Long-time static state identification method and wearable device

Abstract

The invention provides a long-standing immobile state recognition method and wearable equipment, and relates to the technical field of wearable equipment, and the long-standing immobile state recognition method comprises the step of changing the collected heart rate value and blood pressure value in the prior art into the collected body surface vibration wave approximate analog quantity. And judging whether the wearing object is in a long-time static state or not by utilizing the body surface vibration of the wearing object. Whether the wearing object is in a long-time static state or not is directly determined by using the modulus threshold value of the body surface vibration of the wearing object, so that the false alarm rate of the alarm information is reduced. And continuous detection is carried out by using the sliding window, so that the continuous monitoring capability of the wearable equipment is improved. The modulus of the recognition complex variable function of the selected to-be-analyzed window is extracted, and whether the early warning information is sent or not is determined based on the modulus threshold value, so that the computing power demand of a processor of the wearable equipment is reduced. The monitoring result is simple to obtain and high in accuracy. The wearable device has the technical effects of being simple in monitoring result acquisition and high in monitoring result accuracy.

Description

Technical Field

[0001] The present invention relates to the technical field of wearable devices, and in particular to a method for identifying a long-term static state and a wearable device. Background Art

[0002] With the development of economic level, people pay more and more attention to the daily life of the elderly at home. Due to the old age of the elderly, their physical functions have deteriorated. Compared with young people, the ordinary life movements of the elderly will appear slow, and the elderly often choose to lie down and other life postures in daily life. It is often difficult to determine the current physiological state of the elderly in a timely and accurate manner through the monitoring system at home. Traditional wearable devices with blood pressure and heart rate measurement functions often cause false alarms of warning information due to the deterioration of the elderly's physical functions. This is because as the elderly age, the heart rate and blood pressure values ​​of the elderly are generally greater than the heart rate and blood pressure values ​​of young people. When the human body is in a dangerous physiological state, the heart rate and blood pressure values ​​are generally greater than the usual heart rate and blood pressure values, so the traditional wearable devices with blood pressure and heart rate measurement functions have a high warning false alarm rate in this application scenario. To this end, it is necessary to propose a long-term static state recognition method and wearable device to address the defect of high warning false alarm rate of traditional wearable devices with blood pressure and heart rate measurement functions. Summary of the invention

[0003] The technical problem to be solved by the present invention is the high false alarm rate of traditional wearable devices with blood pressure and heart rate measurement functions. In order to overcome the defects of the above prior art, the present invention provides a method for identifying a long-term static state and a wearable device.

[0004] The present invention provides a method for identifying a long-term static state, comprising:

[0005] The vibration wave received by the vibration sensor is approximately analog.

[0006] The vibration wave approximate analog quantity is converted into an identifying complex variable function using Fourier transform.

[0007] Receives a time interval to form a sliding window based on the length of the time interval.

[0008] The time domain of the complex variable function is cut using a sliding window, and the sliding step length of the sliding window is half the length of the time interval.

[0009] Get multiple windows to be analyzed.

[0010] Select a window to be analyzed in chronological order.

[0011] The modulus of the identified complex variable function of the selected window to be analyzed is extracted to form at least one modulus to be measured.

[0012] Select a modulus to be measured.

[0013] Based on the modulus threshold, it is determined whether the modulus to be measured is greater than or equal to the modulus threshold.

[0014] If the modulus to be measured is less than the modulus threshold, a warning message is issued, and the process returns to selecting a modulus to be measured until all moduli to be measured are selected.

[0015] If the modulus to be measured is greater than or equal to the modulus threshold, the process returns to the step of selecting a modulus to be measured until all moduli to be measured are selected.

[0016] Return to the above step of selecting a window to be analyzed in chronological order, one by one, until all windows to be analyzed are selected.

[0017] Compared with the prior art, the long-term static state recognition method of the present invention has the following advantages:

[0018] The existing technology of collecting heart rate and blood pressure values ​​is replaced by collecting approximate analog values ​​of body surface vibration waves.

[0019] The wearer's body surface vibration is used to determine whether the wearer has been in a state of long-term inactivity.

[0020] The modulus threshold of the wearer's body surface vibration is used to directly determine whether the wearer is in a long-term static state, thereby reducing the false alarm rate of the alarm information.

[0021] The vibration wave approximation analog quantity received by the vibration sensor can be the vibration intensity received by the vibration sensor, which greatly reduces the performance requirements of the vibration sensor and reduces the manufacturing cost. The vibration wave approximation analog quantity is converted into an identification complex function using Fourier transform, which reduces the computing power requirements of the processor of the wearable device. The continuous detection using sliding windows improves the continuous monitoring capability of the wearable device.

[0022] The modulus of the identification complex function of the selected window to be analyzed is extracted, and based on the modulus threshold, it is determined whether to issue a warning message, which reduces the computing power requirements of the processor of the wearable device. The monitoring results are easy to obtain and the monitoring results are highly accurate.

[0023] In a possible implementation manner, the receiving of the vibration wave approximate analog quantity of the vibration sensor includes:

[0024] Receive a minimum time period.

[0025] A two-dimensional orthogonal coordinate system of vibration amplitude and time is established.

[0026] The time axis that incorporates the shortest time period into a two-dimensional orthogonal coordinate system.

[0027] Receive the characteristic amplitude within the shortest time period.

[0028] The coordinate points formed by the characteristic amplitude and the shortest time period corresponding to the characteristic amplitude are incorporated into a two-dimensional orthogonal coordinate system.

[0029] Using each coordinate point, a curve of the approximate simulation quantity of the vibration wave is formed.

[0030] Compared with the prior art, the above technical solution can reduce the computing power required by the processor, incorporate the characteristic amplitude and the shortest time period corresponding to the characteristic amplitude into a two-dimensional orthogonal coordinate system, is simple to operate, and requires little computing power for the processor.

[0031] At present, the STM32 series of microprocessors can complete this single-threaded data processing task. Simply put, in the clock signal chain of the processor, it receives and saves the vibration intensity at the corresponding time to complete the data acquisition task.

[0032] The data processing of the curve of the vibration wave approximate simulation quantity is simple, which can greatly improve the data distortion rate and the curve distortion rate. The fidelity of the original data is high, which can improve the accuracy of the subsequent data processing results.

[0033] In a possible implementation manner, the receiving of the vibration wave approximate analog quantity of the vibration sensor further includes:

[0034] The minimum time resolution of the vibration sensor is called. The minimum time resolution is the minimum sampling time interval of the vibration sensor.

[0035] The quotient of the shortest time period and the minimum time resolution of the vibration sensor is defined as the sampling frequency.

[0036] The characteristic coordinates are generated using the minimum time resolution and the amplitude corresponding to the minimum time resolution. The amplitude corresponding to the minimum time resolution is a single vibration amplitude collected by the vibration sensor within the minimum sampling time interval.

[0037] Compared with the prior art, the above technical solution can match the sampling rate of the vibration sensor with the sampling rate of the processor clock, and can also adapt the sampling rate of the processor clock by using the minimum time resolution of the vibration sensor.

[0038] This can reduce the intervention rate of the STM32 watchdog program, thereby improving the integrity of data acquisition, thereby improving the fidelity of the original data, and can improve the accuracy of subsequent data processing results.

[0039] In a possible implementation manner, the receiving the characteristic amplitude within the shortest time period includes:

[0040] The linear regression algorithm is:

[0041]

[0042] In formula 1, h a (t a , f a ) is the coordinate point of a sampling frequency in the shortest time period, is the linear regression equation, ε is the first coefficient of the linear regression equation, α is the second coefficient of the linear regression equation, ε Δ for The function solves for the first coefficient of the linear regression equation, α Δ for The function solves for the second coefficient of the linear regression equation.

[0043] Based on the linear regression algorithm, the characteristic amplitude is generated for all sampling frequencies in each shortest time period.

[0044] Use the linear regression equation to solve for the mean coordinate point in the shortest time period.

[0045] Receive the mean coordinate point corresponding to each shortest time period.

[0046] The mean coordinate point is defined as the coordinate point formed by the shortest time period corresponding to the characteristic amplitude.

[0047] Simply put, the amplitude corresponding to the minimum time resolution is the single vibration amplitude collected by the vibration sensor within the minimum sampling time interval.

[0048] In the clock signal chain of the processor, the vibration intensity at the corresponding time is received and saved to complete the data collection task. Therefore, the vibration sensor will obtain a single vibration amplitude in the minimum sampling time interval. The single vibration amplitude is the vibration intensity of the vibration in sequence. The clock signal chain of the processor can contain vibration intensity information.

[0049] In a processor with low-cost computing power, we can rely on a minimum time to form a mean coordinate point corresponding to the shortest time to make early warning judgments, which can greatly reduce the demand for computing power and reduce the use of battery power.

[0050] The coordinates of all sampling frequencies in the shortest time period are converted into a mean coordinate point through a linear regression equation.

[0051] In this working state, a window to be analyzed can have the number of mean coordinate points that coincide with the shortest time period. In this calculation working state, the modulus of the identification complex function of the selected window to be analyzed can still be extracted to form the modulus to be measured. Based on the modulus threshold, it is determined whether the modulus to be measured is greater than or equal to the modulus threshold. Thus, early warning can be achieved more effectively and accurately.

[0052] In fact, in another computing working state, the quotient of the shortest time period and the minimum time resolution of the vibration sensor is defined as the sampling frequency.

[0053] In a possible implementation manner, the method of converting the vibration wave approximate analog quantity into an identification complex variable function by using Fourier transform includes:

[0054] Define a sampling frequency corresponding to a characteristic coordinate.

[0055] H n (T n , F n )→H n (M n ,θ n ) (2)

[0056] In formula 2, H n (T n , F n ) is the coordinate point of the two-dimensional orthogonal coordinate system, T n is a sampling frequency of the time axis of the two-dimensional orthogonal coordinate system, F n is a vibration amplitude of the vibration amplitude axis of the two-dimensional orthogonal coordinate system. n (M n ,θ n ) is the characteristic coordinate of the complex function, M n is the characteristic coordinate H of the complex function coordinate system n The mold length.

[0057]

[0058] In formula 3, θ n is the inclination angle of the complex function coordinate system.

[0059] All characteristic coordinates are transformed into identified complex functions.

[0060]

[0061] θ k ∈{θ1…θ n}

[0062] In formula 4, ||H k (M k ,θ k)|| is the characteristic coordinate H of the complex function coordinate system n The expansion of the module length.

[0063] Simply, the receiving of a time interval of a sliding window to form a sliding window includes:

[0064] Define a minimum time period with multiple minimum sampling time intervals.

[0065] Define a sliding window with multiple minimum time periods.

[0066] The sliding direction of the sliding window is defined as the order of the time domain.

[0067] The sliding step is half the sum of the lengths of all the shortest time periods in the sliding window.

[0068] The sliding window extracts and identifies the data of the complex variable function in sequence along the time axis with the time length of the sliding step.

[0069] It can be understood that the sampling frequency coincides with the number of minimum sampling time intervals.

[0070] In the minimum sampling time interval, the vibration sensor will perform a sampling. Based on the minimum sampling time interval and the time axis, a coordinate point of the minimum sampling time interval and the vibration amplitude can be generated. This coordinate point is a characteristic coordinate in the complex function coordinate system.

[0071] Formula 3 and Formula 4 are transformation equation systems from a two-dimensional orthogonal coordinate system to a complex function coordinate system.

[0072] Using the conversion equation system can greatly reduce the computing power of the processor and avoid the use of the graphics processor.

[0073] In formula 3, when calculating the modulus, the unit length of the time axis is constant, and the use of a sliding window can greatly reduce the computational complexity of the modulus, thereby improving the accuracy of the result.

[0074] In a possible implementation manner, the receiving a time interval of a sliding window to form a sliding window further includes:

[0075] Minimum time resolution for calling the vibration sensor.

[0076] Take the inverse of the minimum time resolution of the vibration sensor.

[0077] The inverse value of the minimum time resolution of the vibration sensor is obtained to generate a time interval value of the sliding window.

[0078] Assign the hour unit in the time unit to the time interval value.

[0079] The length of the time interval to generate the sliding window.

[0080] Specifically, for example, the minimum time resolution of the vibration sensor is 200 times, the inverse value of the minimum time resolution of the vibration sensor is 1 / 200, the time interval value is 1 / 200, and after the time interval value is assigned to the hour unit in the time unit, the time interval is 1 / 200 hour, that is, the length of the time interval of the sliding window is 18 seconds.

[0081] Simply put, the larger the minimum time resolution of the vibration sensor, the higher the sampling frequency of the vibration sensor. The smaller the reciprocal of the minimum time resolution of the vibration sensor, the smaller the length of the time interval of the sliding window after unit conversion.

[0082] The smaller the minimum time resolution of the vibration sensor, the lower the sampling frequency of the vibration sensor. The larger the reciprocal of the minimum time resolution of the vibration sensor, the longer the time interval of the sliding window after unit conversion.

[0083] By adapting the sampling frequency to the length of the time interval of the sliding window, the accuracy of the judgment result can be controlled by utilizing the sampling precision and the time length of the summary information.

[0084] Simply put, no matter how fast or slow the sampling frequency of the vibration sensor is, the judgment result generated by the obtained amplitude data can be maintained within a stable range.

[0085] That is to say, the accuracy of the judgment result of the long-term static state recognition method is less affected by the sampling frequency.

[0086] In a possible implementation, extracting the modulus of the identified complex variable function of the selected window to be analyzed to form the modulus to be measured includes:

[0087] Calling a sample library, wherein the sample library is a sample library of vibration wave approximate analog quantities.

[0088] The modulus threshold is determined based on a minimum modulus of the vibration wave approximation.

[0089] The minimum sampling time interval is used as the extraction time unit for identifying the modulus of the complex function.

[0090] Call the identified complex variable function.

[0091] Determines the modulus of the identified complex variable function within the selected extraction time unit.

[0092] A one-to-one correspondence is formed between the modulus of the identified complex function within each extraction time unit and the selected extraction time unit.

[0093] The modulus of the complex variable function identified within the extraction time unit is included in the modulus to be measured. The modulus threshold is used to compare the modulus of the complex variable function identified within the extraction time unit.

[0094] Specifically, the heart rate will generate vibrations on the body surface. Using the sample library, the modulus of the maximum amplitude of the body surface generated by external factors such as heart rate changes and heart rate can be used as the modulus threshold, and the modulus threshold is the minimum modulus threshold.

[0095] Simply put, the maximum value of the modulus of the body surface amplitude generated by all external factors is taken, and this maximum value is the minimum modulus threshold. When the body surface amplitude exceeds the minimum modulus threshold, it can be determined that the object is not in a long-term static state.

[0096] The minimum modulus threshold can be used to filter the body surface amplitude generated by external factors such as heart rate changes and heart rate. When the surface vibration amplitude of the wearer is less than the minimum modulus threshold, it can be determined that the wearer is in a long-term static state.

[0097] In a possible implementation manner, the returning to the step of selecting a window to be analyzed in order of time until the selection of the window to be analyzed is completed, the method for identifying a long-term static state further includes:

[0098] Call the current window to be analyzed.

[0099] Determine whether the modulus of the complex variable function identified in all extraction time units of the current window to be analyzed has been selected.

[0100] If there is at least one unfinished selection of the modulus of the complex function identified in all extraction time units of the current window to be analyzed, the method based on the modulus threshold is returned to determine whether the modulus to be measured is greater than or equal to the modulus threshold.

[0101] If the modulus of the complex function identified in all extraction time units in the current window to be analyzed is selected, the number of warning messages issued is counted

[0102] The number of warning messages issued is divided by the total number of extraction time units in the current window to be analyzed to form warning data.

[0103] Determine whether the alarm data is greater than or equal to the alarm threshold.

[0104] If the alarm data is less than the alarm threshold, the process returns to the order of time and selects a window to be analyzed in sequence until the selection of the window to be analyzed is completed.

[0105] If the alarm data is greater than or equal to the alarm threshold, the alarm information is fed back to the data center, and the process returns to selecting a window to be analyzed in chronological order until the selection of the window to be analyzed is completed.

[0106] Simply put, for the current window to be analyzed, which has completed all the extraction time units, the modulus judgment of the complex function, and the number of warning messages issued is greater than the alarm threshold, then the alarm information needs to be fed back to the data center.

[0107] Generally, there will be a small period of time when the wearer is inactive. The design redundancy can be used to improve the accuracy of the judgment result.

[0108] The present invention provides a wearable device, comprising:

[0109] The wearable device body is used to execute the long-term static state recognition method.

[0110] A vibration sensor is communicatively connected to the wearable device body.

[0111] Specifically, the wearable device body includes a processor, a memory, a clock, and a communicator. The processor is connected to the memory for communication. The processor is connected to the clock for communication. The processor is connected to the vibration sensor for communication. The communicator can be used as a communication device between the wearable device body and the vibration sensor, and the communicator can also be used as a communication device for the wearable device body to send data to a host computer monitoring center.

[0112] Wearable devices replace the existing technology of collecting heart rate and blood pressure values ​​with vibration sensors that collect approximate analog values ​​of body surface vibration waves.

[0113] The processor uses the body surface vibration of the wearer to determine whether the wearer is in a long-term static state. The vibration sensor is arranged between the body surface of the wearer and the wearable device body to detect the body surface vibration of the wearer.

[0114] The processor uses the modulus threshold of the wearer's body surface vibration to directly determine whether the wearer is in a long-term static state, thereby reducing the false alarm rate of the alarm information.

[0115] The memory receives the approximate analog quantity of the vibration wave of the vibration sensor according to the clock, which can be the vibration intensity received by the vibration sensor, greatly reducing the performance requirements of the vibration sensor and reducing the manufacturing cost. The approximate analog quantity of the vibration wave is converted into an identification complex function using Fourier transform, reducing the computing power requirements of the processor of the wearable device. The continuous detection using the sliding window improves the continuous monitoring capability of the wearable device.

[0116] The processor extracts the modulus of the identification complex function of the selected window to be analyzed, and determines whether to issue a warning message based on the modulus threshold, which reduces the computing power requirements of the processor of the wearable device. The monitoring results are easy to obtain and the monitoring results are highly accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0117] Figure 1 A method flow chart of a method for identifying a long-term static state proposed in this embodiment;

[0118] Figure 2 A two-dimensional orthogonal coordinate system diagram of a long-time static state recognition method proposed in this embodiment;

[0119] Figure 3 A characteristic coordinate system diagram of a complex variable function of a long-time static state recognition method proposed in this embodiment;

[0120] Figure 4 This is a structural connection diagram of a wearable device proposed in this embodiment.

[0121] Description of reference numerals:

[0122] 1. Wearable device body, 100. Processor, 101. Memory, 102. Clock, 103. Communicator, 2. Vibration sensor. DETAILED DESCRIPTION

[0123] First, those skilled in the art should understand that these implementations are only used to explain the technical principles of the embodiments of the present application, and are not intended to limit the protection scope of the embodiments of the present application. Those skilled in the art can make adjustments to them as needed to adapt to specific application scenarios.

[0124] In the description of the embodiments of the present application, it should be noted that, unless otherwise clearly specified and limited, the terms "connected" and "connection" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection. It can be a mechanical connection or an electrical connection. It can be directly connected or indirectly connected through an intermediate medium. For ordinary technicians in this field, the specific meanings of the above terms in the embodiments of the present application can be understood according to specific circumstances.

[0125] In the embodiments of the present application, unless otherwise clearly specified and limited, a first feature being "above" or "below" a second feature may mean that the first and second features are in direct contact, or the first and second features are in indirect contact through an intermediate medium. Moreover, a first feature being "above", "above" or "above" a second feature may mean that the first feature is directly above or obliquely above the second feature, or simply means that the first feature is higher in level than the second feature. A first feature being "below", "below" or "below" a second feature may mean that the first feature is directly below or obliquely below the second feature, or simply means that the first feature is lower in level than the second feature.

[0126] The present application is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0127] See also Figure 1 to Figure 4As shown, the embodiment of the present application discloses a method for identifying a long-term static state, comprising:

[0128] S100, receiving an approximate analog quantity of a vibration wave from a vibration sensor.

[0129] S200, uses Fourier transform to transform the vibration wave approximate analog quantity into an identification complex variable function.

[0130] S300, receiving a time interval of a sliding window to form a sliding window.

[0131] S400, using a sliding window to cut the time domain of the identified complex variable function. The sliding step length of the sliding window is half of the time interval of the sliding window.

[0132] S500: Obtain multiple windows to be analyzed.

[0133] S600: Select a window to be analyzed in chronological order.

[0134] S700, extracting the modulus of the identified complex variable function of the selected window to be analyzed to form a modulus to be measured.

[0135] S800: Based on the modulus threshold, determine whether the modulus to be measured is greater than or equal to the modulus threshold.

[0136] S810, if the modulus to be measured is less than the modulus threshold, a warning message is issued, and the method returns to the modulus threshold-based method to determine whether the modulus to be measured is greater than or equal to the modulus threshold, until the selection of the modulus to be measured is completed.

[0137] S820, if the modulus to be measured is greater than or equal to the modulus threshold, return to the method based on the modulus threshold to determine whether the modulus to be measured is greater than or equal to the modulus threshold, until the selection of the modulus to be measured is completed.

[0138] S900, returning to the step of selecting a window to be analyzed in chronological order, until the selection of the window to be analyzed is completed.

[0139] See also Figure 1 , the vibration wave approximate analog quantity of the receiving vibration sensor of this embodiment includes:

[0140] Receive a minimum time period.

[0141] A two-dimensional orthogonal coordinate system of vibration amplitude and time is established.

[0142] The time axis that incorporates the shortest time period into a two-dimensional orthogonal coordinate system.

[0143] Receive the characteristic amplitude within the shortest time period.

[0144] The coordinate points formed by the characteristic amplitude and the shortest time period corresponding to the characteristic amplitude are incorporated into a two-dimensional orthogonal coordinate system.

[0145] Using each coordinate point, a curve of the approximate simulation quantity of the vibration wave is formed.

[0146] Please continue to see Figure 1 , the receiving vibration wave approximate analog quantity of the vibration sensor also includes:

[0147] The minimum time resolution of the vibration sensor is called. The minimum time resolution is the minimum sampling time interval of the vibration sensor.

[0148] The quotient of the shortest time period and the minimum time resolution of the vibration sensor is defined as the sampling frequency.

[0149] The characteristic coordinates are generated using the minimum time resolution and the amplitude corresponding to the minimum time resolution. The amplitude corresponding to the minimum time resolution is a single vibration amplitude collected by the vibration sensor within the minimum sampling time interval.

[0150] See also Figure 2 to Figure 4 In this embodiment, the characteristic amplitude received within the shortest time period includes:

[0151] The linear regression algorithm is:

[0152]

[0153] In formula 1, h a (t a , f a ) is the coordinate point of a sampling frequency in the shortest time period, is the linear regression equation, ε is the first coefficient of the linear regression equation, α is the second coefficient of the linear regression equation, ε Δ for The function solves for the first coefficient of the linear regression equation, α Δ for The function solves for the second coefficient of the linear regression equation.

[0154] Based on the linear regression algorithm, the characteristic amplitude is generated for all sampling frequencies in each shortest time period.

[0155] Use the linear regression equation to solve for the mean coordinate point in the shortest time period.

[0156] Receive the mean coordinate point corresponding to each shortest time period.

[0157] Define the mean coordinate point as the coordinate point formed by the shortest time period corresponding to the characteristic amplitude. Figure 2In some embodiments, the method of converting the vibration wave approximate analog quantity into an identification complex variable function by using Fourier transform includes:

[0158] Define a sampling frequency corresponding to a characteristic coordinate.

[0159] H n (T n , F n )→H n (M n ,θ n ) (2)

[0160] In formula 2, H n (T n , F n ) is the coordinate point of the two-dimensional orthogonal coordinate system, T n is a sampling frequency of the time axis of the two-dimensional orthogonal coordinate system, F n is a vibration amplitude of the vibration amplitude axis of the two-dimensional orthogonal coordinate system. n (M n ,θ n ) is the characteristic coordinate of the complex function, M n is the characteristic coordinate H of the complex function coordinate system n The mold length.

[0161]

[0162] In formula 3, θ n is the inclination angle of the complex function coordinate system.

[0163] All characteristic coordinates are transformed into identified complex functions.

[0164]

[0165] θ k ∈{θ1…θ n}

[0166] In formula 4, ||H k (M k ,θ k )|| is the characteristic coordinate H of the complex function coordinate system n The expansion of the module length.

[0167] Preferably, the receiving a time interval of a sliding window to form a sliding window includes:

[0168] Define a minimum time period with multiple minimum sampling time intervals.

[0169] Define a sliding window with multiple minimum time periods.

[0170] The sliding step of the sliding window is defined as the length of the shortest total time period in the sliding window in chronological order, which is half of the time period.

[0171] The sliding window extracts and identifies the data of the complex variable function in sequence along the time axis with the time length of the sliding step.

[0172] In other embodiments, the receiving a time interval of a sliding window to form a sliding window further includes:

[0173] Minimum time resolution for calling the vibration sensor.

[0174] Take the inverse of the minimum time resolution of the vibration sensor.

[0175] The inverse value of the minimum time resolution of the vibration sensor is obtained to generate a time interval value of the sliding window.

[0176] Assign the hour unit in the time unit to the time interval value.

[0177] The length of the time interval to generate the sliding window.

[0178] Please continue to see Figure 3 , the extracting the modulus of the identification complex function of the selected window to be analyzed to form the modulus to be measured includes:

[0179] Calling a sample library, wherein the sample library is a sample library of vibration wave approximate analog quantities.

[0180] The modulus threshold is determined based on a minimum modulus of the vibration wave approximation.

[0181] The minimum sampling time interval is used as the extraction time unit for identifying the modulus of the complex function.

[0182] Call the identified complex variable function.

[0183] Determines the modulus of the identified complex variable function within the selected extraction time unit.

[0184] A one-to-one correspondence is formed between the modulus of the identified complex function within each extraction time unit and the selected extraction time unit.

[0185] The modulus of the complex variable function identified within the extraction time unit is included in the modulus to be measured. The modulus threshold is used to compare the modulus of the complex variable function identified within the extraction time unit.

[0186] It is not difficult to understand that the returning to the step of selecting a window to be analyzed in order of time until the selection of the window to be analyzed is completed, the method for identifying a long-time static state further includes:

[0187] Call the current window to be analyzed.

[0188] Determine whether the modulus of the complex variable function identified in all extraction time units of the current window to be analyzed has been selected.

[0189] If at least one modulus of the complex function identified in all extraction time units of the current window to be analyzed is selected, the method based on the modulus threshold is returned to determine whether the modulus to be measured is greater than or equal to the modulus threshold.

[0190] If the modulus of the complex function identified in all extraction time units in the current window to be analyzed is selected, the number of warning messages issued is counted

[0191] The number of warning messages issued is divided by the total number of extraction time units in the current window to be analyzed to form warning data.

[0192] Determine whether the alarm data is greater than or equal to the alarm threshold.

[0193] If the alarm data is less than the alarm threshold, the process returns to the order of time and selects a window to be analyzed in sequence until the selection of the window to be analyzed is completed.

[0194] If the alarm data is greater than or equal to the alarm threshold, the alarm information is fed back to the data center, and the process returns to selecting a window to be analyzed in chronological order until the selection of the window to be analyzed is completed.

[0195] The invention provides a wearable device.

[0196] See also Figure 4 In some embodiments, a wearable device includes:

[0197] The wearable device body 1 is used to execute the long-term static state recognition method.

[0198] The vibration sensor 2 is communicatively connected to the wearable device body 1 .

[0199] Specifically, the wearable device body 1 includes a processor 100, a memory 101, a clock 102, and a communicator 103. The processor 100 is communicatively connected to the memory 101. The processor 100 is communicatively connected to the clock 102. The processor 100 is communicatively connected to the vibration sensor 2. The communicator 103 can be used as a communication device between the wearable device body 1 and the vibration sensor 2, and the communicator 103 can also be used as a communication device for the wearable device body 1 to send data to the upper computer monitoring center.

[0200] Wearable devices replace the existing technology of collecting heart rate and blood pressure values ​​with vibration sensors that collect approximate analog values ​​of body surface vibration waves.

[0201] The processor 100 uses the body surface vibration of the wearer to determine whether the wearer is in a long-term static state.

[0202] The processor 100 directly determines whether the wearer is in a long-term static state by using the modulus threshold of the wearer's body surface vibration, thereby reducing the false alarm rate of the alarm information.

[0203] According to the clock 102, the memory 101 receives the vibration wave approximate analog quantity of the vibration sensor, which can be the vibration intensity received by the vibration sensor, greatly reducing the performance requirements of the vibration sensor and reducing the manufacturing cost. The vibration wave approximate analog quantity is converted into an identification complex function by Fourier transform, reducing the computing power requirement of the processor of the wearable device. The sliding window is used for continuous detection, which improves the continuous monitoring capability of the wearable device.

[0204] The processor 100 extracts the modulus of the identified complex function of the selected window to be analyzed, and determines whether to issue a warning message based on the modulus threshold, which reduces the computing power requirement of the processor of the wearable device. The monitoring result is easy to obtain and the monitoring result is highly accurate.

[0205] In the description of the embodiments of the present application, it should be noted that in the description of the present application, terms such as "inside" and "outside" indicating directions or positional relationships are based on the directions or positional relationships shown in the drawings. This is only for the convenience of description, and does not indicate or imply that the device or component must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it cannot be understood as a limitation on the present application.

[0206] In the description of the present application, the description with reference to the terms "one embodiment", "some embodiments", "in the present embodiment", "specific example", or "some examples" etc. means that the specific features, mechanisms, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, mechanisms, materials or characteristics described may be combined in a suitable manner in any one or more embodiments or examples. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, without contradiction.

[0207] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A method for identifying a long-term static state, characterized in that: include: Receive the approximate analog quantity of the vibration wave from the vibration sensor; The vibration wave approximate analog quantity is transformed into an identification complex variable function by using Fourier transform; receiving a time interval to form a sliding window based on a length of the time interval; The time domain of the complex variable function is cut by using a sliding window; the sliding step length of the sliding window is half the length of the time interval; Obtain multiple windows to be analyzed; Select a window to be analyzed in chronological order; Taking the modulus of the identified complex variable function of the selected window to be analyzed to form at least one modulus to be measured; Select a modulus to be measured; Based on the modulus threshold, determining whether the modulus to be measured is greater than or equal to the modulus threshold; If the modulus to be measured is less than the modulus threshold, a warning message is issued, and the process returns to selecting a modulus to be measured until all moduli to be measured are selected; If the modulus to be measured is greater than or equal to the modulus threshold, return to the step of selecting a modulus to be measured until all moduli to be measured are selected; Return to the above step of selecting a window to be analyzed in chronological order, one by one, until all windows to be analyzed are selected.

2. The method for identifying a long-term static state according to claim 1, characterized in that: The vibration wave approximate analog quantity of the receiving vibration sensor includes: Receive a minimum time period; Establish a two-dimensional orthogonal coordinate system about vibration amplitude and time; Incorporate the shortest time period into the time axis of a two-dimensional orthogonal coordinate system; Receive the characteristic amplitude within the shortest time period; The coordinate points formed by the characteristic amplitude and the shortest time period corresponding to the characteristic amplitude are incorporated into a two-dimensional orthogonal coordinate system; Using each coordinate point, a curve of the approximate simulation quantity of the vibration wave is formed.

3. The method for identifying a long-term static state according to claim 2, characterized in that: The receiving vibration wave approximate analog quantity of the vibration sensor further includes: Calling the minimum time resolution of the vibration sensor; the minimum time resolution is the minimum sampling time interval of the vibration sensor; The quotient of the shortest time period and the minimum time resolution of the vibration sensor is defined as the sampling frequency; The characteristic coordinates are generated using the minimum time resolution and the amplitude corresponding to the minimum time resolution; the amplitude corresponding to the minimum time resolution is a single vibration amplitude collected by the vibration sensor within the minimum sampling time interval.

4. The method for identifying a long-term static state according to claim 3, characterized in that: The receiving characteristic amplitude within the shortest time period includes: The linear regression algorithm is: In formula 1, h a (t a , f a ) is the coordinate point of a sampling frequency in the shortest time period, is the linear regression equation, ε is the first coefficient of the linear regression equation, α is the second coefficient of the linear regression equation, and εΔ is The function solves for the first coefficient of the linear regression equation, α Δ for The function solves for the second coefficient of the linear regression equation; Based on the linear regression algorithm, the characteristic amplitude is generated for all sampling frequencies in each shortest time period; Use the linear regression equation to solve the mean coordinate point in the shortest time period; Receive the mean coordinate point corresponding to each shortest time period; The mean coordinate point is defined as the coordinate point formed by the shortest time period corresponding to the characteristic amplitude.

5. The method for identifying a long-term static state according to claim 4, characterized in that: The method of converting the vibration wave approximate analog quantity into an identification complex variable function by Fourier transform includes: Define a sampling frequency corresponding to a characteristic coordinate; H n (T n ,F n )→H n (M n ,θ n ) (2) In formula 2, H n (T n , F n ) is the coordinate point of the two-dimensional orthogonal coordinate system, T n is a sampling frequency of the time axis of the two-dimensional orthogonal coordinate system, F n is a vibration amplitude of the vibration amplitude axis of the two-dimensional orthogonal coordinate system; H n (M n ,θ n ) is the characteristic coordinate of the complex function, M n is the characteristic coordinate H of the complex function coordinate system n The module length; In formula 3, θ n is the inclination angle of the complex function coordinate system; Perform identification complex function transformation on all characteristic coordinates; In formula 4, ||H k (M k ,θ k )|| is the characteristic coordinate H of the complex function coordinate system n The expansion of the module length.

6. The method for identifying a long-term static state according to claim 5, characterized in that: The receiving a time interval of a sliding window to form a sliding window includes: Define a minimum time period with multiple minimum sampling time intervals; Define a sliding window with multiple minimum time periods; Define the sliding direction of the sliding window as the order of the time domain; The sum of the lengths of the shortest time periods in the sliding window with a sliding step length of one-half; The sliding window extracts and identifies the data of the complex variable function in sequence along the time axis with the time length of the sliding step.

7. The method for identifying a long-term static state according to claim 6, characterized in that: The receiving a time interval of a sliding window to form a sliding window also includes: Minimum time resolution for calling vibration sensors; Take the inverse of the minimum time resolution of the vibration sensor; Obtain the inverse value of the minimum time resolution of the vibration sensor to generate a time interval value of the sliding window; Assign the hour unit in the time unit to the time interval value; The length of the time interval to generate the sliding window.

8. The method for identifying a long-term static state according to claim 7, characterized in that: The step of extracting the modulus of the identification complex variable function of the selected window to be analyzed to form the modulus to be measured includes: Calling a sample library; the sample library is a sample library of vibration wave approximate analog quantities; Determining a modulus threshold based on a minimum modulus of a vibration wave approximation; The minimum sampling time interval is used as the extraction time unit for identifying the modulus of the complex variable function; Call the identification complex variable function; determining the modulus of the identified complex variable function within the selected extraction time unit; A one-to-one correspondence is formed between the modulus of the identified complex variable function in each extraction time unit and the selected extraction time unit; The modulus of the complex variable function identified within the extraction time unit is included in the modulus to be measured; the modulus threshold is used to compare the modulus of the complex variable function identified within the extraction time unit.

9. The method for identifying a long-term static state according to claim 8, characterized in that: The returning to the step of selecting a window to be analyzed in order of time until the selection of the window to be analyzed is completed, the method for identifying a long-time static state further includes: Call the current window to be analyzed; Determine whether the modulus of the complex variable function identified in all extraction time units of the current window to be analyzed has been selected; If there is at least one unfinished selection of the modulus of the complex function identified in all extraction time units of the current window to be analyzed, returning to the method based on the modulus threshold to determine whether the modulus to be measured is greater than or equal to the modulus threshold; If the modulus of the complex function identified in all extraction time units in the current window to be analyzed is selected, the number of warning messages issued is counted The number of issued warning information is divided by the total number of extracted time units in the current window to be analyzed to form warning data; Determine whether the alarm data is greater than or equal to the alarm threshold; If the alarm data is less than the alarm threshold, return to the order of time and select a window to be analyzed in turn until the window to be analyzed is selected; If the alarm data is greater than or equal to the alarm threshold, the alarm information is fed back to the data center, and the process returns to selecting a window to be analyzed in chronological order until the selection of the window to be analyzed is completed.

10. A wearable device, characterized in that: include: A wearable device body, used to execute the method for identifying a long-term static state as claimed in any one of claims 1 to 9; A vibration sensor is communicatively connected to the wearable device body.

Images