upper limits, this embodiment presents an optional calculation scheme. First, construct the dynamic upper limit. And apply an upper limit constraint to the intensity of inertial motion. .

[0038] in, To correct the uniform time granularity index within the time period, The physiological load intensity is obtained from the heat correction request information R101. The mismatch tolerance factor is provided by the heat correction request information R101. The inertial motion intensity is provided by the heat correction request information R101. The upper limit is dynamic. This represents the intensity of the constrained inertial motion. For example, The value can be taken as 1.0 to 5.0 and kept consistent with step S101. This is because the coefficient affects both the terminal dynamic upper limit multiplier and the wearable side boundary judgment threshold. Keeping it consistent can reduce the drift of cross-terminal recalculation caliber.

[0039] In an optional embodiment, step S103 is performed by the terminal device, and its implementation process is summarized as corrected data synthesis, boundary smoothing compensation, and writing and transmission.

[0040] In the corrected data synthesis, the terminal device reads the basal physiological caloric components and restricted activity caloric components from the caloric component reconstruction result set R102, and adds them point-by-point at a uniform time granularity to obtain the corrected caloric data sequence. This sequence satisfies the dynamic upper limit constraint within the corrected time period, but there may be a cumulative discontinuity between the sequence and historical caloric records after the termination boundary at the termination boundary. Based on this, the terminal device calculates the corrected cumulative value at the end of the termination boundary and reads the adjacent historical cumulative value after the termination boundary from the historical caloric record X103 as the two benchmarks for the difference calculation.

[0041] In boundary smoothing compensation, the terminal device calculates the difference between the cumulative value of the corrected heat data at the end of the termination boundary and the cumulative value of the adjacent historical heat records after the termination boundary, and uses this difference as the compensation amplitude. The terminal device distributes the compensation amplitude within a preset transition length to form a boundary smoothing compensation sequence, and superimposes this compensation sequence onto the heat data of the transition section after the termination boundary, so that the corrected section and the subsequent historical section remain continuous in terms of cumulative caliber.

[0042] During the write and return process, the terminal device writes the compensated, final corrected heat data to the historical heat record X103, overwriting any abnormal heat records within the correction period. The terminal device calibrates and updates the mismatch tolerance coefficient based on the compensation amplitude, ensuring the updated mismatch tolerance coefficient remains applicable to subsequent start boundary determination, end boundary determination, and dynamic upper limit construction. The terminal device then writes the final corrected heat data, boundary smoothing compensation, and updated mismatch tolerance coefficient to the corrected heat and calibration result set R103 and sends it to the smart wearable device.

[0043] To facilitate implementation and standardize the criteria for boundary smoothing compensation and coefficient calibration, this embodiment provides an optional calculation scheme.

[0044] First calculate the compensation amplitude ;in, For the unified time granularity index corresponding to the termination boundary, This is the cumulative value of the corrected heat data. This is the cumulative value of historical heat records. The compensation amplitude for boundary smoothing compensation.

[0045] The compensation amplitude is apportioned over the transition length, for example, for... Pick ;in, This refers to the heat data at the time granularity corresponding to the transition period after the termination boundary, which is the historical heat record. For the transition length, This is the final heat data after smoothing. For example, Three to 12 time granularities can be selected because they smooth the seams without significantly extending the affected area.

[0046] Then, the mismatch tolerance coefficient is calibrated and updated based on the compensation amplitude. Time to take , when Time to take ;in, This represents the current mismatch tolerance factor. This is the updated mismatch tolerance factor. To calibrate the learning rate, For dead zone thresholds that do not require calibration, and These are the lower and upper limits of the updated coefficients. The lower limit of the cumulative value is used to avoid overshooting the update due to an excessively small denominator. It is a symbolic function.

[0047] For example, A value of 0.05 to 0.20 is acceptable because it reduces the risk of update oscillations. Stable period can be adopted The high quantile level is due to the suppression of noise-driven frequent updates. The lower limit of the cumulative value can be taken as being on the same order of magnitude as the smallest measurable unit of the terminal heat record, because this is to avoid amplifying the normalization term by making the cumulative value too small. A value of around 1.0 is acceptable. A value of around 5.0 is acceptable because it limits the reasonable range of the dynamic upper limit multiplier and is consistent with steps S101 to S102.

[0048] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0049] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0050] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and inventive constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0051] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0052] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0053] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for synchronizing health data across terminals in a smart wearable device, characterized in that, include: S101, When a smart wearable device triggers a retrospective correction of historical heat records, the timing of the correction is locked as a physical time anchor point. Using physical time anchors as a reference, an asymmetric bidirectional scan is performed. The starting boundary is the moment when the ratio of inertial motion intensity to physiological load intensity first exceeds the mismatch tolerance coefficient along the historical direction. The attenuation trend of inertial signal energy is tracked along the future direction, and the termination boundary is determined when the inertial signal energy shows a decaying state and the ratio of inertial motion intensity to physiological load intensity falls back to no more than the mismatch tolerance coefficient. The correction time period information is determined based on the starting boundary and the termination boundary. A heat correction request information containing the correction time period information, physiological load data within the correction time period, inertial motion intensity sequence within the correction time period, and mismatch tolerance coefficient is generated and sent to the terminal device. S102, after receiving the heat correction request information, the terminal device obtains the heat data within the correction period from the historical heat record and reconstructs it according to the basic physiological heat component and the restricted activity heat component; and calculates the corresponding basic physiological heat component based on the physiological load data within the correction period. Based on physiological load data and mismatch tolerance coefficient, a dynamic upper limit for inertial motion intensity is constructed, and an upper limit constraint is applied to the inertial motion intensity within the correction time period to calculate the restricted activity heat component. S103, The terminal device generates corrected calorie data based on the basic physiological calorie components and the restricted activity calorie components; Extract the numerical continuity feature of the corrected heat data at the termination boundary, wherein the numerical continuity feature is the difference between the corrected cumulative value at the end of the termination boundary and the cumulative value of the adjacent historical heat record after the termination boundary. Boundary smoothing compensation is performed on the transition section after the termination boundary, and the compensated final data is written into the historical heat record; the mismatch tolerance coefficient is calibrated and updated based on the compensation amplitude of the boundary smoothing compensation; the compensated final data and the updated mismatch tolerance coefficient are sent to the smart wearable device.

2. The method according to claim 1, characterized in that, Before performing ratio calculations, smart wearable devices align inertial motion data and physiological load data at a uniform time granularity, and perform aggregation or resampling on the data within each time granularity to generate an inertial motion intensity sequence and a physiological load intensity sequence under the same time reference. The physical time anchor point is the timestamp corresponding to the modified entry in the historical heat record involved in the retrospective correction.

3. The method according to claim 2, characterized in that, The inertial motion intensity and inertial signal energy are calculated based on data from a triaxial accelerometer. The inertial motion intensity is the value obtained by performing mean or variance calculation on the acceleration magnitude sequence within a sliding window corresponding to the time granularity, and the inertial signal energy is the value obtained by performing energy accumulation on the acceleration magnitude sequence within a sliding window.

4. The method according to claim 2, characterized in that, Physiological load intensity is calculated based on heart rate data extracted from photoplethysmography pulse wave signals. Physiological load intensity is the average heart rate value within a time granularity, or the current metabolic intensity value obtained by mapping the average heart rate value and the user's basal metabolic rate.

5. The method according to claim 1, characterized in that, The initial value of the mismatch tolerance coefficient is derived from the user's historical steady-state profile. The smart wearable device determines the initial value of the mismatch tolerance coefficient based on the coupling dispersion of the inertial motion intensity and physiological load intensity in the uncorrected state. The coupling dispersion is the degree of fluctuation or deviation of the ratio of inertial motion intensity and physiological load intensity at a uniform time granularity. After receiving the updated mismatch tolerance coefficient returned by the terminal device, the historical steady-state profile is iteratively updated.

6. The method according to claim 1, characterized in that, A denominator protection mechanism is introduced when calculating the ratio. A lower limit for physiological intensity is preset. When the physiological load intensity is lower than the lower limit, the lower limit is used instead of the physiological load intensity in the ratio calculation.

7. The method according to claim 1, characterized in that, The determination of the termination boundary includes anti-jitter logic. A termination confirmation window of preset length is established. The endpoint of the termination confirmation window is determined as the termination boundary only when the inertial signal energy in the termination confirmation window shows a decay state and the ratio corresponding to each time granularity falls back to no more than the mismatch tolerance coefficient.

8. The method according to claim 1, characterized in that, The process of constructing and constraining the dynamic upper limit includes: the terminal device multiplies the physiological load intensity by the mismatch tolerance coefficient at each time granularity to obtain the dynamic upper limit, and replaces the inertial motion intensity corresponding to that time granularity with the dynamic upper limit when the inertial motion intensity is greater than the dynamic upper limit.

9. The method according to claim 1, characterized in that, When the terminal device performs boundary smoothing compensation, it calculates the difference between the cumulative value of the corrected heat data at the end of the termination boundary and the cumulative value of the adjacent historical heat records after the termination boundary, and distributes the difference within a preset transition length to form boundary smoothing compensation.

10. The method according to claim 1, characterized in that, When calibrating and updating the mismatch tolerance coefficient, the terminal device adjusts the mismatch tolerance coefficient according to the compensation amplitude of the boundary smoothing compensation, and applies upper and lower limit constraints to the updated mismatch tolerance coefficient.

Images