Sleep adjusting method combining sleep mode learning and optimization

Abstract

The invention relates to the technical field of sleep data processing, and provides a sleep adjustment method combining sleep mode learning and optimization. The method comprises the following steps: learning sleep sign parameters based on deep sleep test data of a user to obtain a deep sleep mode parameter model; in the sleep process, detecting physical sign parameters and analyzing in combination with the model to obtain deep sleep entry degree and entry reliability; when the entry degree is smaller than a threshold value, setting an adjustment step length according to the reliability, and performing optimization adjustment in the sleep environment parameter space until the entry degree is larger than or equal to the threshold value; continuing to adjust the sleep environment parameters until the deep sleep time reaches the threshold value, and completing sleep adjustment. The technical problem that the sleep environment cannot be adjusted in real time according to the user state is solved, and the technical effects that the sleep environment parameters are automatically adjusted by learning and analyzing the deep sleep mode of the user, the deep sleep quality is improved, and the duration time is prolonged are achieved.

Description

Technical Field

[0001] The present application relates to the technical field of sleep data processing, and in particular to a sleep regulation method combining sleep pattern learning and optimization. Background Art

[0002] With the accelerating pace of modern life and increasing work pressure, sleep problems have become a major factor affecting people's health. Many people experience difficulty entering deep sleep or insufficient deep sleep duration, which in turn affects sleep quality. Deep sleep is a crucial stage for the body to restore energy and strength, and its quality is directly related to human health. Therefore, how to effectively improve the quality of a user's deep sleep has become a hot topic of research. Existing sleep adjustment devices can perform simple sleep environment adjustments by monitoring the user's vital parameters (such as heart rate and respiratory rate). However, these adjustment methods are often based on fixed parameter adjustment schemes, lack personalized adjustment capabilities, and are unable to accurately adjust according to the user's real-time sleep state. Furthermore, existing sleep adjustment methods often lack systematic learning and optimization of deep sleep patterns, resulting in suboptimal adjustment effects during complex sleep processes. This is especially true when the user's deep sleep entry level is low, making it difficult to effectively help the user enter deep sleep smoothly. Furthermore, most existing sleep adjustment solutions focus solely on adjusting a single sleep environment parameter, such as temperature or humidity, and fail to comprehensively consider the interplay between multiple sleep environment parameters. This results in poor adjustment and optimization efficiency and makes it difficult to achieve precise deep sleep management. Summary of the Invention

[0003] This application aims to solve the technical problem of being unable to adjust the sleeping environment according to the user's status in real time by providing a sleep adjustment method that combines sleep pattern learning and optimization.

[0004] In view of the above problems, the present application provides a sleep regulation method that combines sleep pattern learning and optimization.

[0005] The present application provides a sleep adjustment method that combines sleep pattern learning and optimization, the method comprising: learning sleep sign parameters in a deep sleep mode based on a user's deep sleep test data to obtain a deep sleep pattern parameter model; detecting the user's physical sign parameters during the user's sleep, and performing deep sleep analysis in combination with the deep sleep pattern parameter model to obtain a deep sleep entry degree and entry reliability; when the deep sleep entry degree is less than an entry degree threshold value, setting a sleep adjustment step size according to the entry reliability, and adjusting and optimizing the sleep environment parameters within a sleep environment parameter space until the deep sleep entry degree is greater than or equal to the entry degree threshold value; and continuing to adjust the sleep environment parameters until the deep sleep time reaches the deep sleep time threshold value, thereby completing the sleep adjustment.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages: The sleep adjustment method described above combines sleep pattern learning and optimization. Based on the user's deep sleep test data, it learns their vital signs parameters in deep sleep mode and constructs a deep sleep pattern parameter model. As the user falls asleep, their vital signs are detected in real time and combined with the model to perform deep sleep analysis, thereby obtaining accurate deep sleep entry and entry reliability. This method accurately captures changes in the user's sleep state. When the user's deep sleep entry level is detected to be below a threshold, the adjustment step size is dynamically set based on the entry reliability, and personalized optimization and adjustment are performed within the sleep environment parameter space. This allows the user to gradually enter their ideal deep sleep state, ensuring the accuracy and effectiveness of the adjustment. On this basis, continuous optimization of the sleep environment parameters further extends the user's deep sleep time, ultimately improving sleep quality and efficiently managing the sleep process.

[0007] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0009] Figure 1 1. A flowchart of a sleep regulation method combining sleep pattern learning and optimization in one embodiment; Figure 2 FIG. 1 is a flow chart of obtaining a deep sleep pattern parameter model in a sleep regulation method combining sleep pattern learning and optimization in one embodiment. DETAILED DESCRIPTION

[0010] The embodiments of the present application solve the technical problem of being unable to adjust the sleeping environment according to the user's status in real time by providing a sleep adjustment method that combines sleep pattern learning and optimization.

[0011] The following will be combined with the accompanying drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only some of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0012] It should be noted that the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or are inherent to these processes, methods, products or devices.

[0013] Examples, such as Figure 1 As shown, the present application provides a sleep regulation method that combines sleep pattern learning and optimization, the method comprising: Based on the user's deep sleep test data, sleep sign parameters in deep sleep mode are learned to obtain a deep sleep mode parameter model.

[0014] In this embodiment of the present application, multiple standard vital sign parameters, such as respiration, body temperature, and heart rate, are collected from multiple deep sleep states over a user's history to obtain the user's deep sleep test data, which serves as the data foundation for the current analysis. Because a person's vital sign parameters may fluctuate over time, multiple samples of these parameters are collected and analyzed to ensure model accuracy. Subsequently, deep sleep mode vital sign parameters are learned based on this data. Deep sleep mode is detected using electroencephalogram (EEG) detection technology, which records and analyzes electrical signals from brain activity. EEG technology distinguishes sleep stages based on different brain wave frequencies, particularly deep sleep (typically the third or fourth NREM stage). In these stages, EEG primarily exhibits low-frequency, high-amplitude slow wave activity, commonly known as delta waves (0.5Hz to 4Hz), which are hallmarks of deep sleep. Using the vital sign parameters collected from the deep sleep test data, a deep sleep mode parameter model is constructed to reflect the user's physiological characteristics during deep sleep, ensuring that the user maintains good deep sleep quality throughout their sleep.

[0015] Further, if Figure 2 As shown, this application provides a method for learning sleep sign parameters in deep sleep mode based on the user's deep sleep test data to obtain a deep sleep mode parameter model, including: Based on the deep sleep test record data of the user in the historical time, the vital sign parameters during deep sleep are collected to obtain multiple sample vital sign parameter sets; based on the multiple sample vital sign parameter sets, multiple sample average vital sign parameters are calculated to obtain multiple learning vital sign parameters; and the multiple learning vital sign parameters are integrated as a deep sleep pattern parameter model.

[0016] Preferably, deep sleep stage data is extracted from the user's past sleep records to form deep sleep test record data. Multiple vital sign parameters for each deep sleep stage are then extracted from the deep sleep test record data. These data are then used as multiple sample vital sign parameter sets, with each sample representing a set of vital sign data from a single deep sleep episode. Subsequently, statistical analysis is performed on these multiple sample vital sign parameter sets. Specifically, the average value of each vital sign parameter (such as respiratory rate and heart rate) across multiple samples is calculated to obtain the average vital sign parameters for these multiple samples. These parameters represent the typical physiological characteristics of the user during deep sleep. The calculated average vital sign parameters are then used as learned vital sign parameters. These learned vital sign parameters are used to characterize the user's stable physiological characteristics during deep sleep and to construct a deep sleep pattern model. Subsequently, the multiple learned vital sign parameters are integrated and packaged to form a deep sleep pattern parameter model. This model reflects the user's typical deep sleep state and serves as the basis for subsequent sleep regulation and analysis. The model comprises an input layer, a processing layer, and an output layer. The input layer receives the user's vital signs, the processing layer performs deep sleep analysis based on the learned parameters, and the output layer outputs the deep sleep analysis results. This deep sleep pattern parameter model allows for real-time determination of the difference between the user's current vital signs and a typical deep sleep state, providing a basis for initiating and regulating deep sleep.

[0017] During the user's sleep, the user's physical sign parameters are detected, and deep sleep analysis is performed in combination with the deep sleep pattern parameter model to obtain the deep sleep entry degree and entry reliability.

[0018] In one embodiment, after the user falls asleep, the user's vital sign parameters, such as heart rate, respiratory rate, body temperature, etc., are monitored in real time through sensors. These parameters are an important basis for judging the user's current sleep state. The currently detected vital sign parameters are input into the deep sleep mode parameter model for comparative analysis. By performing a difference comparison between each vital sign parameter and the corresponding learning vital sign parameter within the model, and then performing a mean calculation on the difference comparison results, the current deep sleep entry degree is obtained, which indicates the degree of closeness between the user's current sleep state and the deep sleep state. Because the deep sleep entry degree only reflects the degree of closeness between the current state and deep sleep, in order to ensure the reliability of the analysis results, based on the difference comparison results, the entry reliability is calculated by combining the volatility of the deep sleep entry degree to ensure the subsequent accurate analysis and adjustment optimization of the user's current deep sleep state.

[0019] Furthermore, the present application provides a method for detecting the user's physical parameters during the user's sleep and performing deep sleep analysis in combination with the deep sleep pattern parameter model, including: During the user's sleep, the user's vital sign parameters are detected; based on the vital sign parameters, combined with multiple learned vital sign parameters in the deep sleep pattern parameter model, a deep sleep entry analysis is performed to obtain multiple basic deep sleep entry degrees, and the deep sleep entry degree is calculated; based on the multiple basic deep sleep entry degrees, the reliability of the deep sleep entry analysis is calculated to obtain entry reliability.

[0020] Optionally, during the user's sleep, sensors detect the user's vital signs in real time, and the collection period for each vital sign parameter is recorded regularly according to a pre-set period. The collected vital sign parameters are then input into a deep sleep mode parameter model, which transmits these vital sign parameters to the processing layer via the input layer. In the processing layer, each vital sign parameter is compared with the corresponding learned vital sign parameter, and the similarity between them is calculated to obtain multiple basic deep sleep entry degrees. The basic deep sleep entry degree reflects the degree of proximity between the current vital sign parameter and the learned vital sign parameter. The basic deep sleep entry degrees for each vital sign parameter are then averaged to obtain a comprehensive deep sleep entry degree. This value represents the degree of proximity between the user's current overall sleep state and a deep sleep state. The higher the similarity, the higher the deep sleep entry degree, indicating that the user is currently close to a deep sleep state. To assess the reliability of the deep sleep entry degree, the reliability of the deep sleep entry analysis is also calculated. Specifically, a first basic deep sleep entry level is selected from multiple basic deep sleep entry levels as the initial comparison object. This first basic deep sleep entry level is then subjected to deviation analysis with the remaining basic deep sleep entry levels. This process is repeated for all basic deep sleep entry levels, resulting in multiple deviation analysis results. Based on these deviation analysis results, the entry reliability corresponding to the current deep sleep entry level is calculated to measure the accuracy and stability of the deep sleep entry level.

[0021] Furthermore, the present application provides a method for performing deep sleep entry analysis based on the physical sign parameters in combination with multiple learning physical sign parameters in the deep sleep pattern parameter model to obtain multiple basic deep sleep entry degrees, including: Calculating similarities between the physical sign parameter and multiple learning physical sign parameters in the deep sleep pattern parameter model to obtain multiple physical sign parameter similarities as multiple basic deep sleep entry degrees; and calculating an average of the multiple basic deep sleep entry degrees as the deep sleep entry degree.

[0022] Optionally, in the processing layer of the deep sleep mode parameter model, the physical sign parameter is compared with a plurality of learning physical sign parameters of the user. Each learning physical sign parameter represents a typical physiological characteristic of the user in a deep sleep state. For each physical sign parameter, the Euclidean distance is used to calculate the physical sign parameter. Calculate its similarity with the corresponding learned vital sign parameter, where x is the collected vital sign parameter and y is the learned vital sign parameter. Repeat this process to calculate a corresponding vital sign parameter similarity for each vital sign parameter. This vital sign parameter similarity is the corresponding basic deep sleep entry degree. Subsequently, calculate the average of all basic deep sleep entry degrees to obtain the current overall deep sleep entry degree. This value can be used to determine the degree of match between the user's current vital sign state and the ideal deep sleep state. The higher the deep sleep entry degree, the closer the user is to the ideal deep sleep state.

[0023] Furthermore, the present application provides a reliability calculation for deep sleep entry analysis based on the multiple basic deep sleep entry degrees, including: A first basic deep sleep entry degree is selected from the multiple basic deep sleep entry degrees, and several basic deep sleep entry degrees are randomly selected and the average is calculated as the first comparative deep sleep entry degree; the difference between the first basic deep sleep entry degree and the first comparative deep sleep entry degree is calculated as the first analysis deviation; the calculation is continued to obtain multiple analysis deviations, and the average of the multiple analysis deviations is calculated to obtain the average analysis deviation; and the entry reliability is obtained by subtracting the average analysis deviation from 1.

[0024] Optionally, from multiple calculated basic deep sleep entry levels, a random selection of basic deep sleep entry levels is used as the first basic deep sleep entry level. Several (e.g., 3 to 5) random selections are then made from the other basic deep sleep entry levels. The mean of these randomly selected basic deep sleep entry levels is calculated and used as the first comparative deep sleep entry level. This mean represents the comprehensive similarity calculated based on multiple vital sign parameters and is used to compare the differences between the first basic deep sleep entry level and the first comparative deep sleep entry level. Subsequently, the difference between the first basic deep sleep entry level and the first comparative deep sleep entry level is calculated to obtain a first analysis deviation. A larger first analysis deviation indicates a greater difference between the similarity of the current vital sign parameter and the average similarity of other standard vital sign parameters, indicating lower stability, or reliability, of the current deep sleep entry level analysis. Repeat the above steps, selecting different basic deep sleep entry levels as the first basic deep sleep entry level for each of the multiple basic deep sleep entry levels. Several more random selections are then made, and the deviation between each selected basic deep sleep entry level and its corresponding comparative deep sleep entry level is calculated. Through multiple calculations, multiple analysis deviations are obtained, and these deviations are used for the subsequent average deviation calculation. After that, all calculated analysis deviations are averaged to obtain the average analysis deviation. This average analysis deviation is subtracted from 1 to obtain the entry reliability. The smaller the average analysis deviation, the smaller the difference between the current basic deep sleep entry levels and the learned vital sign parameters. The reliability of the currently analyzed deep sleep entry level is higher, and the entry reliability is also higher. Conversely, if the deviation is large, it means that the similarity between different vital sign parameters is large, the analysis reliability of the current deep sleep entry level is low, and the entry reliability is also low.

[0025] When the deep sleep entry degree is less than the entry degree threshold, a sleep adjustment step is set according to the entry reliability, and sleep environment parameters are adjusted and optimized within the sleep environment parameter space until the deep sleep entry degree is greater than or equal to the entry degree threshold.

[0026] In one embodiment, the calculated deep sleep entry level is monitored in real time and compared with a preset entry level threshold (e.g., 75%). If the deep sleep entry level is greater than or equal to the threshold, it indicates that the user has entered a deep sleep state, and the adjustment process ends. If the deep sleep entry level is less than the threshold, the sleep environment parameters need to be optimized to help the user enter a deep sleep state. When adjusting the sleep environment parameters, the adjustment step size depends on the reliability of the deep sleep entry level. The adjustment step size controls the amplitude of the environmental parameter adjustments. When the entry reliability is high, indicating that the current deep sleep entry level analysis is more reliable, the adjustment step size is smaller, meaning the environmental parameter adjustments are smaller and more precise. The goal is to fine-tune the environmental parameters to ensure accurate optimization of the user's sleep state. When the entry reliability is low, indicating that the current deep sleep entry level analysis may have significant errors, the adjustment step size is larger, meaning the environmental parameter adjustments are larger. The goal is to quickly optimize the environment to help the user enter a deep sleep state more quickly. Subsequently, the user's sleep environment parameters, including but not limited to ambient temperature, light intensity, humidity, or noise, are adjusted according to the set adjustment step size. Regarding ambient temperature, if the user's body temperature is detected to be high or low, the room temperature can be adjusted to help the user enter deep sleep. For example, a smaller step size results in smaller temperature changes, maintaining accurate regulation; a larger step size results in larger temperature adjustments, which can quickly affect the user's body temperature. Regarding light intensity, the user's sleeping environment can be optimized by adjusting the indoor lighting intensity. For example, lowering the light intensity can help the user relax and gradually enter deep sleep. After adjusting environmental parameters, the user's vital signs are continuously monitored, the deep sleep entry level is recalculated, and the new entry reliability is evaluated. Based on the latest deep sleep entry level, it is determined whether the preset threshold has been reached. If the deep sleep entry level is greater than or equal to the threshold, the user has entered deep sleep, and the optimization adjustment ends. If the deep sleep entry level remains below the threshold, the environmental parameter optimization continues, and the adjustment step size is updated based on the new entry reliability, until the user enters deep sleep.

[0027] Furthermore, the present application provides setting a sleep adjustment step size according to the entry reliability, and optimizing the adjustment of sleep environment parameters within the sleep environment parameter space, including: Based on the user's sleep analysis data over a historical period, a historical average entry reliability is calculated; based on the ratio of the historical average entry reliability to the entry reliability, a preset sleep adjustment step size is corrected and calculated to obtain a sleep adjustment step size; a sleep environment parameter space is obtained; using the sleep adjustment step size, sleep environment parameters are adjusted and optimized within the sleep environment parameter space, and after the adjustment, the user's vital sign parameters are tested and a deep sleep entry degree is calculated until the deep sleep entry degree is greater than or equal to the entry degree threshold value.

[0028] Preferably, the user's historical sleep entry reliability for different sleep stages is calculated based on historical sleep analysis data. This historical entry reliability reflects the user's deep sleep entry reliability over multiple past sleep sessions. The specific calculation process involves extracting the entry reliability for each deep sleep stage from the user's historical sleep data, averaging these historical entry reliability values, and calculating the historical average entry reliability. This serves as the basis for adjusting the step size during the current adjustment process. Subsequently, the ratio between the current entry reliability and the historical average entry reliability is multiplied by the preset sleep adjustment step size to complete the adjustment. This adjustment method allows for dynamic adjustment of the adjustment step size to accommodate varying entry reliability conditions. Before performing adjustment optimization, the current sleep environment parameter space, including temperature, humidity, light, noise, etc., is obtained. In this step, the current values of these environmental parameters are read to form an adjustable sleep environment parameter space, providing a basis for subsequent adjustment optimization. The sleep environment parameters are then adjusted based on the adjusted sleep adjustment step size. For example, if the current ambient temperature is lower than the optimal sleep temperature, the temperature will be gradually adjusted based on the corrected step size to ensure that the temperature optimization can help the user enter a deep sleep state faster. After each adjustment of the environmental parameters, the user's vital sign parameters will be retested, and the user's deep sleep entry level will be calculated in real time. If the deep sleep entry level is greater than or equal to the entry level threshold, it means that the user has entered a deep sleep state and the adjustment optimization is complete. If the deep sleep entry level is still lower than the threshold, the environmental parameters will continue to be optimized using the corrected step size until the deep sleep entry level reaches or exceeds the threshold. Through this dynamic adjustment cycle, it can ensure that the user enters a deep sleep state in a short period of time, and through continuous optimization of environmental parameters, it helps the user maintain deep sleep, thereby improving the user's overall sleep quality. This method not only improves the accuracy and adaptability of the adjustment, but also ensures the stability and continuity of deep sleep, making sleep adjustment more efficient and intelligent.

[0029] Furthermore, the present application provides a method for using the sleep adjustment step size to adjust and optimize the sleep environment parameters within the sleep environment parameter space, and after the adjustment, testing the user's physical sign parameters and calculating the deep sleep entry degree, including: Collect current initial sleep environment parameters; within the sleep environment parameter space, use the sleep adjustment step to adjust the initial sleep environment parameters to obtain first sleep environment parameters; adjust the sleep environment parameters according to the first sleep environment parameters, detect the user's vital sign parameters, and perform deep sleep analysis in combination with the deep sleep mode parameter model to obtain a first deep sleep entry degree and a first entry reliability; determine whether the first deep sleep entry degree is greater than or equal to the entry degree threshold value, and if so, complete the adjustment optimization; if not, update the sleep adjustment step according to the first entry reliability, and continue to adjust and optimize the first sleep environment parameters; until the deep sleep entry degree is greater than or equal to the entry degree threshold value, the adjustment optimization of the sleep environment parameters is completed.

[0030] Optionally, after the user falls asleep, the current sleep environment parameters are collected as initial sleep environment parameters. These parameters will serve as the starting point for subsequent adjustment and optimization. Based on the current initial sleep environment parameters and the corrected sleep adjustment step size, the environment parameters are initially adjusted to obtain first sleep environment parameters. After the sleep environment parameter adjustment is complete, the user's vital signs are detected and input into the deep sleep mode parameter model. Using the same deep sleep analysis described above, a first deep sleep entry degree and a first entry reliability are calculated. The first deep sleep entry degree indicates how close the user's current sleep state is to a deep sleep state after adjustment, while the first entry reliability indicates the reliability of the analysis of the adjusted current deep sleep entry degree. The calculated first deep sleep entry degree is then compared with a set entry degree threshold. If the first deep sleep entry degree is greater than or equal to the entry degree threshold, it indicates that the user has successfully entered a deep sleep state, completing the adjustment and optimization, and no further adjustments are required. If the first deep sleep entry degree is less than the entry degree threshold, it indicates that the user has not yet entered the ideal deep sleep state and further optimization of the environment parameters is required. At this point, the sleep adjustment step size is updated using the first sleep entry reliability according to the same sleep adjustment step size adjustment method as described above, and the first sleep environment parameter adjustment and optimization are continued. After each adjustment, the user's vital sign parameters are retested and a new deep sleep entry level is calculated until the deep sleep entry level reaches or exceeds the threshold value, completing the entire sleep environment parameter adjustment and optimization process to ensure the user's deep sleep quality.

[0031] Continue adjusting the sleep environment parameters until the deep sleep time reaches the deep sleep time threshold value, and the sleep adjustment is completed.

[0032] In one embodiment, after a user's deep sleep entry level reaches a threshold and they enter a deep sleep state, the user's deep sleep duration is monitored in real time. During this time, the user's vital signs are continuously monitored to ensure that their deep sleep entry level remains above the deep sleep time threshold. This deep sleep time threshold is the user's target deep sleep duration, representing the minimum duration the user needs to maintain in deep sleep to ensure adequate recovery and rest. By continuously monitoring the user's deep sleep entry level, the duration the user maintains in deep sleep can be calculated and a determination can be made as to whether the user's deep sleep time has reached a preset deep sleep time threshold. If the deep sleep time reaches or exceeds the threshold, the user has achieved sufficient deep sleep, and the adjustment process is complete. Active adjustment of environmental parameters is no longer required, and sleep adjustment ends. If the deep sleep time has not yet reached the threshold, the current environmental parameters are maintained or fine-tuned to ensure the user remains in a deep sleep state until the accumulated deep sleep time reaches or exceeds the preset threshold. The deep sleep adjustment process is considered successfully completed when the user's duration in deep sleep reaches or exceeds the preset time threshold. At this time, active adjustment of environmental parameters will stop, but the user's vital signs will continue to be monitored to maintain their sleep quality until the user wakes up naturally.

[0033] Furthermore, the present application provides for continuing to adjust the sleep environment parameters until the deep sleep time reaches the deep sleep time threshold, including Obtaining a deep sleep time threshold value of the user; and continuing to adjust and optimize the sleep environment parameters until the deep sleep entry degree is greater than or equal to the entry degree threshold value and the duration is greater than or equal to the deep sleep time threshold value.

[0034] Preferably, after the user falls asleep, a set deep sleep time threshold is obtained. This threshold is typically the minimum time the user needs to remain in deep sleep, such as two or three hours. Subsequently, the user's sleeping environment is optimized. Even if the user's deep sleep entry level has reached or exceeded the threshold, environmental parameters will continue to be optimized to ensure the user remains in deep sleep for a prolonged period. During this process, the user's vital signs are continuously monitored, and the real-time deep sleep entry level is calculated. When the user's deep sleep entry level remains consistently greater than or equal to the preset threshold, a timer begins to monitor the user's total duration in deep sleep. If the entry level decreases during this process, environmental parameters are further adjusted to help the user return to a deep sleep state. Once the deep sleep duration reaches the set time threshold, for example, if the user has accumulated two hours of deep sleep, active adjustments are terminated. At this point, environmental parameter adjustments cease, signaling the completion of deep sleep optimization. However, the user's vital signs will continue to be monitored to ensure a smooth transition to subsequent sleep stages. The entire process helps users stay in a deep sleep state long enough by continuously optimizing environmental parameters, thereby effectively improving sleep quality and ensuring that users get adequate rest.

[0035] In summary, the embodiments of the present application have at least the following technical effects: The embodiment of the present application constructs a deep sleep pattern parameter model by learning the user's deep sleep test data. During the sleep process, the user's physical sign parameters are detected in real time and combined with the model to perform deep sleep analysis, and the deep sleep entry degree and entry reliability are calculated. When the entry degree is lower than the threshold value, the adjustment step size is dynamically set according to the reliability, and the sleep environment parameters are optimized until the user enters deep sleep and maintains it for a certain period of time. By continuously adjusting the sleep environment parameters, it is ensured that the user reaches and maintains the preset deep sleep time, and ultimately achieves intelligent optimization of sleep quality. These technical effects jointly solve the technical problem of being unable to adjust the sleep environment according to the user's status in real time, and achieve the technical effect of automatically adjusting the sleep environment parameters and improving the quality and duration of deep sleep by learning and analyzing the user's deep sleep pattern.

[0036] It should be noted that the order in which the embodiments of the present application are presented is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. Furthermore, the foregoing descriptions of specific embodiments of this specification are provided. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential sequence shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0037] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

[0038] This specification and drawings are merely illustrative of the present application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Obviously, those skilled in the art may make various modifications and variations to this application without departing from the scope of this application. Thus, this application is intended to include such modifications and variations as fall within the scope of this application and its equivalents.

Claims

1. A sleep regulation method combining sleep pattern learning and optimization, characterized in that: The method comprises: Based on the user's deep sleep test data, sleep sign parameters in deep sleep mode are learned to obtain a deep sleep mode parameter model; During the user's sleep, the user's physical parameters are detected, and deep sleep analysis is performed in combination with the deep sleep pattern parameter model to obtain a deep sleep entry degree and entry reliability; When the deep sleep entry degree is less than the entry degree threshold, setting a sleep adjustment step size according to the entry reliability, and adjusting and optimizing the sleep environment parameters within the sleep environment parameter space until the deep sleep entry degree is greater than or equal to the entry degree threshold; Continue adjusting the sleep environment parameters until the deep sleep time reaches the deep sleep time threshold value, and the sleep adjustment is completed.

2. The sleep regulation method combining sleep pattern learning and optimization according to claim 1, characterized in that: Based on the user's deep sleep test data, sleep sign parameters in deep sleep mode are learned to obtain a deep sleep mode parameter model, including: Based on the deep sleep test record data of the user in the historical time, the vital sign parameters during deep sleep are collected to obtain multiple sample vital sign parameter sets; Calculating and obtaining a plurality of sample average physical sign parameters according to the plurality of sample physical sign parameter sets as a plurality of learning physical sign parameters; The multiple learned vital sign parameters are integrated as a deep sleep pattern parameter model.

3. The sleep regulation method combining sleep pattern learning and optimization according to claim 2, characterized in that: During the user's sleep, the user's physical parameters are detected and combined with the deep sleep pattern parameter model to perform deep sleep analysis, including: During the user's sleep, detecting the user's vital signs parameters; Performing a deep sleep entry analysis based on the physical sign parameters and in combination with a plurality of learned physical sign parameters in the deep sleep pattern parameter model to obtain a plurality of basic deep sleep entry degrees, and calculating the deep sleep entry degree; The reliability of the deep sleep entry analysis is calculated based on the multiple basic deep sleep entry degrees to obtain the entry reliability.

4. The sleep regulation method combining sleep pattern learning and optimization according to claim 3, characterized in that: According to the physical sign parameters, combined with multiple learning physical sign parameters in the deep sleep pattern parameter model, deep sleep entry analysis is performed to obtain multiple basic deep sleep entry degrees, including: calculating similarities between the physical sign parameter and a plurality of learned physical sign parameters in the deep sleep pattern parameter model, and obtaining a plurality of physical sign parameter similarities as a plurality of basic deep sleep entry degrees; An average of the multiple basic deep sleep entry levels is calculated as the deep sleep entry level.

5. The sleep regulation method combining sleep pattern learning and optimization according to claim 3, characterized in that: Calculating the reliability of deep sleep entry analysis based on the multiple basic deep sleep entry degrees includes: Selecting a first basic deep sleep entry degree from the multiple basic deep sleep entry degrees, and randomly selecting a plurality of basic deep sleep entry degrees and calculating an average thereof as a first comparison deep sleep entry degree; calculating a difference between the first basic deep sleep entry degree and the first comparison deep sleep entry degree as a first analysis deviation; Continuing to calculate to obtain multiple analysis deviations, calculating the average of the multiple analysis deviations to obtain an average analysis deviation; The entry reliability was obtained by subtracting the mean analytical deviation from 1.

6. The sleep regulation method combining sleep pattern learning and optimization according to claim 1, characterized in that: According to the entry reliability, a sleep adjustment step is set, and adjustment optimization of the sleep environment parameters is performed within the sleep environment parameter space, including: Calculate the historical average entry reliability based on the user's historical sleep analysis data; performing correction calculation on a preset sleep adjustment step length according to the ratio of the historical average entry reliability to the entry reliability to obtain a sleep adjustment step length; Get the sleep environment parameter space; The sleep adjustment step is used to adjust and optimize the sleep environment parameters in the sleep environment parameter space, and after the adjustment, the user's vital sign parameters are tested and the deep sleep entry degree is calculated until the deep sleep entry degree is greater than or equal to the entry degree threshold.

7. The sleep regulation method combining sleep pattern learning and optimization according to claim 6, characterized in that: The sleep adjustment step is used to adjust and optimize the sleep environment parameters within the sleep environment parameter space, and after the adjustment, the user's physical sign parameters are tested and the deep sleep entry degree is calculated, including: Collect the current initial sleeping environment parameters; In the sleep environment parameter space, the initial sleep environment parameter is adjusted using the sleep adjustment step to obtain a first sleep environment parameter; adjusting the sleep environment parameters according to the first sleep environment parameters, detecting the user's physical parameters, and performing deep sleep analysis in combination with the deep sleep pattern parameter model to obtain a first deep sleep entry degree and a first entry reliability; determining whether the first deep sleep entry degree is greater than or equal to the entry degree threshold; if so, completing the adjustment optimization; if not, updating the sleep adjustment step size according to the first entry reliability, and continuing to adjust and optimize the first sleep environment parameter; Until the deep sleep entry degree is greater than or equal to the entry degree threshold, the adjustment and optimization of the sleep environment parameters are completed.

8. The sleep regulation method combining sleep pattern learning and optimization according to claim 1, characterized in that: Continue adjusting the sleep environment parameters until the deep sleep time reaches the deep sleep time threshold, including: Get the user's deep sleep time threshold; The sleep environment parameters are adjusted and optimized continuously until the duration during which the deep sleep entry degree is greater than or equal to the entry degree threshold value is greater than or equal to the deep sleep time threshold value.

Images