A sleep onset time correction method for sleep monitoring

Abstract

This invention belongs to the field of sleep monitoring technology and discloses a method for correcting the sleep onset time for sleep monitoring, including: obtaining a calculated value S of the sleep onset time for the current test. M ;Calculate values ​​S1 to S based on the sleep onset time of the previous M-1 tests. M‑1 For S M The first correction is performed, resulting in the sleep time correction value S. M Output values ​​S'1 to S' based on the sleep time corrected in the previous M-1 iterations. M‑1 To S” M After a second correction, the output value S' for the current sleep onset time is obtained. M The correction process is completed. The sleep onset time correction method of this invention is applied to sleep monitoring devices, where the sleep monitoring device directly calculates and determines the calculated sleep onset time value S based on the detection data. M Subsequently, the sleep onset time value S can be calculated based on historical data recorded by the sleep monitoring device. M Two corrections are performed to eliminate detection bias in the equipment to some extent and improve the output value S' of the sleep onset time. M The accuracy.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of sleep monitoring, and in particular relates to a sleep-in time correction method for sleep monitoring. BACKGROUND

[0002] Sleep quality is an important factor to ensure physical health, and with the improvement of people's living standards, the importance of sleep quality is also increasing. In order to monitor the sleep quality, with the development of science and technology, the existing technology has appeared the technical scheme of measuring various physiological indexes of human body in sleep by instrument, and then realizing sleep quality evaluation. Specifically, the current sleep monitoring methods mainly include polysomnography, electroencephalogram double frequency index, sleep state video monitoring, blood oxygen saturation sleep monitoring, temperature change recorder, activity recorder, etc.

[0003] The evaluation of sleep quality needs to collect various data by sleep monitoring equipment during the user's sleep process for detection, and one of the important detections is to distinguish the sleep state of the user, so as to determine the sleep-in time, wake-up time, sleep stage, sleep-in duration, bed non-sleep duration, and total sleep duration of the user in this sleep process. Among them, the accurate detection of sleep-in time and wake-up time will directly affect the accuracy of the judgment of the total sleep duration of the user by the sleep monitoring equipment.

[0004] However, the existing sleep monitoring equipment inevitably has detection errors. For example, one of the existing technical solutions is to judge the sleep-in time by detecting the reduction of body movement frequency, and if the user lies in bed for a long time without obvious movement, the sleep monitoring equipment may determine the current time node as the sleep-in time even if the user has not actually fallen asleep. The error generated in such a situation is difficult to eliminate by improving the accuracy of equipment detection.

[0005] Therefore, on the basis of the existing sleep monitoring technology, how to correct the sleep-in time detected by the sleep monitoring equipment so that the corrected sleep-in time is closer to the actual sleep-in time of the user has become a problem to be solved.

[0006] Therefore, the present application is proposed. SUMMARY

[0007] The technical problem to be solved by the present application is to overcome the shortcomings of the prior art, and to provide a sleep-in time correction method for sleep monitoring, which can correct the sleep-in time calculation value directly calculated according to the detection data twice according to the historical data of the user's sleep, thereby improving the accuracy of the sleep-in time judgment.

[0008] To solve the above technical problems, the basic idea of the technical solution of the present application is:

[0009] A method for correcting sleep onset time in sleep monitoring includes:

[0010] Obtain the calculated value S of the time of falling asleep during this test. M ;

[0011] The values ​​S1 to S2 are calculated based on the sleep onset time from the previous M-1 tests. M-1 For S M The first correction is performed, resulting in the sleep time correction value S. M ;

[0012] Based on the sleep time correction completed in the previous M-1 iterations, output values ​​S'1 to S' M-1 To S” M A second correction is performed to obtain the output value S' for the current sleep onset time. M The correction process is now complete.

[0013] Furthermore, the first revision includes:

[0014] According to ΔS i =S M -S i Calculate the value S of the time when you fall asleep this time. M The calculated sleep time S of the i-th time among the calculated sleep time values ​​from the previous M-1 tests. i The difference ΔS between i Where i is any integer in the interval [1, M-1];

[0015] Based on the calculated M-1 differences ΔS i For S M The correction is performed to obtain the sleep time correction value S”. M .

[0016] Furthermore, according to ΔS i For S M The corrections include:

[0017] Calculate the M-1 differences ΔS i Their respective absolute values ​​|ΔS i | is used for comparison;

[0018] Determine the minimum value |ΔS i | min and the minimum value |ΔS i | min The corresponding sleep onset time calculation value S i,min ;

[0019] According to |ΔS i | min and / or S i,min For S MThe sleep time correction value S" is obtained by correcting the sleep time calculation value S M .

[0020] Further, a first time threshold T1 is preset.

[0021] If |ΔS i | min ≥ T1, the sleep time correction value S" M is obtained by adding or subtracting an offset v1 to the sleep time calculation value S M ; wherein the offset v1 satisfies: 0 < v1 ≤ T1.

[0022] If |ΔS i | min < T1, the sleep time correction value S" M retains the value of the sleep time calculation value S M .

[0023] Further, when |ΔS i | min ≥ T1, if S M > S i,min , the sleep time correction value S" M = S M - v1; if S M < S i,min , the sleep time correction value S" M = S M + v1.

[0024] Further, if S1 to S M-1 are not completely equal, and the M-1 absolute values |ΔS i | obtained by calculation are all equal, the sleep time correction value S" M retains the value of the sleep time calculation value S M .

[0025] And / or, when there are x minimum values |ΔS i | and y values of equal |ΔS i | min in the M-1 absolute values |ΔS i | eq , if x < y, the sleep time correction value S" M retains the value of the sleep time calculation value S M ; wherein |ΔS i | eq > |ΔS i | min , and x + y ≤ M-1, y > 1.

[0026] Further, the second correction comprises:

[0027] According to ΔS' i =S” M -S' i Calculate the sleep onset time correction value S” for this time. M The sleep time output value S' of the i-th sleep time output value among the previous M-1 corrected sleep time output values. i The difference between ΔS' i Where i is any integer in the interval [1, M-1];

[0028] Based on the calculated M-1 differences ΔS' i To S” M The correction yields the output value S' for the time of falling asleep. M .

[0029] Furthermore, according to ΔS' i To S” M The corrections include:

[0030] Calculate the M-1 differences ΔS' i Their respective absolute values ​​|ΔS' i | is used for comparison;

[0031] Determine the minimum value |ΔS' i | min and the minimum value |ΔS' i | min The corresponding output value S' for the time of falling asleep i,min ;

[0032] According to |ΔS' i | min and / or S' i,min To S” M The correction yields the output value S' for the time of falling asleep. M .

[0033] Furthermore, a second duration threshold T2 is preset;

[0034] If |ΔS' i If | < T2, then the output value S' at the time of falling asleep M Retain sleep time correction value S” M The value;

[0035] If |ΔS' i | min If T2 is greater than or equal to T2, then the correction value S for the time of falling asleep is... M The offset processing is performed to obtain the output value S' at the time of falling asleep. M The offset processing includes:

[0036] If S” M >S'i,min then the sleep time output value S' M =S" M -v2

[0037] If S" M <S' i,min then the sleep time output value S' M =S" M +v2

[0038] Wherein, the offset v2 satisfies: 0

[0039] Further, if S'1 to S' M-1 are not completely equal, and M-1 absolute values |ΔS' i | calculated are equal, then the sleep time output value S' M retains the value of the sleep time correction value S" M .

[0040] And / or, when M-1 absolute values |ΔS' i | are, there are x' minimum values |ΔS' i | min , and y' values equal |ΔS' i | eq , if x' < y', then the sleep time output value S' M retains the value of the sleep time correction value S" M . Wherein, |ΔS' i | eq > |ΔS' i | min , and x'+y'≤M-1, y'>1.

[0041] After the above technical solution, the present application has the following beneficial effects compared with the prior art.

[0042] In the present application, the sleep time is corrected twice, the first correction is according to the historical record of the sleep time calculation value, and the second correction is according to the historical record of the sleep time output value. For most users, their sleep habits have certain regularity, even if the actual sleep time is not the same every day, it will not have a large fluctuation. Therefore, according to the historical record of the sleep related data to correct the sleep time of this time, the sleep time after correction is closer to the real situation, which is beneficial to improve the accuracy of the sleep monitoring device in judging the sleep time.

[0043] In the present application, by setting the time length threshold, when the deviation of the current sleep time from the sleep time in the historical record exceeds the time length threshold, the current sleep time is corrected, otherwise the value of the current sleep time is retained. The above correction method can match the situation that the actual sleep time of the user has a certain fluctuation range every day, and when the sleep time deviates greatly from the historical record, the probability is that the device detection error is caused, at this time, the sleep time is corrected, and a more accurate sleep time can be obtained.

[0044] The specific embodiments of the present application will be further described in detail below with reference to the accompanying drawings. BRIEF DESCRIPTION OF DRAWINGS

[0045] The accompanying drawings, which are part of the present application, serve to provide a further understanding of the present application, and the schematic embodiments of the present application and their descriptions serve to explain the present application, but do not constitute an improper limitation on the present application. Obviously, the accompanying drawings in the following description are only some embodiments, and other drawings can be obtained by those skilled in the art without creative labor on the basis of these drawings. In the drawings:

[0046] Figure 1 is a flowchart of the sleep time correction method in the embodiments of the present application.

[0047] It should be noted that these drawings and written descriptions are not intended to limit the scope of the concept of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0048] In order to make the purpose, technical scheme and advantages of the embodiments of the present application more clear, the technical scheme in the embodiments will be described clearly and completely below with reference to the drawings in the embodiments of the present application. The following embodiments are used to illustrate the present application, but not to limit the scope of the present application.

[0049] As shown in Figure 1 , the sleep time correction method for sleep monitoring in the present application includes the following steps:

[0050] obtain the sleep time calculation value S M of the current detection;

[0051] According to the sleep time calculation values S1 to S M-1 of the previous M-1 detections, the first correction is performed on S M to obtain the sleep time correction value S" M ;

[0052] According to the sleep time output values S'1 to S' M-1 of the previous M-1 corrected sleep times, the second correction is performed on S" MThe second correction is performed to obtain the sleep time output value S' of this time M , and the correction process is completed.

[0053] Since most users have relatively stable sleep habits, the actual sleep time of each day often fluctuates within a certain range. In the above scheme, the sleep time calculation value S M is directly calculated according to the detection data, and the sleep time is corrected twice according to the historical sleep time, especially for sleep monitoring devices with large measurement deviations, the accuracy of sleep time judgment can be significantly improved after correction, thereby improving the monitoring effect of user sleep.

[0054] As a preferred embodiment of the present application, the first correction includes:

[0055] According to ΔS i = S M -S i , the difference ΔS M between the sleep time calculation value S i of this time and the sleep time calculation value S i of the i-th time in the previous M-1 times of detection is calculated; wherein i is any integer in the interval range [1, M-1];

[0056] According to the calculated M-1 differences ΔS i , the sleep time calculation value S M is corrected to obtain the sleep time correction value S" M .

[0057] Further, the correction of S M according to M-1 differences ΔS i includes:

[0058] The absolute values |ΔS i | of the M-1 differences ΔS i are calculated and compared;

[0059] The minimum value |ΔS i | min and the sleep time calculation value S i corresponding to the minimum value |ΔS min | i,min are determined.

[0060] According to |ΔS i | min and / or S i,min , the sleep time calculation value S M is corrected to obtain the sleep time correction value S" M .

[0061] As another preferred embodiment of the present invention, the second modification includes:

[0062] According to ΔS' i =S” M -S' i Calculate the sleep onset time correction value S” for this time. M The sleep time output value S' of the i-th sleep time output value among the previous M-1 corrected sleep time output values. i The difference between ΔS' i Where i is any integer in the interval [1, M-1];

[0063] Based on the calculated M-1 differences ΔS' i To S” M The correction yields the output value S' for the time of falling asleep. M .

[0064] Furthermore, based on M-1 differences ΔS' i To S” M The corrections include:

[0065] Calculate the M-1 differences ΔS' i Their respective absolute values ​​|ΔS' i | is used for comparison;

[0066] Determine the minimum value |ΔS' i | min and the minimum value |ΔS' i | min The corresponding output value S' for the time of falling asleep i,min ;

[0067] According to |ΔS' i | min and / or S' i,min To S” M The correction yields the output value S' for the time of falling asleep. M .

[0068] In the description of this invention, the sleep time data can refer to a calculated sleep time value, a corrected sleep time value, or a sleep time output value. The specific meaning of sleep time data in this invention can be understood according to the specific circumstances. In the above solution, the current sleep time data is compared with multiple historical sleep time data, and the closest historical sleep time data is found. The current sleep time data is corrected based on the closest historical sleep time data. This eliminates the influence of historical sleep time data that deviates significantly from the general situation on the correction result, which is beneficial for obtaining a more accurate sleep time output value S'. M .

[0069] Example 1

[0070] This embodiment provides a method for correcting the sleep onset time for sleep monitoring. The method can be applied to sleep monitoring devices to correct the calculated sleep onset time value obtained directly from user data by the sleep monitoring device, thereby obtaining a sleep onset time output value that is closer to the user's actual sleep situation.

[0071] In this embodiment, the sleep monitoring device can acquire the time of falling asleep in any way available in the prior art, such as collecting the user's body movement frequency data and calculating the time of falling asleep by the decrease in body movement frequency. No specific limitation is made here.

[0072] like Figure 1 As shown, the sleep time correction method described in this embodiment includes the following steps:

[0073] Obtain the calculated value S of the time of falling asleep during this test. M ;

[0074] The values ​​S1 to S2 are calculated based on the sleep onset time from the previous M-1 tests. M-1 For S M The first correction is performed, resulting in the sleep time correction value S. M ;

[0075] Based on the sleep time correction completed in the previous M-1 iterations, output values ​​S'1 to S' M-1 To S” M A second correction is performed to obtain the output value S' for the current sleep onset time. M Complete the correction process;

[0076] In this embodiment, M is a positive integer greater than or equal to 3. That is, the value S for the current sleep time must be calculated based on at least the previous two sleep times recorded in the historical data. M Make corrections.

[0077] Most users have relatively stable sleep habits. Excluding special circumstances, the same user generally goes to bed at a fixed time each day, and their actual sleep onset time should fluctuate within a certain time range, with very few, if any, instances of significant differences between the previous and current sleep onset times. Sleep monitoring devices typically store daily sleep onset time data. In this embodiment, the sleep onset time data from the historical records of the sleep monitoring device can be used to correct the current sleep onset time data twice, which helps to improve the accuracy of sleep onset time detection results.

[0078] In a further embodiment, the first modification specifically includes:

[0079] According to ΔS i =S M -S i Calculate the value S of the time when you fall asleep this time. M The calculated sleep time S of the i-th time among the calculated sleep time values ​​of the previous M-1 tests. i The difference ΔS between i Where i is any integer in the interval [1, M-1];

[0080] Based on the calculated M-1 differences ΔS i For S M The correction is performed to obtain the sleep time correction value S”. M .

[0081] Furthermore, based on M-1 differences ΔS i For S M The corrections include:

[0082] Calculate the M-1 differences ΔS i Their respective absolute values ​​|ΔS i | is used for comparison;

[0083] Determine the minimum value |ΔS i | min and the minimum value |ΔS i | min The corresponding sleep onset time calculation value S i,min ;

[0084] According to |ΔS i | min and / or S i,min For S M The correction is performed to obtain the sleep time correction value S”. M .

[0085] When M=5, the values ​​of i are 1, 2, 3, and 4 respectively. That is to say, the current sleep time data is corrected based on the previous four sleep time data.

[0086] The sleep monitoring device records the calculated sleep onset times S1, S2, S3, and S4 from the previous four times. These calculated values ​​are retrieved and subtracted from the current calculated sleep onset time S5, resulting in four differences ΔS1, ΔS2, ΔS3, and ΔS4. The absolute values ​​of these four differences are then taken to obtain |ΔS1|, |ΔS2|, |ΔS3|, and |ΔS4|, which are compared to determine the minimum absolute value, such as |ΔS4|. The corresponding calculated sleep onset time is S4. Finally, the current calculated sleep onset time S5 is corrected based on |ΔS4| and S4, resulting in a corrected sleep onset time value S”5.

[0087] In the above scheme, the current sleep time data is subtracted from the historical sleep time data respectively, so as to determine the historical sleep data closest to the current one, and correct the current sleep time data based on the historical sleep data. In this way, if there is sleep time data recorded in the historical record under special circumstances, which deviates greatly from the user's general habit, the influence of the sleep time data on the correction result can be excluded, and a more accurate sleep time output value can be obtained. For example, when the user rests early due to heavy workload during the day on a certain day, the sleep time data of the day will be significantly earlier than the sleep time data of other days in the historical record. If the current correction is based on the above-mentioned sleep time data which deviates significantly early, it is difficult to obtain a sleep time correction result closer to the real situation.

[0088] In a further scheme of the embodiment, a first time threshold T1 is preset, and when |ΔS i | min and S i,min , |ΔS i | min is compared with the first time threshold T1, and the correction process is determined according to the comparison result.

[0089] Specifically, if |ΔS i | min ≥ T1, the sleep time correction value S" is obtained by adding or subtracting the offset v1 to or from S M ; M wherein the offset v1 satisfies: 0 < v1 ≤ T1.

[0090] If |ΔS i | min < T1, the sleep time correction value S" M retains the value of the sleep time calculation value S M .

[0091] Further, when |ΔS i | min ≥ T1, if S M > S i,min , the sleep time correction value S" M = S M - v1; if S M < S i,min , the sleep time correction value S" M = S M + v1.

[0092] In the above scheme, when |ΔS i | min ≥ T1, the sleep time calculation value S Mthe value after the increase / decrease offset v1 (i.e., the correction value S M the value after the increase / decrease offset v1 (i.e., the correction value S i | min the value after the increase / decrease offset v1 (i.e., the correction value S i,min the value after the increase / decrease offset v1 (i.e., the correction value S i,min , which conforms to the correction principle of obtaining a more accurate correction result.

[0093] In this embodiment, the offset v1 is a preset constant, that is, in the first correction, if it is determined that the current value of the falling-asleep time calculation value S M needs to be offset, the specific offset v1 is the same each time.

[0094] In detail, in this embodiment, the value of the offset v1 can be set according to the accuracy of the detection result of the sleep monitoring device when the sleep monitoring device is shipped. When it is determined each time that the current value of the falling-asleep time calculation value S M needs to be offset in the first correction, the current falling-asleep time calculation value is offset by a fixed size.

[0095] Alternatively, the user can manually set the value of the offset v1 when using the sleep monitoring device, for example, after using the sleep monitoring device for a certain period of time, the user manually sets the value of the offset v1 according to the deviation between the output result of the falling-asleep time and the actual situation. After the user sets the value of the offset v1, when it is determined each time that the current value of the falling-asleep time calculation value S M needs to be offset in the first correction, the current falling-asleep time calculation value is offset by a fixed size according to the value of the offset v1 set by the user.

[0096] In this embodiment, when the sleep monitoring device calculates the falling-asleep time according to the collected data, the actual calculation result is first approximately processed to obtain the falling-asleep time calculation value to be corrected. In detail, the approximate processing refers to processing in units of 10 minutes according to the principle of rounding off. For example, if the actual calculation result is 08:47:40, the falling-asleep time calculation value after the approximate processing is 08:50:00. It should be noted that the specific value of the unit length can also be other lengths of time.

[0097] The first correction process in this embodiment is described below, where the value of M is 5.

[0098] The second row in Table 1-1 shows the falling-asleep time calculation value S5 of this time and the falling-asleep time calculation values S1, S2, S3 and S4 of the previous four times, and the third row shows the calculated absolute values |ΔS1|, |ΔS2|, |ΔS3| and |ΔS4|. As can be seen from Table 1, the smallest absolute value is |ΔS2| = 20 min.

[0099] Table 1-1

[0100] i 5 4 3 2 1 [SA i ]] 08:50:00 08:10:00 07:40:00 09:10:00 10:10:00 | ΔS i |]]> — 00:40:00 01:10:00 00:20:00 01:20:00

[0101] Assuming the first duration threshold T1 is 30 min, |AS2| < T1, at this time the sleep time correction value S"5 retains the sleep time calculation value S5, i.e. the sleep time correction value S"5 = S5 = 08:50:00.

[0102] Table 1-2 gives another example of sleep time data, at this time the minimum absolute value of the difference is |AS4| = 40 min.

[0103] Table 1-2

[0104] i 5 4 3 2 1 [SA i ]]> 08:50:00 09:30:00 07:20:00 07:50:00 10:50:00 | ΔS i |]]> — 00:40:00 01:30:00 01:00:00 02:00:00

[0105] In the case where the first duration threshold T1 is 30 min, |AS4| > T1 at this time, the offset correction needs to be made on the basis of S5. In this embodiment, the value of the offset v1 is 10 min, and since S5 < S4, the first correction result is: the sleep time correction value S"5 = S5 + v1 = 09:00:00.

[0106] In this embodiment, the sleep monitoring device can also detect wake time data, and can also correct the directly detected wake time data. In correcting the wake time data, a wake correction threshold T w wake time data is less than the wake correction threshold T w wake time data is less than the wake correction threshold T w , the current value of the wake time data is retained.

[0107] But in correcting the sleep time data, the correction principle is just the opposite of the correction method of the wake time data, i.e. for the case where the minimum deviation is greater than the first duration threshold T1, the current sleep time data is offset to a certain extent to achieve correction, and in the case where the minimum deviation is less than or equal to the first duration threshold T1, the current value of the sleep time data is retained.

[0108] Extensive data analysis reveals that while the wakefulness and sleep onset times of most users generally remain stable within a certain timeframe, the fluctuation in wakefulness is relatively smaller than the fluctuation in sleep onset. Statistics show that sleep onset time is closely related to daily activity levels; therefore, a user's daytime activity level influences the timing of their sleep onset, while wakefulness is primarily controlled by their biological clock. Since daily activity levels vary, the fluctuation in sleep onset time is often greater than the fluctuation in wakefulness.

[0109] Therefore, regarding the time of falling asleep, if the calculated time of falling asleep this time deviates slightly from the calculated times of falling asleep in the previous days, then the current calculated time of falling asleep is likely reliable and highly accurate, and no offset correction is needed. However, if the calculated time of falling asleep this time deviates significantly from the calculated times of falling asleep in the previous days, even from the closest calculated time of falling asleep, then the probability of an error in the current calculation is high. In this case, a certain degree of offset needs to be made based on the current calculated time of falling asleep to obtain more accurate data.

[0110] The first correction process in the sleep time correction method of this embodiment has been described above. The following will further explain how to perform the second correction after the first correction is completed.

[0111] Specifically, in this embodiment, the second modification includes:

[0112] According to ΔS' i =S” M -S' i Calculate the sleep onset time correction value S” for this time. M The sleep time output value S' of the i-th sleep time output value among the previous M-1 corrected sleep time output values. i The difference between ΔS' i Where i is any integer in the interval [1, M-1];

[0113] Based on the calculated M-1 differences ΔS' i To S” M The correction yields the output value S' for the time of falling asleep. M .

[0114] The second correction process is similar to the first correction process, with the difference being that the correction is based on the historical record of multiple sleep-onset time output values, i.e., each time the sleep-onset time data is corrected, rather than the sleep-onset time calculation value. By using the multiple sleep-onset time calculation values and sleep-onset time output values recorded by the sleep monitoring device to double correct the sleep-onset time data, the closeness of the final corrected sleep-onset time data to the actual situation can be further improved.

[0115] In a further aspect of the embodiment, the M-1 difference values ΔS' i are calculated according to the sleep-onset time output values S' M recorded by the sleep monitoring device. The correction of S" i includes:

[0116] calculating M-1 difference values ΔS' i each absolute value |ΔS' i | min and comparing them;

[0117] determining the minimum value |ΔS' i | min corresponding to the sleep-onset time output value S' i,min ;

[0118] correcting S" i | min and / or S' i,min according to |ΔS' M to obtain the sleep-onset time output value S' M .

[0119] Further, a second time threshold T2 is preset.

[0120] If |ΔS' i | < T2, the sleep-onset time output value S' M retains the value of the sleep-onset time correction value S" M .

[0121] If |ΔS' i | min ≥ T2, the sleep-onset time correction value S" M is subjected to offset processing to obtain the sleep-onset time output value S' M , and the offset processing includes:

[0122] If S" M > S' i,min , the sleep-onset time output value S' M = S" M - v2.

[0123] If S" M < S' i,minthe sleep time correction value S" M = S" M + v2;

[0124] wherein the offset v2 satisfies: 0 < v2≤ T2.

[0125] In the embodiment, the offset v2 is similar to the offset v1, and is also a preset constant. That is, when performing the second correction, the sleep time correction value S" M The offset v2 is the same when performing the offset processing.

[0126] In detail, the value of the offset v2 can be set when the sleep monitoring device is manufactured, or can be manually set by the user. In each second correction process, if it is determined that the sleep time correction value S" M The offset processing is performed, that is, the current sleep time correction value is offset by a fixed size.

[0127] The second correction process in the embodiment is exemplarily described below according to the case that M=5.

[0128] Table 2-1 is related sleep time data after the second correction based on Table 1-1 above, wherein the sleep time correction value S"5 of this time, the sleep time output values S'1 to S'4 of the previous four times, and the absolute values |AS'1| to |AS'4| of the four calculated difference values are shown. It can be seen that the minimum difference absolute value is |AS'2|=20min at this time.

[0129] Table 2-1

[0130] i 5 4 3 2 1 S i ]] 08:50:00 — — — — S i ]] — 08:20:00 07:40:00 09:10:00 10:00:00 | ΔS i |]]> — 00:30:00 01:10:00 00:20:00 01:10:00

[0131] When the preset second time threshold T2 is 30min, |AS'2|<T2, the sleep time output value S'5 retains the value of the sleep time correction value S"5 at this time, that is, the sleep time output value S'5=S"5=08:50:00.

[0132] Table 2-2 is related sleep time data after the second correction based on Table 1-2 above, wherein the minimum difference absolute value calculated is |AS'4|=40min.

[0133] Table 2-2

[0134] i 5 4 3 2 1 S i ]] 09:00:00 — — — — <![CDATA[S’ i ]]> — 09:40:00 07:30:00 07:50:00 10:40:00 | ΔS i |]]> — 00:40:00 01:30:00 01:10:00 01:40:00

[0135] When the second duration threshold T2 is 30 minutes, |ΔS'4| > T1, so the correction value for the sleep time S”5 needs to be offset. In this embodiment, the offset v2 is 10 minutes. Since S”5 < S'4, the second correction result is: the output value for the sleep time S'5 = S”5 + v1 = 09:10:00.

[0136] The second correction process is similar to the first. If the current sleep time data is relatively close to the historical data, the current value is retained. However, if there is a large difference from the historical data, it is likely that there is a significant error in the device's detection this time. The current sleep time correction value is then offset to obtain the final sleep time output value, which is more likely to match the actual situation.

[0137] In this embodiment, the calculated sleep time S obtained directly from the collected data is used. M First, calculate values ​​S1 to S2 based on multiple sleep onset times recorded in historical data. M-1 Perform the first correction, and then output values ​​S'1 to S' based on multiple sleep times in the historical records. M-1 A second correction is performed. Through this double correction method, the output value S' at the time of falling asleep after correction can be significantly improved. M This improves the accuracy of sleep monitoring devices in determining sleep time, making it closer to the user's actual sleep time. By adopting the sleep time correction method in this embodiment, the accuracy of sleep time determination can be improved, which is conducive to obtaining more accurate sleep-related data and enabling sleep monitoring devices to more effectively evaluate the user's sleep quality.

[0138] Example 2

[0139] This embodiment is a further limitation of the first embodiment described above. In the first correction process, when the sleep time is calculated as S1 to S2, the calculation values ​​are... M-1 When they are all different, the absolute value of the M-1 differences, |ΔS, is calculated. i |and compare them to determine the minimum value|ΔS i | min According to |ΔS i | min The result of comparing the value with the first duration threshold T1 determines whether S needs to be adjusted. M Perform offset processing, i.e., whether or not it is necessary to perform offset processing in S. M Add or subtract offset v1 from the base.

[0140] But if S1 to S M-1 They are not exactly equal, but the calculated M-1 absolute values ​​|ΔS i If all are equal, then the sleep time correction value S” M Retain the calculated value S of the time of falling asleepM The value.

[0141] Furthermore, if it is S1 to S M-1 When all are equal, we can also use |ΔS i Compare the result with the first duration threshold T1 to determine whether it is appropriate for S. M Perform offset processing.

[0142] Specifically, for example, Table 3-1 below gives an example of sleep time data, where S1 to S4 are all equal to 09:00:00, and thus |ΔS1| to |ΔS4| are all 60 minutes.

[0143] Table 3-1

[0144] i 5 4 3 2 1 [SA i ]]> 08:00:00 09:00:00 09:00:00 09:00:00 09:00:00 | ΔS i |]]> — 01:00:00 01:00:00 01:00:00 01:00:00

[0145] When the first time threshold T1 is 30 minutes, then |ΔS i If |>T1, offset correction is still needed based on S5. In this embodiment, the offset value v1 is 10min, since S5<S i The first correction result is: the correction value for the time of falling asleep is S”5=S5+v1=08:10:00.

[0146] Table 3-2 below provides another example of sleep onset time data, where S1 to S4 are not exactly equal, but the calculated |ΔS1| to |ΔS4| are all 120 minutes. In this case, the sleep onset time correction value S”5 retains the value of the calculated sleep onset time S5, that is, the sleep onset time correction value S”5 = S5 = 08:00:00.

[0147] Table 3-2

[0148] i 5 4 3 2 1 [SA i ]]> 08:00:00 06:00:00 06:00:00 10:00:00 10:00:00 | ΔS i |]]> — 02:00:00 02:00:00 02:00:00 02:00:00

[0149] In a further embodiment, during the second correction process, the output values ​​S'1 to S' are changed when the sleep time begins. M-1 When they are all different, M-1 absolute values ​​of the differences |ΔS' are calculated. i |and compare them to determine the minimum value|ΔS' i | min According to |ΔS' i | min The result of comparing the value with the second duration threshold T2 determines whether S needs to be adjusted. M Perform offset processing, i.e., whether or not it is necessary to perform offset processing in S. M Add or subtract offset v2 from the base.

[0150] And if S'1 to S' M-1not equal, but the calculated M-1 absolute values |ΔS i | are equal, the output value S M of the sleep time is kept. M The sleep time correction value S M-1 is kept. i The value of the sleep time S

[0151] Further, if S'1 to S' M are all equal, the size comparison result of |ΔS i | and the second time threshold T2 can be used to determine whether to perform the offset processing on S M .

[0152] The second correction process is similar to the first correction process, and will not be described again.

[0153] In this embodiment, a solution is provided for the special case that the deviation of the sleep time data of this time and the sleep time data of the historical records are all the same. When the sleep time data of the historical records are all earlier or later than the sleep time data of this time, the size comparison of the deviation and the time threshold is still used to determine whether to perform the offset processing. However, if some of the sleep time data of the historical records are earlier than the sleep time data of this time, and some are later than the sleep time data of this time, the correction error may occur no matter which direction is used to perform the offset. In this case, keeping the value of the sleep time data of this time can ensure the accuracy of the sleep time detection result with a higher probability.

[0154] Embodiment Three

[0155] This embodiment is a further limitation of the above-mentioned embodiments one and two. Among the M-1 absolute values |ΔS i |, there are x values of the minimum absolute value |ΔS i | min , and y values of the absolute value |ΔS i | eq If x < y, the sleep time correction value S M is kept as the sleep time calculation value S M , wherein |ΔS i | eq > |ΔS i | min , and x + y ≤ M-1, y > 1.

[0156] Preferably, the x absolute values |ΔS i | min correspond to x S i,min , which are all greater than S M , and the y absolute values |ΔS i |eq corresponding to y i,eq at the same time are less than S M or when x |ΔS i min corresponding to x i,min at the same time are less than S M and y |ΔS i eq corresponding to y i,eq at the same time are greater than S M when the sleep time correction value S" M retains the value of the sleep time calculation value S M .

[0157] For example, a specific case of sleep time data is shown in Table 4-1.

[0158] Table 4-1

[0159] i 5 4 3 2 1 [SA i ]]> 08:00:00 06:00:00 09:00:00 06:00:00 06:00:00 | ΔS i |]]> — 02:00:00 01:00:00 02:00:00 02:00:00

[0160] In the case shown in Table 4-1, the minimum value of the absolute value is |ΔS3| = 60 min, and there is only one, i.e. x = 1. The other |ΔS4| = |ΔS2| = |ΔS1| = 120 min, i.e. y = 3. At the same time, S3> S5 and S4= S2= S1< S5. In the above case, the special condition x < y is satisfied, and at this time the sleep time correction value S"5retains the value of the sleep time calculation value S5, i.e. the sleep time correction value S"5= S5= 08:00:00 after the first correction.

[0161] In a further aspect of the embodiment, before calculating |ΔS i |, the sleep time calculation values S1 to S M-1 are first grouped according to whether there are identical values. The number of sleep time calculation values in each group of identical values is counted, and the absolute value of the difference from S M is calculated. Then it is determined whether the above special condition is satisfied, and when the above special condition is satisfied, the sleep time correction value S"5retains the value of the sleep time calculation value S5.

[0162] For example, for the sleep time data in Table 4-1, it is first determined that S4= S2= S1and the corresponding absolute value of the difference is 120 min. The absolute value of the difference corresponding to S3is 60 min < 120 min. Therefore, in the current case x = 1 and y = 3, the special condition x < y is satisfied.

[0163] In a preferred aspect of the embodiment, for S1 to S M-1 ​​For values ​​that are different but relatively close (e.g., by setting an approximate threshold to determine whether different values ​​are close), two or more sleep time calculations can be grouped into the same group with the same value. Then, based on the number of data points in each group and their correlation with S... M The absolute value of the difference determines whether the value of the sleep time calculation S5 should be retained in the first correction.

[0164] Similarly, in this embodiment, when a similar special case occurs in the second correction, the sleep time output value S' is also set. M Retain sleep time correction value S” M The value.

[0165] Specifically, when there are M-1 absolute values ​​|ΔS' i In |ΔS'|, there exist x' minimum values ​​with the same value. i | min and y' equal values ​​of |ΔS' i | eq If x' < y', then the output value S' is used at the time of falling asleep. M Retain sleep time correction value S” M The value of |ΔS'; where, i | eq >|ΔS' i | min And x'+y'≤M-1, y'>1.

[0166] Preferably, when x' |ΔS' i | min The corresponding x' S' i,min At the same time greater than S” M And y' |ΔS' i | eq The corresponding y' S' i,eq At the same time, it is less than S” M When, or when x' |ΔS' i | min The corresponding x' S' i,min At the same time, it is less than S” M And y' |ΔS' i | eq The corresponding y' S' i,eq At the same time greater than S” M At that time, the output value S' is displayed at the time of falling asleep. M Retain sleep time correction value S” M The value.

[0167] Furthermore, during the second correction, when calculating |ΔS' i First, output values ​​S'1 to S' for the time of falling asleep. M-1Group the data based on whether there are identical values. Count the number of output values ​​at the time of sleep in each group with identical values ​​(the number should include at least x' and y'), and calculate its relationship with S". M The absolute value of the previous difference. Then it is determined whether the special condition x'<y' is met. If the above special condition is met, the output value S'5 at the time of falling asleep retains the value of the correction value S”5 at the time of falling asleep.

[0168] Preferably, an approximate threshold can also be set for S'1 to S' M-1 Two or more sleep-on time output values ​​that are different but relatively close are statistically grouped into the same group with the same value.

[0169] The process of the second revision is similar to that of the first revision, and will not be illustrated with examples.

[0170] In this embodiment, we consider the special case where the probability of a certain range of sleep time data appearing in the historical records is relatively high, but the deviation from the current sleep time data is not the smallest deviation in the historical records. In this case, it is possible that the historical sleep time data with the smallest deviation is due to a special case, and correcting according to the historical sleep time data with the smallest deviation may cause problems. In this case, we do not shift the current sleep time data, but retain the current value, which can avoid the above-mentioned erroneous correction problems.

[0171] Example 4

[0172] The difference between this embodiment and embodiments one to three above is that the offsets v1 and v2 are calculated based on the current region and / or the time of falling asleep. M Parameters related to the time interval in which it is located.

[0173] Among them, the calculated value S of the time to fall asleep M The time interval refers to: dividing the day into multiple time periods based on a set duration, with the sleep time calculated as the value S. M The time period it falls into is its time interval. For example, dividing the time period into 2-hour intervals, we can get multiple time periods, such as 05:00:00~07:00:00, 07:00:00~09:00:00, 09:00:00~11:00:00, etc., which are time intervals. If S M =08:30:00, which falls within the time interval of 07:00:00 to 09:00:00.

[0174] It should be noted that for a calculated sleep time value that falls exactly between two time periods, it is determined to fall into the latter time period. For example, S... M= 09:00:00, the time interval in which it is located is 09:00:00~11:00:00. But it can be understood that the sleep-onset time calculation value on the time node falling between two time periods can also be determined to fall into the previous time period.

[0175] Specifically, in the embodiment, the sleep monitoring device is in communication connection with the server, the server can receive and record the sleep-onset time data uploaded by multiple sleep monitoring devices, and can also record the respective values of the offset v1 and the offset v2 adopted in the correction process of the sleep-onset time data.

[0176] In detail, the above data information is recorded in the server in the form of a statistical table, and the server stores different statistical tables corresponding to different regions or geographical time zones. In each statistical table, the sleep-onset time calculation values S M in different time intervals are classified, and the corresponding relationship between the sleep-onset time calculation values S M and the respective values of the offset v1 and the offset v2 adopted in the correction thereof in different time intervals is recorded, and the probability of occurrence of different values of the offset v1 and the probability of occurrence of different values of the offset v2 in the same time interval are counted.

[0177] In the sleep-onset time correction, for the first correction process, if it is judged that offset processing is needed, the corresponding statistical table is first called from the server according to the region or geographical time zone in which the sleep monitoring device is located, and then the time interval in which the current sleep-onset time calculation value S M is located is judged, and the data in the same time interval is called in the statistical table. The value of the offset v1 with the highest probability of occurrence is determined from the called data, that is, the value of the offset v1 adopted in the majority of correction results in the same time interval, and the offset processing is performed on the current sleep-onset time calculation value S M .

[0178] Similarly, for the second correction process, if it is judged that offset processing is needed, the corresponding statistical table is first called from the server according to the region or geographical time zone in which the sleep monitoring device is located, and then the time interval in which the current sleep-onset time calculation value S M is located is judged, and the data in the same time interval is called in the statistical table. The value of the offset v2 with the highest probability of occurrence is determined from the called data, that is, the value of the offset v2 adopted in the majority of correction results in the same time interval, and the offset processing is performed on the sleep-onset time correction value S M .

[0179] Sleep habits of users often exist regional differences, the embodiment in the determination of the offset v1 and v2 value used in the correction process, the current region as a factor into the scope of consideration, help to improve the accuracy of the correction result. At the same time, for falling asleep time interval into different time interval, the range of detection error often also exist certain differences. The embodiment in the determination of the offset v1 and v2 value is obtained by statistical data, the final offset v1 and v2 value used is the same time interval in most correction results of the offset v1 and v2 value, can further ensure the accuracy of the correction result.

[0180] Embodiment five

[0181] The difference between the embodiment and the above embodiment four is that the offset v1 and the offset v2 are parameters related to the current use environment and / or application scenario.

[0182] Similar to embodiment two, the server can collect a large number of falling asleep time data uploaded by different sleep monitoring devices, and record as multiple statistical tables according to different use environments / application scenarios. And in the statistical table, the probability of different values of offset v1 and the probability of different values of offset v2 under the same use environment / application scenario are counted.

[0183] When correcting the falling asleep time data, the sleep monitoring device determines the current use environment and / or application scenario, retrieves the corresponding statistical table from the server, and retrieves the value of the offset v1 with the highest probability for the first correction and the value of the offset v2 with the highest probability for the second correction.

[0184] In a further scheme of the embodiment, the above method of determining the offset v1 and v2 value according to the current use environment and / or application scenario can also be combined with the method of determining the offset v1 and v2 value in embodiment four, considering the current region, the time interval where the falling asleep time calculation value S M The time interval, and the current use environment and application scenario, help to further improve the correction accuracy.

[0185] In the embodiment, the offset size of the falling asleep time data in the correction process is determined according to the current use environment and / or application scenario, which is conducive to obtaining a correction result more in line with the current actual situation.

[0186] Embodiment six

[0187] The difference between the embodiment and the above embodiments one to three is that the offset v1 and the offset v2 value are determined according to the historical falling asleep time data.

[0188] In one embodiment, during the first correction, the sleep time correction method further includes: based on |ΔS i | min Determine the value of offset v1.

[0189] Specifically, the value of offset v1 varies with |ΔS i | min It increases with the increase of [something]. The calculated value S for the time of falling asleep this time. M The larger the difference between the calculated sleep time and the historical data, the larger the corresponding offset v1 value, which is the value of the calculated sleep time S. M The larger the offset, the better.

[0190] Similarly, when making the second correction, the method for correcting the sleep time also includes: based on |ΔS' i | min Determine the value of offset v2.

[0191] Specifically, the value of offset v2 varies with |ΔS' i | min It increases with the increase of [something]. The current sleep time correction value S” M The larger the difference between the output value and the one closest to the historical sleep time, the larger the corresponding offset v2 value, which is the correction value S for the sleep time. M The larger the offset, the better.

[0192] Preferably, in this embodiment, the values ​​of offsets v1 and v2 can be determined in the following way: v1 = α1 × |ΔS i | min v2=α2×|ΔS' i | min , where α1 and α2 are preset proportional coefficients.

[0193] In the above scheme, the offset range for correcting the sleep time data is determined based on the difference between the current sleep time data and the historical sleep time data. The greater the difference between the current sleep time data and the historical data, the greater the corresponding offset range, so that the final corrected sleep time output value is stable within a certain range, and the correction result is more accurate.

[0194] In another embodiment, according to S1 to S... M-1 The values ​​of one or more determined offsets v1 in the equation are determined according to S'1 to S'. M-1 One or more values ​​of a specific offset v2 in the equation.

[0195] Specifically, when S M >S i,minWhen S1≤S≤S2, all the sleep time calculation values in S1 to S2 that satisfy S1≤S≤S2 are selected, and the respective |ΔS| is weightedly averaged to obtain the value of the offset v1. The greater the value of |ΔS|, the smaller the corresponding weight coefficient. M-1 M When S1≤S≤S2, all the sleep time calculation values in S1 to S2 that satisfy S1≤S≤S2 are selected, and the respective |ΔS| is weightedly averaged to obtain the value of the offset v1. The greater the value of |ΔS|, the smaller the corresponding weight coefficient. i i When S1≤S≤S2, all the sleep time calculation values in S1 to S2 that satisfy S1≤S≤S2 are selected, and the respective |ΔS| is weightedly averaged to obtain the value of the offset v1. The greater the value of |ΔS|, the smaller the corresponding weight coefficient. i

[0196] When S1≤S≤S2, all the sleep time calculation values in S1 to S2 that satisfy S1≤S≤S2 are selected, and the respective |ΔS| is weightedly averaged to obtain the value of the offset v1. The greater the value of |ΔS|, the smaller the corresponding weight coefficient. M i,min When S1≤S≤S2, all the sleep time calculation values in S1 to S2 that satisfy S1≤S≤S2 are selected, and the respective |ΔS| is weightedly averaged to obtain the value of the offset v1. The greater the value of |ΔS|, the smaller the corresponding weight coefficient. M-1 M When S1≤S≤S2, all the sleep time calculation values in S1 to S2 that satisfy S1≤S≤S2 are selected, and the respective |ΔS| is weightedly averaged to obtain the value of the offset v1. The greater the value of |ΔS|, the smaller the corresponding weight coefficient. i i When S1≤S≤S2, all the sleep time calculation values in S1 to S2 that satisfy S1≤S≤S2 are selected, and the respective |ΔS| is weightedly averaged to obtain the value of the offset v1. The greater the value of |ΔS|, the smaller the corresponding weight coefficient. i

[0197] When S1≤S≤S2, all the sleep time calculation values in S1 to S2 that satisfy S1≤S≤S2 are selected, and the respective |ΔS| is weightedly averaged to obtain the value of the offset v1. The greater the value of |ΔS|, the smaller the corresponding weight coefficient. M i,min When S1≤S≤S2, all the sleep time calculation values in S1 to S2 that satisfy S1≤S≤S2 are selected, and the respective |ΔS| is weightedly averaged to obtain the value of the offset v1. The greater the value of |ΔS|, the smaller the corresponding weight coefficient. M-1 M When S1≤S≤S2, all the sleep time calculation values in S1 to S2 that satisfy S1≤S≤S2 are selected, and the respective |ΔS| is weightedly averaged to obtain the value of the offset v1. The greater the value of |ΔS|, the smaller the corresponding weight coefficient. i i When S1≤S≤S2, all the sleep time calculation values in S1 to S2 that satisfy S1≤S≤S2 are selected, and the respective |ΔS| is weightedly averaged to obtain the value of the offset v1. The greater the value of |ΔS|, the smaller the corresponding weight coefficient. i

[0198] When S1≤S≤S2, all the sleep time calculation values in S1 to S2 that satisfy S1≤S≤S2 are selected, and the respective |ΔS| is weightedly averaged to obtain the value of the offset v1. The greater the value of |ΔS|, the smaller the corresponding weight coefficient. M i,min When S1≤S≤S2, all the sleep time calculation values in S1 to S2 that satisfy S1≤S≤S2 are selected, and the respective |ΔS| is weightedly averaged to obtain the value of the offset v1. The greater the value of |ΔS|, the smaller the corresponding weight coefficient. M-1 M When S1≤S≤S2, all the sleep time calculation values in S1 to S2 that satisfy S1≤S≤S2 are selected, and the respective |ΔS| is weightedly averaged to obtain the value of the offset v1. The greater the value of |ΔS|, the smaller the corresponding weight coefficient. i i When S1≤S≤S2, all the sleep time calculation values in S1 to S2 that satisfy S1≤S≤S2 are selected, and the respective |ΔS| is weightedly averaged to obtain the value of the offset v1. The greater the value of |ΔS|, the smaller the corresponding weight coefficient. i

[0199] In the above scheme, the specific values of the offset v1 and the offset v2 are determined according to the comprehensive analysis of multiple sleep time data in the historical record, which can avoid the error correction that may occur when the offset value is determined according to a single historical data.

[0200] ​​​​​​​​​​​​​​​In the embodiment, the values of the offsets v1 and v2 are determined according to the deviation between the historical sleep time data and the current sleep time data. Compared with the historical data, the greater the deviation, the greater the correction range. The final corrected sleep time output value can fall within a certain range, which is consistent with the actual situation of the general user's sleep process, and the correction success rate is higher.

[0201] The above merely describes the preferred embodiments of the present application, and does not limit the present application in any form. Although the present application has been disclosed as above with the preferred embodiments, it is not intended to limit the present application. Any person skilled in the art can make some changes or modifications to the above-mentioned technical content without departing from the technical solution of the present application, and any simple modification, equivalent change and modification of the above embodiments made according to the technical essence of the present application still belong to the scope of the present application.

Claims

1. A sleep onset time correction method for sleep monitoring, characterized in that, The method comprises the following steps: acquiring a sleep onset latency value S for the current detection M ; According to the values S1 to S calculated at the time of the previous M-1 detections M-1 S M The first correction is performed to obtain the sleep time correction value S" M ; According to the previous M-1 times of completed correction, the sleep time output value S'1 to S' M-1 S'' M is corrected for the second time to obtain the sleep time output value S' of this time M , and the correction process is completed. The first correction comprises: According to ΔS i =S M -S i Calculate the value S of the time when you fall asleep this time. M The calculated sleep time S of the i-th time among the calculated sleep time values ​​of the previous M-1 tests. i The difference ΔS between i Where i is any integer in the interval [1, M-1]; According to the calculated M-1 difference values ΔS i S M corrected to obtain a sleep time correction value S'' M , specifically includes: Calculate M-1 differences ΔS i Each absolute value |ΔS i |, compare; determining the minimum value |ΔS i | min and the minimum value |ΔS i | min the corresponding sleep onset time calculation value S i,min ; If |ΔS i | min ≥ T1, then the sleep time S M is increased or decreased by an offset v1 to obtain a sleep time correction value S'' M ; If |ΔS i | min If <T1, then the sleep time correction value S'' M Retain the calculated value S of the time of falling asleep M The value; Wherein, the offset v1 satisfies: 0 < v1 ≤ T1.

2. The sleep onset time correction method for sleep monitoring of claim 1, wherein, In |ΔS i | min ≥ T1, if S M > S i,min , then the sleep time correction value S'' M = S M - v1; if S M < S i,min , then the sleep time correction value S'' M = S M + v1.

3. The sleep onset time correction method for sleep monitoring of claim 1, wherein, If S1 to S M-1 are not exactly equal, and the calculated M-1 absolute values |ΔS i are not equal, then the sleep time correction value S'' M is retained as the value of the sleep time calculation value S M ; and / or, when M-1 absolute values |AS i | exist, x minima |AS i | min and y equal values |AS i | eq If x < y, then the sleep time correction value S" M retains the value of the sleep time calculation value S M ; wherein |AS i | eq > |AS i | min , and x+y = M-1, y > 1.

4. The sleep onset time correction method for sleep monitoring according to any one of claims 1 to 3, characterized in that, The second correction comprises: According to ΔS' i =S'' M -S' i Calculate the correction value S'' for this sleep onset time. M The sleep time output value S' of the i-th sleep time output value among the previous M-1 corrected sleep time output values. i The difference between ΔS' i Where i is any integer in the interval [1, M-1]; According to the calculated M-1 difference values ΔS' i corrected to obtain the sleep onset time output value S' M M .​ 5. The sleep onset time correction method for sleep monitoring according to claim 4, characterized in that, According to M-1 differences ΔS' i correction of S'' M includes M-1 difference values ΔS' are calculated i the respective absolute values |ΔS' i are compared; determining the minimum value |ΔS' of the difference |ΔS' = S' - S i | min and the minimum value |ΔS' i | min corresponding to the falling asleep time output value S' i,min ; According to |ΔS' i | min and / or S' i,min corrects S'' M to obtain the sleep onset time output value S' M .

6. The sleep onset time correction method for sleep monitoring according to claim 5, characterized in that, A second time threshold T2 is preset; If |ΔS'| i M S' = S' + ΔS' M S' = S' + ΔS'​ If |ΔS'| i | min ≥ T2, then the falling-asleep time correction value S'' M is subjected to offset processing to obtain a falling-asleep time output value S' M , and the offset processing includes: If S" M > S' i,min then the falling asleep time instant output value S' M = S" M - v2; If S" M <S' i,min then the sleep time output value S' M =S" M +v2; Wherein, the offset v2 satisfies: 0 < v2 ≤ T2.

7. The sleep onset time correction method for sleep monitoring of claim 5, wherein, If S'1 to S' M-1 are not all equal, and the calculated M-1 absolute values |ΔS' i are not all equal, then the output value S' M at the time of falling asleep is outputted. M The value of the correction value S'' and / or, when M-1 absolute values |ΔS' i | exist, x' minimum values |ΔS' i | min and y' values equal |ΔS' i | eq values, if x' < y', then the sleep time output value S' M retains the value of the sleep time correction value S'' M ; wherein |ΔS' i | eq > |ΔS' i | min , and x' + y' ≤ M-1, y' > 1.

Images