PPG-based method for detecting wearability of wrist-worn devices

Abstract

This invention belongs to the field of smart wearable technology, and provides a method for detecting the wearing of wrist-worn wearable devices based on PPG. By activating the wearing detection module, the method detects the three-axis acceleration signal A. X A Y and A Z The mean squared error s AX s AY and s AZ With the three-axis resting mean square error threshold Th sAX ,Th sAY and Th sAZ The infrared light signal P from the PPG module was compared. IRa The mean m IRa , mean squared error IRa and minimum value Min IRa With infrared light upper threshold Th IRu Threshold Th under infrared light IRd Infrared light threshold Th IRd Infrared light stability mean square error threshold Th IRs Compare and obtain the infrared light signal P from the PPG module. IRb The signal P′ is obtained by performing mean filtering. IRb According to signal P′ IRb Frame segmentation and windowing are performed. The mean square error and Fourier amplitude of the signal obtained after frame segmentation and windowing are calculated. The wearing state S is then analyzed based on the calculated mean square error and Fourier amplitude. w Compared to the previous round of wearing detection marker S igw By configuring settings, false or missed detections of not wearing the device can be effectively avoided, thus improving the reliability of the wear detection algorithm for wrist-worn devices.

Description

Technical Field

[0001] This invention relates to the field of smart wearable technology, and in particular to a method, apparatus, and wrist-worn wearable device based on PPG for detecting wear. Background Technology

[0002] Smart wearable devices, such as smartwatches and fitness trackers, can track users' daily activities and sleep patterns to help them monitor their health, manage their bodies, and share their workout experiences, thus sparking a smart wrist revolution. Currently, smart wearable devices generally incorporate various sensors, such as PPG and G-sensors. For PPG sensors, in particular, the presence or absence of a light and the sampling rate significantly impact power consumption. For example, when a certain brand of smart bracelet enters a specific workout mode, the green light and infrared light of the PPG sensor remain on. In this scenario, the bracelet's battery life is less than 3 hours. However, when the continuous heart rate monitoring mode is turned off, the battery life can be extended to up to a week. This demonstrates the significant impact of the PPG light's on-time on power consumption and battery life. Wear detection algorithms detect when the device is being worn; disabling the PPG light when the device is not being worn can greatly reduce ineffective power consumption and significantly improve battery life.

[0003] However, existing wear detection algorithms primarily rely on accelerometer-based activity levels to determine if a device is not being worn. Clearly, these algorithms are prone to false positives or false negatives regarding wear status. For example, if a user is wearing the device but in deep sleep, a false negative will occur, causing the device to disable the PPG light and thus fail to continuously monitor physiological information such as heart rate and HRV during deep sleep. Conversely, if the device is removed and placed in a backpack while walking or engaging in other activities, a false negative will occur because the PPG light is not disabled, rendering the wear detection algorithm ineffective and failing to reduce unnecessary power consumption when the device is not being worn. Alternatively, thresholding the PPG infrared signal and its simple temporal characteristics can be used to detect wear status; however, this approach does not significantly improve false positives or false negatives in scenarios such as wearing the device outdoors in strong light or placing it on its side indoors without wearing it.

[0004] In summary, existing wear detection algorithms for smart wrist-worn devices suffer from low reliability. Summary of the Invention

[0005] To solve the above-mentioned technical problems, the present invention provides the following solution.

[0006] On one hand, the present invention provides a method for detecting the wearing of a wrist-worn wearable device based on PPG, comprising the following steps:

[0007] S100, Activate the wear detection module to check the wear status S w Initialization, check the Sig flag from the previous round of wear detection. w initialization;

[0008] S101. Calculate the triaxial acceleration signal A X A Y and A Z The mean squared error s AX s AY and s AZ With respect to the three-axis resting mean square error threshold Th sAX ,Th sAY and Th sAZ Compare;

[0009] S102, in s AX <Th sAX And s AY <Th sAY And s AZ <Th sAZ If this condition is not met, calculate the infrared light signal P from the PPG module. IRa The mean m IRa , mean squared error IRa and minimum value Min IRa With the upper threshold Th of infrared light IRa Threshold Th under infrared light IRd Infrared light threshold Th IRl Infrared light stability mean square error threshold Th IRs Compare;

[0010] S103, in m IRa >Th IRu or m IIRa <Th IRd or Min IRa <Th IRl or s IRa >Th IRs None of these are valid, and the previous round of wearing detection markers (SIs) was also incorrect. w =1 and s IRa <Th IRs If this condition is not met, acquire the infrared light signal P from the PPG module. IRb The signal P′ is obtained by performing mean filtering. IRb According to signal P′ IRb Frame segmentation and windowing are performed. The mean square error and Fourier amplitude of the signal obtained after frame segmentation and windowing are calculated. The wearing state S is then analyzed based on the calculated mean square error and Fourier amplitude.w Compared to the previous round of wearing detection marker SIg w Configure the settings.

[0011] On one hand, the present invention provides a PPG-based wrist-worn wearable device wearing detection device, comprising:

[0012] Wear detection activation module, used to activate the wear detection module and detect the wearing status S w Initialization, check the Sig flag from the previous round of wear detection. w initialization;

[0013] The triaxial acceleration signal comparison module is used to calculate the triaxial acceleration signal A. X A Y and A Z The mean squared error s AX s AY and s AZ With respect to the three-axis resting mean square error threshold Th sAX Yh saY and Th sAZ Compare;

[0014] Infrared light signal comparison module, used for s AX <Th sax And s AY <Th saY And s AZ <Th sAZ If this condition is not met, calculate the infrared light signal P from the PPG module. IRa The mean m IRa , mean squared error IRa and minimum value Min IRa With the upper threshold Th of infrared light IRu Threshold Th under infrared light IRd Infrared light threshold Th IRl Infrared light stability mean square error threshold Th IRs Compare;

[0015] Wearing settings module, used in m IRa >Th IRu or m IRa <Th IRd or Min IRa <Th IRl or s IRa >Th IRs None of these are valid, and the previous round of wearing the detection marker Sig... w =1 and s IRa <Th IRs If this condition is not met, acquire the infrared light signal P from the PPG module. IRb The signal P′ is obtained by performing mean filtering.IRb According to signal P′ IRb Frame segmentation and windowing are performed. The mean square error and Fourier amplitude of the signal obtained after frame segmentation and windowing are calculated. The wearing state S is then analyzed based on the calculated mean square error and Fourier amplitude. w Compared to the previous round of wearing detection marker Sig w Configure the settings.

[0016] On one hand, the present invention provides a wrist-worn device, comprising: a memory for storing a computer program, a processor, a PPG module communicating with the processor, a three-axis accelerometer module, and a wear detection module, wherein the processor runs the computer program to implement the method described in any of the above-mentioned embodiments.

[0017] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0018] The present invention provides a PPG-based method for detecting the wearing status of a wrist-worn wearable device. This method activates the wearing detection module to detect the wearing status S. w Initialization, check the Sig flag from the previous round of wear detection. w Initialize and calculate the triaxial acceleration signal A. X A Y and A Z The mean squared error s AX s AY and s AZ With respect to the three-axis resting mean square error threshold Th sAX ,Th sAY and Th sAZ Comparison, in s AX <Th sAX And s AY <Th sAY And s AZ <Th sAZ If this condition is not met, calculate the infrared light signal P from the PPG module. IRa The mean m IRa , mean squared error IRa and minimum value Min IRa With the upper threshold Th of infrared light IRu Threshold Th under infrared light IRd Infrared light threshold Th IRl Infrared light stability mean square error threshold Th IRs Comparison, in m IRa >Th IRu or m IRa <Th IRd or Min IRa <Th IRl or s IRa >Th IRsNone of these are valid, and the previous round of wearing the testing marker Slg was also invalid. w =1 and s IRa <Th IRs If this condition is not met, acquire the infrared light signal P from the PPG module. IRb The signal P′ is obtained by performing mean filtering. IRb According to signal P′ IRb Frame segmentation and windowing are performed. The mean square error and Fourier amplitude of the signal obtained after frame segmentation and windowing are calculated. The wearing state S is then analyzed based on the calculated mean square error and Fourier amplitude. w Compared to the previous round of wearing detection marker Slg w By configuring settings, false or missed detections of not wearing the device can be effectively avoided, thus improving the reliability of the wear detection algorithm for wrist-worn devices. Attached Figure Description

[0019] Figure 1 This is a flowchart illustrating a method for detecting the wearing of wrist-worn wearable devices based on PPG.

[0020] Figure 2 This is a flowchart of a wearable device detection system based on PPG.

[0021] Figure 3 This is a schematic diagram of the architecture of a wearable device on the wrist. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, 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.

[0023] The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. It should be understood that in the various embodiments of the invention, the sequence number of each process does not imply a specific order of execution; the order of execution of each process should be determined by its function and inherent logic, and should not constitute any limitation on the implementation process of the embodiments of the invention. It should be understood that in this invention, "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices. It should be understood that in this invention, "a plurality of" means two or more. "And / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, "and / or B" can represent: A alone, A and B simultaneously, or B alone. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "Contains A, B, and C", "Contains A, B, and C" means that all three A, B, and C are contained; "Contains A, B, or C" means that one of A, B, and C is contained; "Contains A, B, and / or C" means that any one, two, or three of A, B, and C are contained. It should be understood that in this invention, "B corresponding to A", "B corresponding to A", "A corresponds to B", or "B corresponds to A" indicates that B is associated with A, and B can be determined based on A. Determining B based on A does not mean determining B solely based on A; B can also be determined based on A and / or other information. Matching A and B is defined as a similarity between A and B that is greater than or equal to a preset threshold. Depending on the context, the word "if" as used herein can be interpreted as "when," "when," "in response to determination," or "in response to detection." The technical solutions of the present invention will be described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments.

[0024] Example 1

[0025] This embodiment provides a method for detecting the wearing of a wrist-worn wearable device based on PPG. The wrist-worn wearable device is equipped with a PPG module, a three-axis accelerometer module, and a wearing detection module. The method for detecting the wearing of a wrist-worn wearable device based on PPG includes steps S100, S101, S102, and S103, as detailed below.

[0026] See Figure 1 A method for detecting the wearing of a wrist-worn wearable device based on PPG includes the following steps:

[0027] S100, Activate the wear detection module to check the wear status S w Initialization, check the Sig flag from the previous round of wear detection. w initialization;

[0028] S101. Calculate the triaxial acceleration signal A X A Y and A Z The mean squared error s AX s AY and s AZ With respect to the three-axis resting mean square error threshold Th sAX ,Th sAY and Th sAZ Compare;

[0029] S102, in s AX <Th sAX And s AY <Th sAY And s AZ <Th sAZ If this condition is not met, calculate the infrared light signal P from the PPG module. IRa The mean m IRa , mean squared error IRa and minimum value Min IRa With the upper threshold Th of infrared light IRu Threshold Th under infrared light IRd Infrared light threshold Th IRl Infrared light stability mean square error threshold Th IRs Compare;

[0030] S103, in m IRA >Th IRu or m IRA <Th IRd or Min IRa <Th IRl or s IRa >Th IRs None of these are valid, and the previous round of wearing the detection marker Sig... w =1 and s IRa <Th IRs If this condition is not met, acquire the infrared light signal P from the PPG module. IRb The signal P′ is obtained by performing mean filtering. IRb According to signal P′ IRbFrame segmentation and windowing are performed. The mean square error and Fourier amplitude of the signal obtained after frame segmentation and windowing are calculated. The wearing state S is then analyzed based on the calculated mean square error and Fourier amplitude. w Compared to the previous round of wearing detection marker Sig w Configure the settings.

[0031] It should be noted that the PPG-based wrist-worn device wearing detection method provided in this embodiment can run on a wrist-worn device. As a type of smart wearable device, the wrist-worn device can be the executing entity for all or part of the steps in the PPG-based wrist-worn device wearing detection method. Besides executing steps S100, S101, S102, and S103 in this embodiment, the wrist-worn device can also run some or all of the steps of the method described below. In this embodiment, by activating the wearing detection module, the wearing state s is detected. w Initialization, check the Sig flag from the previous round of wear detection. w Initialize and calculate the triaxial acceleration signal A. X A Y and A Z The mean squared error s AX s AY and s AZ With respect to the three-axis resting mean square error threshold Th sAX ,Th sAY and Th sAZ Comparison, in s AX <Th sAX And s AY <Th sAY And s AZ <Th sAZ If this condition is not met, calculate the infrared light signal P from the PPG module. IRa The mean m IRa , mean squared error IRa and minimum value Min IRa With the upper threshold Th of infrared light IRs Threshold Th under infrared light IRd Infrared light threshold Th IRl Infrared light stability mean square error threshold Th IRs Comparison, in m IRa >Th IRu or m IRa <Th IRd or Min IRa <Th IRl or s IRa >Th IRs None of these are valid, and the previous round of wearing the detection marker Sig... w =1 and s IRa <Th IRsIf this condition is not met, acquire the infrared light signal P from the PPG module. IRb The signal P′ is obtained by performing mean filtering. IRb According to signal P′ IRb Frame segmentation and windowing are performed. The mean square error and Fourier amplitude of the signal obtained after frame segmentation and windowing are calculated. The wearing state S is then analyzed based on the calculated mean square error and Fourier amplitude. w Compared to the previous round of wearing detection marker Slg w By configuring settings, false or missed detections of not wearing the device can be effectively avoided, thus improving the reliability of the wear detection algorithm for wrist-worn devices.

[0032] In step S100, the wearing detection module is activated to detect the wearing status S. w Initialization, check the Sig flag from the previous round of wear detection. w Initialization. In some embodiments, step S100 includes step one: setting a wear detection module on the wrist-worn wearable device, activating the wear detection module, and checking the wear state S. w Initialization, check the Sig flag from the previous round of wear detection. w Initialization. Further, step one includes the following steps: turning on the wrist-worn device to activate the wear detection module; setting the wear state S... w Initialize to 1, and set the previous round of wear detection flag Sig w Initialize to 0; when the wearing detection module detects wearing state S w When the value is 0, it indicates that the current detection result of the wearing detection module regarding the wrist-worn device is an unworn state; when the wearing detection module detects a wearing state S... w When the value is 1, it means that the current detection result of the wearing detection module regarding the wrist-worn device is that it is in a worn state.

[0033] In step S101, the wrist-worn device calculates the triaxial acceleration signal A. X A Y and A Z The mean squared error s AX s AY and s AZ With respect to the three-axis resting mean square error threshold Th sAX ,Th sAY and Th sAZ Comparison. In some embodiments, step S101 includes: Step two, acquiring triaxial acceleration signal A through the triaxial accelerometer module. X A Y and A Z The signal A′ is obtained by performing mean filtering on each signal. x A′ Y and A′ ZStep 3: Process the signal A′ respectively. x A′ Y and A′ Z Calculate its standard deviation and obtain the standard deviation s. AX s AY and s AZ Step 4: Set the three-axis resting mean square error threshold Th sAX ,Th sAY and Th sAZ If s AX <Th sAX And s AY <Th sAY And s AZ <Th sAZ If the condition is met, proceed to step two; otherwise, proceed to step S102. Further, step two includes the following steps: setting the sampling rate of the triaxial accelerometer module to f... A The sampling time period is T; the triaxial accelerometer module uses a sampling rate f. A Acquire triaxial acceleration signal A within time period T X A Y and A Z ; respectively for the triaxial acceleration signal A X A Y and A Z Signal A′ is obtained by performing mean filtering. x A′ Y and A′ z ;in,

[0034]

[0035] In step S102, the wrist-worn device is in s AX <Th sAX And s AY <Th sAY And s AZ <Th sAZ If this condition is not met, calculate the infrared light signal P from the PPG module. IRa The mean m IRa , mean squared error S IRa and minimum value Min IRa With the upper threshold Th of infrared light IRu Threshold Th under infrared light IRd Infrared light threshold Th IRl Infrared light stability mean square error threshold Th IRs Comparison. In some embodiments, step S102 includes the following steps: Step 5: Acquire infrared light signal P through the PPG module. IRa And perform mean filtering to obtain signal P′ IRaStep 6: For signal P′ IRa Calculate its mean m IRa , mean squared error IRa and its minimum value Min IRa ;in,

[0036]

[0037]

[0038] Step 7: Set the infrared light threshold Th IRu Threshold Th under infrared light IRd Infrared light threshold Th IRl Infrared light stability mean square error threshold Th IRs If TR IRa >Th IRu or m IRa <Th IRd or Min IRa <Th IRl or s IRa >Th IRs Then first set the wearing state S w Assign the value to the previous round of wear detection flag Sig w Then the wearing state S w Set to 0 to disable the PPG module and proceed to step two; otherwise, if the previous round of wear detection flag Sig... w =1 and s IRa <Th IRs Then the wearing state S w If the value remains 1, proceed to step two; otherwise, proceed to step S103; where Th IRl <Th IRd <Th IRu Furthermore, step five includes the following steps: setting the sampling rate of the PPG module to f. P0 The sampling time period is T; the PPG module is enabled and the sampling rate is f. P5 Acquire infrared light signal P within time period T IRa And perform mean filtering to obtain signal P′ IRa .

[0039] In step S103, the wrist-worn device is in m IRa >Th IRu or m IRa <Th IRd or Min IRa <Th IRl or s IRa >Th IRs None of these are valid, and the previous round of wearing the detection marker Sig...w =1 and s IRa <Th IRs If this condition is not met, acquire the infrared light signal P from the PPG module. IRb The signal P′ is obtained by performing mean filtering. IRb According to signal P′ IRb Frame segmentation and windowing are performed. The mean square error and Fourier amplitude of the signal obtained after frame segmentation and windowing are calculated. The wearing state S is then analyzed based on the calculated mean square error and Fourier amplitude. w Compared to the previous round of wearing detection marker Sig w Configure the settings. In some embodiments, step S103 includes the following steps:

[0040] Step 8, using the PPG module at a sampling rate f P25 Acquire infrared light signal P within time period T IRb And perform mean filtering to obtain signal P′ IRb ;

[0041] in,

[0042] Step 9, set the frame length F plx Frame shift F pmx For the signal P′ IRb Perform frame segmentation and windowing processing to obtain and respectively the signals Calculate its standard deviation to obtain the standard deviation.

[0043] Among them, F pnmx <F plx ,

[0044]

[0045]

[0046] Step 10, set the frame length F ply Frame shift F pmy For the signal P′ IRb Perform frame segmentation and windowing processing to obtain and respectively the signals Calculate its standard deviation to obtain the standard deviation. Among them, F pmx <F plx And F plx <F ply And F pmy <Fply ,

[0047] Step 11: Set the upper threshold Th for the mean squared error. su Threshold Th under mean square error sd ,statistics The value greater than the mean square error threshold Th su The number N sux ,statistics The value of Th is less than the threshold value of the mean square error. sd The number N sdx ,statistics The value of Th is less than the threshold value of the mean square error. sd The number N sdy ;

[0048] Step 12, set the percentage threshold Th Nsux ,Th Nsdx and Th Nsdy like or or First, the wearing state Sw is assigned to the previous wearing detection flag Sig. w Then the wearing state S w If set to 0, the PPG module is disabled, and the process proceeds to step two; otherwise, the process proceeds to step thirteen; where 0 < Th Nsux ,Th Nsdx ,Th Nsdy <1;

[0049] Step 13, Set frame length F plz Frame shift F pmz For the signal P′ IRb Perform frame segmentation and windowing processing to obtain and respectively the signals Perform a Fourier transform to obtain the Fourier amplitude. Among them, F pmz <F plz And log2(F plz ) is a positive integer greater than 4.

[0050] Step fourteen, respectively for Obtaining valid elements in,

[0051] Step 15, obtain respectively The index number corresponding to the maximum value

[0052] Step 16, Set the subscript threshold Th Nf ,statistics Less than the subscript threshold Th Nf The number N Nf ;

[0053] Step 17, Set the percentage threshold Th NNf ,like Then the wearing state S w Set to 0, and set the previous round of wear detection flag Sig w If set to 0, the PPG module is turned off, and the process proceeds to step two; otherwise, the process proceeds to step eighteen.

[0054] Step 18, first set the wearing state S w Assign the value to the previous round of wear detection flag Sig w Then the wearing state S w Set it to 1 and proceed to step two.

[0055] It should also be noted that, in the above embodiments, the PPG-based wrist-worn wearable device wearing detection method acquires infrared light signals through the PPG module set on the wrist-worn wearable device, and accurately detects the device wearing status based on threshold judgment and decision tree classification. It effectively disables the PPG light when the device is not worn, greatly reducing the device's ineffective power consumption and thus significantly improving the device's battery life. Furthermore, the PPG-based wrist-worn wearable device wearing detection method is applicable to all-weather real-time detection of wrist-worn wearable devices in any working or non-working scenario. It uses only the time-frequency domain features extracted from the infrared light signal to achieve device wearing detection, exhibiting high algorithm compatibility and applicability to any working or non-working mode of the device. It is completely decoupled from other functional modules of the device (such as heart rate, blood oxygen, and exercise), demonstrating strong anti-interference capabilities and high reliability. Based on threshold judgment and decision tree classification, the algorithm significantly reduces computational complexity and space requirements while ensuring high accuracy, requiring low hardware configuration and making it suitable for various embedded systems.

[0056] Example 2

[0057] See Figure 2 This embodiment provides a PPG-based wrist-worn wearable device wearing detection device, comprising:

[0058] The wear detection activation module is used to activate the wear detection module, initialize the wear state Sw, and set the wear detection flag Sig from the previous round. winitialization;

[0059] The triaxial acceleration signal comparison module is used to calculate the triaxial acceleration signal A. X A Y and A Z The mean squared error s AX s AY and s AZ With respect to the three-axis resting mean square error threshold Th sAX ,Th sAY and Th sAZ Compare;

[0060] Infrared light signal comparison module, used for s AX <Th sAX And s AY <Th sAY And s AZ <Th sAZ If this condition is not met, calculate the infrared light signal P from the PPG module. IRa The mean m IRa , mean squared error IRa and minimum value Min IRa With the upper threshold Th of infrared light IRu Threshold Th under infrared light IRd Infrared light threshold Th IRl Infrared light stability mean square error threshold Th IRs Compare;

[0061] Wearing settings module, used in m IRa >Th IRu or m IRa <Th IRd or Min IRa <Th IRa or s IRa >Th IRs None of these are valid, and the previous round of wearing the detection marker Sig... w =1 and s IRa <Th IRs If this condition is not met, acquire the infrared light signal P from the PPG module. IRb The signal P′ is obtained by performing mean filtering. IRb According to signal P′ IRb Frame segmentation and windowing are performed. The mean square error and Fourier amplitude of the signal obtained after frame segmentation and windowing are calculated. The wearing state S is then analyzed based on the calculated mean square error and Fourier amplitude. w Compared to the previous round of wearing detection marker Slg w Configure the settings.

[0062] In this embodiment, by activating the wearing detection module, the wearing state S is monitored. wInitialization, check the Sig flag from the previous round of wear detection. w Initialize and calculate the triaxial acceleration signal A. X A Y and A Z The mean squared error s AX s AY and s AZ With respect to the three-axis resting mean square error threshold Th sΛX ,Th sAr and Th sAZ Comparison, in s AX <Th sAX And s AY <Th sAY And s AZ <Th sAZ If this condition is not met, calculate the infrared light signal P from the PPG module. IRa The mean m IRa , mean squared error IRa and minimum value Min IRa With the upper threshold Th of infrared light IRu Threshold Th under infrared light IRd Infrared light threshold Th IRl Infrared light stability mean square error threshold Th IRs Comparison, in m IRa >Th IRu or m IRa <Th IRd or Min IRa <Th IRl or s IRa >Th IRs None of these are valid, and the previous round of wearing the testing marker Slg was also invalid. w =1 and s IRa <Th IRs If this condition is not met, acquire the infrared light signal P from the PPG module. IRb The signal P′ is obtained by performing mean filtering. IRb According to signal P′ IRb Frame segmentation and windowing are performed. The mean square error and Fourier amplitude of the signal obtained after frame segmentation and windowing are calculated. The wearing state S is then analyzed based on the calculated mean square error and Fourier amplitude. w Compared to the previous round of wearing detection marker Sig w By configuring settings, false or missed detections of not wearing the device can be effectively avoided, thus improving the reliability of the wear detection algorithm for wrist-worn devices.

[0063] Example 3

[0064] See Figure 3This embodiment provides a wrist-worn wearable device, including: a memory storing a computer program, a processor, a PPG module communicating with the processor, a three-axis accelerometer module, and a wear detection module. The processor is used to execute the computer program stored in the memory to implement the various steps of the above method.

[0065] The memory can also be flash memory. Memory can be standalone or integrated with the processor. When the memory is a device independent of the processor, the device may also include a bus for connecting the memory and the processor. Computer programs are, for example, application programs or functional modules that implement the methods described above.

[0066] In this embodiment, by activating the wearing detection module, the wearing state S is monitored. w Initialization, check the Sig flag from the previous round of wear detection. w Initialize and calculate the triaxial acceleration signal A. X A Y and A Z The mean squared error s AX s AY and s AZ With respect to the three-axis resting mean square error threshold Th sAX ,Th sAY and Th sAZ Comparison, in s AX <Th sAX And s AY <Th sAY And s AZ <Th sAZ If this condition is not met, calculate the infrared light signal P from the PPG module. IRa The mean m IRa , mean squared error IRa and minimum value Min IRa With the upper threshold Th of infrared light IRu Threshold Th under infrared light IRd Infrared light threshold Th IRl Infrared light stability mean square error threshold Th IRs Comparison, in m IRa >Th IRu or m IRu <Th IRd or Min IRa <Th IRl or s IRa >Th IRs None of these are valid, and the previous round of wearing the testing marker Slg was also invalid. w =1 and s IRa <Ih IRs If this condition is not met, acquire the infrared light signal P from the PPG module. IRbThe signal P′ is obtained by performing mean filtering. IRb According to signal P′ IRb Frame segmentation and windowing are performed. The mean square error and Fourier amplitude of the signal obtained after frame segmentation and windowing are calculated. The wearing state S is then analyzed based on the calculated mean square error and Fourier amplitude. w Compared to the previous round of wearing detection marker Slg w By configuring settings, false or missed detections of not wearing the device can be effectively avoided, thus improving the reliability of the wear detection algorithm for wrist-worn devices.

[0067] Example 4

[0068] This embodiment also provides a readable storage medium storing a computer program, which, when executed by a processor, is used to implement the methods provided in the various embodiments described above.

[0069] In this embodiment, when the computer program is executed by the processor to implement the methods provided in the various embodiments described above, it activates the wearing detection module to detect the wearing state S. w Initialization, check the Sig flag from the previous round of wear detection. w Initialize and calculate the triaxial acceleration signal A. X A Y and A Z The mean squared error s AX s AY and s AZ With respect to the three-axis resting mean square error threshold Th sAX ,Th sAY and Th sAZ Comparison, in s AX <Th sAX And s AY <Th sAY And s AZ <Th sAZ If this condition is not met, calculate the infrared light signal P from the PPG module. IRa The mean m IRa , mean squared error IRa and minimum value Min IRa With the upper threshold Th of infrared light IRu Threshold Th under infrared light IRd Infrared light threshold Th IRl Infrared light stability mean square error threshold Th IRs Comparison, in m IRa >Th IRu or m IRa <Th IRd or Min IRa <Th IRl or s IRa >ThIRs None of these are valid, and the previous round of wearing the detection marker Sig... w =1 and s IRa <Th IRs If this condition is not met, acquire the infrared light signal P from the PPG module. IRb The signal P′ is obtained by performing mean filtering. IRb According to signal P′ IRb Frame segmentation and windowing are performed. The mean square error and Fourier amplitude of the signal obtained after frame segmentation and windowing are calculated. The wearing state S is then analyzed based on the calculated mean square error and Fourier amplitude. w Compared to the previous round of wearing detection marker Sig w By configuring settings, false or missed detections of not wearing the device can be effectively avoided, thus improving the reliability of the wear detection algorithm for wrist-worn devices.

[0070] It should be noted that a readable storage medium can be a computer storage medium or a communication medium. A communication medium includes any medium that facilitates the transfer of computer programs from one location to another. A computer storage medium can be any available medium accessible to a general-purpose or special-purpose computer. For example, a readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, a readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside in an Application Specific Integrated Circuit (ASIC). Alternatively, the ASIC can be located in a user equipment. Of course, the processor and the readable storage medium can also exist as discrete components in a communication device. A readable storage medium can be a read-only memory (ROM), random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage device, etc. In the embodiments of the above-described device, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly manifested as execution by a hardware processor, or execution by a combination of hardware and software modules within the processor.

[0071] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A PPG-based wrist-worn device wearing detection method, characterized in that, comprising the steps of: S100, activate the wearing detection module, and determine the wearing state S w initialized to 1, and the wearing detection flag Sig of the last round is w initialized to 0; when the wearing detection module detects a wearing state S w is 0, it represents that the current detection result of the wearing detection module about the wrist-worn device is the unwearing state; when the wearing detection module detects a wearing state S w is 1, it represents that the current detection result of the wearing detection module about the wrist-worn device is the wearing state; S101. Calculate the triaxial acceleration signal A X A Y and A Z The mean squared error s AX s AY and s AZ With respect to the three-axis resting mean square error threshold Th sAX ,Th sAY and Th sAZ Compare; S102、In S AX Th sAX and s AY Th sAY and s AZ Th sAZ If not, calculate the mean m IRa , mean square deviation s IRa , and minimum value Min IRa of the infrared light signal P IRa of the PPG module, and compare them with the infrared light upper threshold value Th IRu , the infrared light lower threshold value Th IRd , the infrared light too low threshold value Th IRl , and the infrared light stable mean square deviation threshold value Th IRs ; S103, in m IRa > Th IRu or m IRa < Th IRd or Min IRa < Th IRl or s IRa > Th IRs are not all true, and when the last round wearing detection flag Sig w is 1 and s IRa < Th IRs is not true, the infrared light signal P IRb of the PPG module is acquired, the signal P′ IRb is obtained by performing mean filtering processing, the signal P′ IRb is processed by frame division and windowing, the mean square deviation and the Fourier amplitude of the signal obtained after the frame division and windowing processing are calculated, and the wearing state S w and the last round wearing detection flag Sig w are set according to the calculated mean square deviation and Fourier amplitude.

2. The method of claim 1, wherein, Step S100 comprises: Step one, wearing the device on the wrist sets wearing detection module, activates the wearing detection module, and detects the wearing state S w Initialization, the wearing detection flag Sig of the last round w Initialization.

3. The method of claim 2, wherein, Step one comprises the steps of: Turning on the wrist-worn device to activate the wear detection module.

4. The method of claim 1, wherein, Step S101 comprises: Step two, collect three-axis acceleration signal A through three-axis accelerometer module X 、 Y and A Z respectively, and carry out mean filtering processing to obtain signal A' X 、 A' Y and A' Z ; Step three, calculate the mean square error of the signals A' X , A' Y and A' Z respectively, and obtain the mean square errors s AX , s AY and s AZ ; Step four, set three-axis rest mean square deviation threshold Th sAX , Th sAY , and Th sAZ If s AX < Th sAX and s AY < Th sAY and s AZ < Th sAZ , jump to step two; otherwise, jump to step S102.

5. The method of claim 4, wherein, Step two comprises the steps of: The sampling rate of the three-axis accelerometer module is set to f A , and the time period of sampling is T. by the triaxial accelerometer module at a sampling rate f A obtaining a triaxial acceleration signal A X Y and A Z ;​ The triaxial acceleration signals A X , A Y and A Z are subjected to mean filtering respectively to obtain signals A' X , A' Y and A' Z ; wherein, 6. The method of claim 5, wherein, Step S102 comprises the steps of: Step five: collect infrared light signal P by PPG module IRa and carry out mean filtering to obtain signal P' IRa ; Step six: the signal P' IRa calculating its mean m IRa , the mean square deviation s IRa and its minimum Min IRa ; wherein, f P5 fs is the sampling rate of the PPG module; Step seven: set the upper threshold Th of infrared light IRu , the lower threshold Th of infrared light IRd , the lower threshold Th of infrared light IRl , the stable mean square error threshold Th of infrared light IRs , if m IRa > Th IRu or m IRa < Th IRd or Min IRa < Th IRl or s IRa > Th IRs , the wearing state S w is assigned to the last round wearing detection flag Sig w , and the wearing state S w is set to 0, the PPG module is closed, and the step two is jumped to; otherwise, if the last round wearing detection flag Sig w is 1 and s IRa < Th IRs , the wearing state S w is kept as 1, and the step two is jumped to; otherwise, the step S103 is jumped to; wherein Th IRI < Th IRd < Th IRu .

7. The method of claim 6, wherein, Step five comprises the steps of: Setting the time period of sampling of the PPG module as T; Turning on the PPG module and sampling at a sampling rate f P5 Obtaining the infrared light signal P within a time period T IRa and performing mean filtering to obtain a signal P' IRa .

8. The method of claim 7, wherein, Step S103 comprises the steps of: Step eight, sampling rate f P25 Obtaining infrared light signal P in time period T IRb And performing mean filtering to obtain signal P' IRb ; wherein Step nine, set frame length F plx , frame shift F pmx , frame the signal P' IRb , and perform windowing processing, to obtain , and calculate the mean square error of the signal , respectively, to obtain the mean square error wherein F pmx <F plx , Step ten, set frame length F ply , frame shift F pmy , frame the signal P' IRb and window processing, get and the signal Calculate the mean square error, get mean square error Where, F pmx <F pIx and F plx <F ply and F pmy <F ply , Step eleven, set the upper threshold of mean square deviation Th su and the lower threshold of mean square deviation Th sd , count the number N sux of the data in the data set greater than the upper threshold of mean square deviation Th su , count the number N sdx of the data in the data set less than the lower threshold of mean square deviation Th sd , count the number N sdy of the data in the data set less than the lower threshold of mean square deviation Th sd ; Step twelve, set the proportion threshold Th Nsux , Th Nsdx , Th Nsdy , if (N sux / [T·f P25 / (F plx -F pmx ))<Th Nsux or (Th Nsdx / [T·f P25 / (F plx -F pmx ))>Th Nsdx or (N sdy / [T·f P25 / (F plx -F pmx ))>Th Nsdy , first assign the wearing state S w to the last round wearing detection flag Sig w , then set the wearing state S w to 0, close the PPG module, and jump to step two; otherwise, jump to step thirteen; wherein 0 Th Nsux ,Th Nsdx ,Th Nsdy <1; Step three, setting frame length F plz , frame shift F pmz , performing frame division and windowing on the signal P' IRb , obtaining , and performing Fourier transform on the signals , respectively, to obtain Fourier amplitude wherein F pmz <F plz , and log2(F plz ) is a positive integer greater than 4; Step fourteen, respectively to Retained active elements obtained Step fifteen, respectively acquiring the subscript sequence number corresponding to the maximum value of Step sixteen, setting a subscript threshold Th Nf , statistics The number N Nf of small subscript values less than the subscript threshold Th Nf ; Step seventeen, set the proportion threshold Th NNf , if (N Nf / [T·f P25 / (F plz -F pmz )) < Th NNf , set the wearing state S w to 0, set the last round wearing detection flag Sig w to 0, close the PPG module, and jump to step two; otherwise, jump to step eighteen; Step eight, the wearing state S w is assigned to the last round wearing detection flag bit Sig w , then the wearing state S w is set to 1, and jump to step two. 9.A PPG-based wrist-worn device wearing detection apparatus, characterized by comprising: The wearing detection activation module is configured to activate the wearing detection module, and the wearing state S w is initialized to 1, and the last round wearing detection flag bit Sig w is initialized to 0. when the wearing detection module detects a wearing state S w is 0, it represents that the current detection result of the wearing detection module about the wrist-worn device is the unwearing state; when the wearing detection module detects a wearing state S w is 1, it represents that the current detection result of the wearing detection module about the wrist-worn device is the wearing state; a triaxial acceleration signal comparison module for calculating the mean square differences s X , A Y , and s Z of the triaxial acceleration signals A AX , s AY , and s AZ for comparison with triaxial rest mean square difference thresholds Th sAX , Th sAY , and Th sAZ ; Infrared light signal comparison module, used for s AX <Th sAX And s AY <Th sAY And s AZ <Th sAZ If this condition is not met, calculate the infrared light signal P from the PPG module. IRa The mean m IRa , mean squared error IRa and minimum value Min IRa With the upper threshold Th of infrared light IRu Threshold Th under infrared light IRd Infrared light threshold Th IRl Infrared light stability mean square error threshold Th IRs Compare; Wearing setting module, used for setting m IRa Th IRu or m IRa Th IRd or Min IRa Th IRl or s IRa Th IRs All are not established, and when the last round wearing detection flag Sig w is 1 and s IRa Th IRs is not established, obtaining the infrared light signal P IRb of PPG module, performing mean filtering processing to obtain signal P' IRb , performing frame division and windowing processing according to signal P' IRb , calculating mean square deviation and Fourier amplitude of the signal obtained after frame division and windowing processing, and setting wearing state S w and last round wearing detection flag Sig w according to the calculated mean square deviation and Fourier amplitude.

10. A wrist-worn device, comprising: comprising: a memory storing a computer program, a processor, a PPG module, a three-axis accelerometer module and a wear detection module in communication with the processor, the processor running the computer program to implement the method according to any one of claims 1-8.

Images