Intelligent wake-up method, apparatus, system, medium, and device

Abstract

An intelligent wake-up method, device, system, medium and equipment. The method comprises: determining a non-deep sleep period closest to a preset wake-up time; generating a wake-up signal in the determined non-deep sleep period for wake-up. The above scheme can accurately determine the wake-up time to improve the wake-up effect.

Description

Technical Field

[0001] This invention relates to the field of smart device technology, specifically to a smart wake-up method, device, system, medium, and equipment. Background Technology

[0002] With the continuous development of smart wearable device technology, wearable devices (such as wristbands) have been widely used in the field of sleep monitoring and wake-up, providing users with more convenient and intelligent sleep management and wake-up services.

[0003] Currently, most wearable devices on the market determine wake-up time primarily through motion monitoring technology. Specifically, after the user sets a desired wake-up time period, the wearable device can monitor the frequency of the user's body movements to determine whether the user is in a relatively active or light sleep state. Once the conditions are met, wake-up is triggered.

[0004] However, the aforementioned wearable devices rely solely on the frequency of body movements to determine the wake-up timing, resulting in poor wake-up performance and failing to fully meet users' needs for comfortable and accurate wake-up. Summary of the Invention

[0005] The problem this invention aims to solve is: how to accurately determine the wake-up timing in order to improve the wake-up effect.

[0006] To address the above problems, embodiments of the present invention provide an intelligent wake-up method, the method comprising: Determine the non-deep sleep period closest to the preset wake-up time; A wake-up signal is generated during the defined non-deep sleep period for use in waking.

[0007] In one possible embodiment, determining the non-deep sleep period closest to the preset wake-up time includes: Real-time acquisition of monitoring data for wake-up parameters; Based on the collected wake-up parameter monitoring data, the current sleep stage is identified to determine the non-deep sleep stage closest to the preset wake-up time.

[0008] In one possible embodiment, determining the non-deep sleep period closest to the preset wake-up time includes: Obtain the phase prediction results of the circadian rhythm phase prediction model; Using the phase prediction results, the non-deep sleep period closest to the preset wake-up time is determined.

[0009] In one possible embodiment, determining the non-deep sleep period closest to a preset wake-up time using the phase prediction result includes: Determine whether the phase prediction result satisfies the wake-up decision constraint; When the phase prediction result satisfies the wake-up decision constraint, the non-deep sleep period closest to the preset wake-up time is obtained based on the phase prediction result.

[0010] In one possible embodiment, determining the non-deep sleep period closest to the preset wake-up time based on the phase prediction result further includes: When the phase prediction result does not meet the wake-up decision constraints, the monitoring data based on the wake-up parameters identifies the non-deep sleep period closest to the preset wake-up time.

[0011] In one possible embodiment, the wake-up decision constraint includes: The matching error between the phase prediction result and the sleep period identification result obtained from the monitoring data based on the wake-up parameters is within a preset error threshold.

[0012] In one possible embodiment, determining the non-deep sleep period closest to a preset wake-up time using the phase prediction result includes: Determine whether the actual application duration of the circadian rhythm phase prediction model has reached the preset duration; When the actual application duration of the circadian rhythm phase prediction model reaches the preset duration, the non-deep sleep period closest to the preset wake-up time is obtained based on the phase prediction results.

[0013] In one possible embodiment, the method further includes: Collect user wake-up status data; The model parameters of the circadian rhythm phase prediction model are optimized based on the collected wake-up state data.

[0014] In one possible embodiment, the wake-up parameters include: heart rate parameters and body movement parameters.

[0015] In one possible embodiment, the non-deep sleep period is the REM sleep period or light sleep period.

[0016] In one possible embodiment, generating a wake-up signal during a determined non-deep sleep period includes: generating a wake-up signal during the light sleep period closest to the preset wake-up time when a light sleep period closest to the preset wake-up time is identified; and generating a wake-up signal during the REM period closest to the preset wake-up time when a light sleep period closest to the preset wake-up time is not identified but a REM period closest to the preset wake-up time is identified.

[0017] In one possible embodiment, the method further includes: Send the wake-up signal.

[0018] This invention also provides an intelligent wake-up device, the device comprising: The determination unit is suitable for identifying the non-deep sleep period closest to the preset wake-up time; A wake-up unit is adapted to generate a wake-up signal during a defined non-deep sleep period for use in waking up.

[0019] A smart wake-up system, comprising: The aforementioned intelligent wake-up device; And wake up the execution terminal; The wake-up execution terminal is connected to the intelligent wake-up device and is used to receive the wake-up signal generated by the intelligent wake-up device and perform the wake-up operation.

[0020] In one possible embodiment, the wake-up execution terminal includes: a wake-up light, which is used to determine the corresponding wake-up illumination parameters according to the wake-up signal and output the corresponding wake-up light.

[0021] In one possible embodiment, the wake-up light includes: The data acquisition unit is used to collect the user's wake-up status data; An adjustment unit is used to adjust the wake-up illumination parameters using the wake-up state data.

[0022] This invention also provides a computer-readable storage medium having a computer program stored thereon, the computer program being executed by a processor to implement the steps of any of the methods described above.

[0023] This invention also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program that can run on the processor, and the processor executes the steps of any of the methods described above when running the computer program.

[0024] Compared with the prior art, the technical solution of the embodiments of the present invention has the following advantages: When waking a user using the solution of this invention, the nearest non-deep sleep period to the preset wake-up time is first determined, and then a wake-up signal is generated within the determined non-deep sleep period. Directly determining the nearest non-deep sleep period to the preset wake-up time not only limits the wake-up time range, better aligns with the user's sleep schedule, narrows the sleep stage judgment interval, reduces interference errors from body movement signals, and simultaneously considers both wake-up comfort and timing accuracy. Therefore, compared to determining the wake-up timing solely based on the frequency of body movement, the wake-up time determined by the solution of this invention is more precise, thereby improving the wake-up effect and meeting the user's needs for comfortable and precise wake-up. Attached Figure Description

[0025] Figure 1This is a flowchart of an intelligent wake-up method according to an embodiment of the present invention; Figure 2 This is a schematic diagram of a sleep period according to an embodiment of the present invention; Figure 3 This is a flowchart of a method for determining the non-deep sleep period closest to a preset wake-up time in an embodiment of the present invention; Figure 4 This is a flowchart of another method for determining the non-deep sleep period closest to a preset wake-up time in an embodiment of the present invention; Figure 5 This is a flowchart of another method for determining the non-deep sleep period closest to a preset wake-up time in an embodiment of the present invention; Figure 6 This is a schematic diagram of the structure of an intelligent wake-up device according to an embodiment of the present invention; Figure 7 This is an exploded view of a smart bracelet according to an embodiment of the present invention; Figure 8 This is a schematic diagram of the structure of an intelligent wake-up system according to an embodiment of the present invention; Figure 9 This is an exploded structural diagram of a smart lamp according to an embodiment of the present invention. Detailed Implementation

[0026] Traditional wearable devices force users to wake up at a fixed time. If the user happens to be in the REM period, this sudden interruption can easily lead to "sleep inertia". After waking up, the user may feel groggy, tired, depressed, or even have a headache, which seriously affects their mental state and work efficiency during the day.

[0027] As a result, smart wearable devices have emerged on the market, which mainly use motion monitoring technology to determine the wake-up time.

[0028] Specifically, after the user sets a desired wake-up time period, the wearable device continuously monitors the frequency of the user's body movements throughout the night using its built-in accelerometer. Once the user-defined wake-up time period begins, the wearable device starts analyzing the user's activity level. Once the judgment conditions are met, the wearable device sends a trigger signal to the smart light fixture. After receiving the signal, the smart light fixture gradually brightens its light from dim to moderate over 20 to 30 minutes by simulating the sunrise process, gently waking the user.

[0029] However, the above-mentioned wake-up methods for wearable devices have obvious shortcomings, resulting in inaccurate judgment of wake-up timing and unstable wake-up effects.

[0030] Specifically, the wake-up time is determined based on the frequency of the user's body movements at night, but there is no strict one-to-one correspondence between the frequency of body movements and the user's sleep period.

[0031] After the N1 sleep stage, modern sleep science divides adult nighttime sleep into multiple cycles (approximately 90 minutes per cycle). Each cycle includes non-rapid eye movement (NREM) sleep (divided into light sleep N2 and deep sleep N3) and rapid eye movement (REM) sleep. N3 is deep sleep, making awakening difficult. REM and N2 stages show brain activity close to wakefulness, making awakening easier and resulting in minimal discomfort upon waking.

[0032] During REM sleep, users may experience rapid eye movements but remain still. In this case, the wearable devices mentioned above may miss the optimal wake-up time because they cannot detect significant body movement. In N3 sleep, users may experience brief body movements. At this time, the wearable devices may misinterpret these movements as light sleep or a relatively active state, thus triggering a wake-up during the user's deep sleep period. This can lead to discomfort such as drowsiness and irritability after the user is forcibly awakened.

[0033] Therefore, the wearable devices mentioned above cannot accurately determine the wake-up time, resulting in poor wake-up effect and instability, and cannot fully meet users' needs for comfortable and accurate wake-up.

[0034] To address this problem, the present invention provides an intelligent wake-up method. By directly determining the nearest non-deep sleep period to the preset wake-up time, the user can be woken up more accurately, thereby improving the wake-up effect.

[0035] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0036] Reference Figure 1 This invention provides a smart wake-up method, which may include the following steps: Step 11: Determine the non-deep sleep period closest to the preset wake-up time.

[0037] Step 12: Generate a wake-up signal during the determined non-deep sleep period for use in waking up.

[0038] Steps 11 and 12 are described in detail below: In practice, the preset wake-up time can be set by the user, for example, through an interactive interface. The preset wake-up time can also be a default time.

[0039] After obtaining the preset wake-up time, the nearest non-deep sleep period can be determined. Sleep can be divided into multiple cycles (approximately 90 minutes per cycle), each cycle including light sleep (N2), deep sleep (N3), and REM sleep. The typical sequence within each sleep cycle is: N2→N3→N2→REM.

[0040] For example, refer to Figure 2 The preset wake-up time is t0. After a user falls asleep, there can be four sleep cycles: the first sleep cycle T1, the second sleep cycle T2, the third sleep cycle T3, and the fourth sleep cycle T4. Each sleep cycle T1 can include a light sleep stage N2, a deep sleep stage N3, and a REM sleep stage. The duration of the same sleep stage within each sleep cycle can be the same or different. For example, the duration of stage N2 within the first sleep cycle T1 can be longer than the duration of stage N2 within the second sleep cycle T2.

[0041] The so-called identification of the non-deep sleep period closest to the preset wake-up time refers to identifying the nearest N2 or REM stage. A wake-up signal can then be generated during the nearest N2 or REM stage to wake the user. Compared to N1 and N3 stages, N2 and REM stages are considered easily awakened stages. N2 stages offer better physical comfort and faster cognitive recovery after wakefulness, while REM stages, as alternative identification stages, can also achieve a smooth wake-up at the end of a sleep cycle not occupied by deep sleep.

[0042] Preferably, the so-called determination of the non-deep sleep period closest to the preset wake-up time refers to determining the N2 stage closest to the preset wake-up time. Therefore, a wake-up signal can be generated in the N2 stage closest to the preset wake-up time to wake up the user. The N2 stage belongs to the light sleep stage. When the user is woken up, the body's muscle tension is well restored, the sleep inertia is low, the awakening level is the highest after waking up, and the emotions are stable, resulting in a better wake-up state.

[0043] In practice, when a wake-up signal is generated during a defined non-deep sleep period, it may include: When a light sleep period closest to the preset wake-up time is detected, a wake-up signal is generated within that light sleep period. When the light sleep period closest to the preset wake-up time is not detected, but the REM period closest to the preset wake-up time is detected, a wake-up signal is generated within the REM period closest to the preset wake-up time.

[0044] In other words, during actual wake-up, a wake-up signal can be generated preferentially within the N2 period closest to the preset wake-up time. Only if no N2 period closest to the preset wake-up time is identified will a wake-up signal be generated within the REM period closest to the preset wake-up time. If neither the N2 nor the REM period closest to the preset wake-up time is identified, the user can be woken up when the preset wake-up time arrives.

[0045] In practice, various methods can be used to determine the non-deep sleep period closest to the preset wake-up time, and no restrictions are imposed here.

[0046] In one embodiment of the present invention, reference is made to... Figure 3 The steps for determining the non-deep sleep period closest to the preset wake-up time may include: Step 31: Collect real-time monitoring data of wake-up parameters; Step 32: Based on the collected wake-up parameter monitoring data, identify the current sleep stage to determine the non-deep sleep stage closest to the preset wake-up time.

[0047] By collecting real-time monitoring data of wake-up parameters, and then identifying the current sleep cycle and which sleep stage of the current sleep cycle based on the collected wake-up parameter monitoring data, it is possible to determine the non-deep sleep stage closest to the preset wake-up time.

[0048] For ease of description, in the embodiments of the present invention, the scheme of identifying sleep stages by collecting monitoring data of wake-up parameters will be referred to as sleep stage identification.

[0049] In practice, a triaxial accelerometer can be used to collect monitoring data of the user's body movement, and a photoelectric heart rate sensor (PPG) can be used to collect monitoring data of the user's heart rate. The user's current sleep stage can then be identified by combining the monitoring data of the user's body movement and the monitoring data of the user's heart rate.

[0050] Specifically, by using motion monitoring data, it is possible to determine whether a user is in a sleep or exercise state. By using heart rate level and heart rate variability, it is possible to distinguish different sleep stages when the user is in a sleep state. By combining this with the sleep cycle pattern, the staged results can be corrected to achieve accurate sleep stage identification.

[0051] For example, refer to Figure 2 Based on real-time monitoring data of wake-up parameters, it can be determined that the current stage is N1 of the second sleep cycle T2, which is not the non-deep sleep stage closest to the preset wake-up time t0. Therefore, no wake-up signal needs to be generated. A wake-up signal will only be generated when the current stage is REM or N2 of the fourth sleep cycle T4.

[0052] In another embodiment, reference is made to Figure 4 The steps for determining the non-deep sleep period closest to the preset wake-up time may include: Step 41: Obtain the phase prediction results of the circadian rhythm phase prediction model; Step 42: Using the phase prediction results, determine the non-deep sleep period closest to the preset wake-up time.

[0053] The circadian rhythm phase prediction model can be implemented using various machine learning models. The input data for this model is the monitoring data of wake-up parameters. It can analyze this monitoring data and output phase prediction results.

[0054] In this embodiment, a two-layer Long Short-Term Memory (LSTM) network is used as the circadian rhythm phase prediction model. The specific model structure and parameter settings are as follows: The input layer receives N-dimensional physiological time-series signals sampled continuously over 24 hours, covering features such as RR intervals (R-wave intervals), triaxial activity, ambient light intensity, and sampling frequency. The input data dimensions are time series length T and the number of features (T, F). The hidden layer uses a two-layer LSTM structure. The first layer has 128 neurons and is set to output the complete sequence (return_sequences=True) to fully extract continuous temporal features. The second layer has 64 neurons and is used to compress temporal features and mine deep correlation information.

[0055] After feature extraction, a random deactivation layer (Dropout) and a fully connected layer (Dense) are sequentially connected to complete feature mapping and output: First, a Dropout (0.3) random deactivation layer is set to suppress model overfitting with a deactivation probability of 0.3; then, a fully connected layer with 32 neurons and ReLU activation function is connected to achieve non-linear feature transformation; finally, a fully connected layer with 1 neuron and Linear activation function is connected to output the model prediction result, that is, the circadian rhythm phase angle in the range of 0~24 h, corresponding to the Dark Light Melatonin Onset (DLMO) offset.

[0056] The model was trained using an Adaptive Moment Estimation (Adam) optimizer with an initial learning rate of 1×10⁻⁶. -3 The batch size is 32, the number of training epochs is 300, and the mean square error (MSE) is used as the loss function.

[0057] The training data uses a domain-standard circadian rhythm dataset, which mainly includes two types: one is a measured physiological dataset, which collects 24-hour wrist movement, electrocardiogram (ECG), and eye light exposure data from multiple subjects, and uses DLMO calibrated by 6-hour intensive saliva sampling as a supervised learning label; the other is a publicly available transcriptome dataset, which selects circadian rhythm-related datasets from the Gene Expression Omnibus (GEO) database, and uses signal cosine periodicity as an unsupervised constraint.

[0058] The sample size in this study follows the standard settings of the field, with a total sample size of no less than 30 cases. All samples are stratified according to age and gender to ensure the balance of data distribution and the effectiveness of model training.

[0059] In practical applications, after a user uses the smart wake-up device, monitoring data of the user's wake-up parameters and the user's state data after wake-up can be collected in real time. Using the real-time collected data and the state data after wake-up, the initial phase prediction model is trained, thereby adjusting the model parameters of the initial phase prediction model, and finally obtaining a phase prediction model with wake-up timing accuracy that meets the requirements, which serves as a circadian rhythm phase prediction model.

[0060] In practice, the user's post-wake-up state data can include time out of bed, heart rate recovery rate after wake-up, and subjective feedback data input by the user. By continuously fitting the phase prediction results of the initial phase prediction model with the user's post-wake-up state data, the accuracy of the wake-up timing is confirmed, thereby continuously optimizing the model parameters.

[0061] In yet another embodiment of the invention, reference is made to... Figure 5 The steps for determining the non-deep sleep period closest to the preset wake-up time may include: Step 51: Obtain the phase prediction results of the circadian rhythm phase prediction model.

[0062] Step 52: Determine whether the phase prediction result meets the wake-up decision constraints.

[0063] In practical implementation, the accuracy of the circadian rhythm phase prediction model can be determined by judging whether the phase prediction results of the circadian rhythm phase prediction model meet the wake-up decision constraints.

[0064] In practice, the wake-up decision constraints can be set using various methods.

[0065] In one embodiment, the wake-up decision constraint can be that the matching error between the phase prediction result and the sleep period identification result obtained based on the monitoring data of the wake-up parameters is within a preset error threshold.

[0066] Specifically, the phase prediction results can be compared with... Figure 3 The sleep phase recognition results obtained in the illustrated embodiment are compared to calculate the matching error between the two. If the matching error between the phase prediction result and the sleep phase recognition result is within a preset error threshold, the accuracy of the phase prediction result is considered to meet the requirements, and step 53 is executed; otherwise, the accuracy of the phase prediction result is considered not to meet the requirements, and step 54 is executed.

[0067] In another embodiment, the wake-up decision constraint can be: whether the actual application duration of the circadian rhythm phase prediction model reaches a preset duration.

[0068] In practical applications, the actual usage time of the circadian rhythm phase prediction model is equivalent to the actual usage time of the smart wake-up device. As the actual usage time increases, the training data for the circadian rhythm phase prediction model also increases, naturally leading to higher accuracy in phase prediction. However, when the actual usage time of the smart wake-up device falls short of the preset duration, the training data for the circadian rhythm phase prediction model is limited, resulting in lower accuracy in phase prediction.

[0069] To ensure the accuracy of the circadian rhythm phase prediction model, after obtaining the phase prediction results, it can be first determined whether the actual application duration of the circadian rhythm phase prediction model has reached the preset duration. If the actual application duration of the circadian rhythm phase prediction model has reached the preset duration, then step 53 is executed; otherwise, step 54 is executed.

[0070] Step 53: Based on the phase prediction results, obtain the non-deep sleep period closest to the preset wake-up time.

[0071] In practical implementation, the circadian rhythm phase prediction model can predict a user's sleep stage at a future moment based on historical wakefulness parameter monitoring data and sleep cycle patterns, thereby outputting phase prediction results representing the sleep stages at multiple future moments. Based on these phase prediction results, the time range corresponding to the non-deep sleep stage closest to the preset wakefulness time can be clearly determined.

[0072] Step 54: Identify the non-deep sleep period closest to the preset wake-up time based on the monitoring data of the wake-up parameters.

[0073] In the specific implementation, when the phase prediction result does not meet the wake-up decision constraints, the non-deep sleep period closest to the preset wake-up time can be identified based on the monitoring data of the wake-up parameters. That is, by collecting the monitoring data of the wake-up parameters in real time, the current sleep cycle and which sleep period of the sleep cycle are determined, and then the non-deep sleep period closest to the preset wake-up time can be determined by using the distance from the current time to the preset wake-up time.

[0074] Step 55: Generate a wake-up signal based on the sleep period determination results.

[0075] The sleep stage determination result can be determined by phase prediction results or confirmed by a sleep stage identification scheme.

[0076] In some embodiments, the smart wake-up device can output the wake-up signal based on the wake-up signal through audio or screen illumination, vibration, or other means to wake up the user.

[0077] In some embodiments, the method may further include: Step 56: Send the wake-up signal to the wake-up execution terminal.

[0078] In practice, the wake-up execution terminal differs from the terminal where the smart wake-up device is located. This wake-up execution terminal can be at least one smart home device, such as a smart curtain, smart light fixture, or smart speaker. A smart curtain can wake the user by gradually opening the curtains to allow natural light to enter. A smart light fixture can wake the user by controlling the lighting components to emit light. A smart speaker can wake the user by controlling the sound-emitting components to play set music.

[0079] Understandably, in practical applications, the wake-up execution terminal can include multiple smart home devices mentioned above. For example, it can control the curtains to gradually open to introduce natural light, while simultaneously using lighting fixtures and progressive sound effects to create a multi-sensory wake-up environment.

[0080] In some embodiments, the wake-up signal may include wake-up time information. The wake-up execution terminal may perform a wake-up operation at the indicated wake-up time based on the wake-up time information.

[0081] In other embodiments, the wake-up signal may not include wake-up time information. Upon receiving the wake-up signal, the wake-up execution terminal immediately performs the wake-up operation.

[0082] In some embodiments, the wake-up signal may further include first ambient light detection data. This first ambient light detection data refers to the detection data of ambient light around the smart wake-up device. Therefore, the first ambient light detection data can be sent to the wake-up execution terminal simultaneously with the wake-up signal, enabling the wake-up execution terminal to perform the wake-up operation in conjunction with the first ambient light detection data.

[0083] To enable those skilled in the art to better understand and implement the present invention, the apparatus, system, electronic device and computer-readable storage medium corresponding to the above method are described in detail below.

[0084] Reference Figure 6 This invention provides an intelligent wake-up device 60, which may include a determining unit 61 and a wake-up unit 62. Wherein: The determining unit 61 is adapted to determine the non-deep sleep period closest to the preset wake-up time; The wake-up unit 62 is adapted to generate a wake-up signal during a defined non-deep sleep period for use in waking up.

[0085] The determination unit 61 and the wake-up unit 62 can be implemented in accordance with the description of the corresponding method steps above, and will not be repeated here.

[0086] In some embodiments, the smart wake-up device 60 may be a wearable device, including but not limited to a smart bracelet.

[0087] This invention also provides an exploded view of a smart bracelet 70. (Refer to...) Figure 7 The smart bracelet 70 may include a wearing component, a control component, a sensor component, and a display and interaction component, with each component working together to achieve sleep monitoring and smart wake-up functions.

[0088] Specifically, the wearable components are used to achieve structural encapsulation, waterproof protection, and secure wearing of the entire device, and mainly include: a wristband 701, a protective shell 702, an outer shell 703, and a base 704. The wristband 701 can be made of silicone, and the protective shell 702 can be made of aluminum alloy. The outer shell 703 and base 704 can be made of lightweight materials. The wristband 701 and protective shell 702 are integrally molded, and the surface of the wristband 701 has adjustment holes to accommodate different wearing sizes, improving wearing comfort and stability. The protective shell 702 wraps around the outer shell 703, providing structural strength and aesthetic appeal. The outer shell 703 and base 704 are assembled to form a closed cavity, providing installation space for internal hardware and simultaneously possessing IP68 waterproof performance, adaptable to various wearing scenarios.

[0089] The control component, the core computing and function scheduling unit of the entire machine, is integrated on the motherboard and may include: a main control chip 705, a communication module 706, a storage chip 707, and a power supply 708. The main control chip 705, a core processor, combines a preset wake-up time with sleep data collected by sensors to identify the sleep period and calculate the optimal wake-up time, thereby generating a wake-up signal. The communication module 706 can be a Bluetooth or other wireless communication module. The communication module 706 can establish a wireless connection with mobile terminals, wake-up execution terminals, etc., to achieve data synchronization and linkage control, such as sending a wake-up signal to the wake-up execution terminal. The storage chip 707 stores user sleep data, device configuration information, and historical records, ensuring data traceability and offline use. The power supply 708 can be a lithium battery to provide continuous power to the entire machine. A power button 709 can be located on the side of the casing 703, used for powering on / off and triggering quick functions.

[0090] The sensor components are used to collect physiological and environmental data, providing a perceptual basis for sleep staging and wakefulness determination. Specifically, they may include: an optical heart rate sensor 710, an accelerometer and gyroscope 711, an ambient light sensor 712, and a blood oxygen and temperature sensor 713. The optical heart rate sensor 710 uses optical detection to collect the user's heart rate data in real time. The accelerometer and gyroscope 711 detects changes in the user's body movement and posture, i.e., collects body movement parameter data. The ambient light sensor 712 senses the intensity of ambient light, thus collecting ambient light monitoring data. The blood oxygen and temperature sensor 713 monitors the user's blood oxygen concentration, body temperature, and ambient temperature, respectively, thus obtaining blood oxygen, body temperature, and ambient temperature data. The data from the sensor components can be sent to the main control chip 705, which then makes sleep stage decisions.

[0091] The display and interaction components are located on the front of the device and are used for information display and user interaction. Specifically, they may include a protective layer 714 and a touchscreen 715. The protective layer 714, as the outermost protective structure, is wear-resistant and scratch-resistant while ensuring light transmittance. The touchscreen 715, which can be an OLED touchscreen, is located below the protective layer 714 and is used to display sleep data, time, and the interactive interface. It supports touch operation, enabling user settings and function triggering.

[0092] The smart bracelet in this embodiment of the invention uses a main control chip 705 that can acquire richer physiological characteristic data based on multi-dimensional wake-up parameter monitoring data such as blood oxygen, heart rate, body temperature, and body movement. This allows for more accurate judgment of the sleep period. Compared with a single data point to determine the wake-up time, the accuracy of sleep period monitoring is higher. Consequently, when the user is awakened, the brain's physiological state is close to that of wakefulness, significantly reducing drowsiness and fatigue after waking up, improving daytime cognitive function, and significantly reducing the risk of false triggering in the N3 phase or missed triggering in the REM phase.

[0093] In other embodiments, the smart wake-up device can be installed on a cloud server. Data collected by wearable devices such as smart bracelets can be uploaded to the cloud server, where sleep phase prediction is performed. The phase prediction results are then sent to the wake-up execution terminal, thereby reducing the power consumption of the wearable device and making it lighter.

[0094] This invention also provides an intelligent wake-up system, which includes the intelligent wake-up device 60 described in the above embodiments and a wake-up execution terminal 80. The wake-up execution terminal 80 is connected to the intelligent wake-up device 60 and is used to receive the wake-up signal generated by the intelligent wake-up device 60 and perform a wake-up operation.

[0095] In specific implementations, the wake-up execution terminal 80 can be at least one smart home device, such as a smart curtain, smart light fixture, or smart speaker. A smart curtain can wake the user by gradually opening the curtains to allow natural light to enter. A smart light fixture can wake the user by controlling the lighting components to emit light. A smart speaker can wake the user by controlling the sound-emitting components to play set music.

[0096] This invention also provides an exploded structural diagram of a smart lighting fixture. (Refer to...) Figure 9 The smart lighting fixture may include: In one embodiment, reference is made to Figure 8 The wake-up execution terminal can be composed of three main parts: a protective shell component, an optical module, and a control module. Each part works together to achieve the function of soft wake-up light output and ambient light adjustment.

[0097] Specifically, the outer casing provides structural support, external protection, and a mounting base for the entire device, mainly including a backplate 801, an outer frame 802, and a protective cover 803. The backplate 801, made of aluminum alloy or other metal materials, serves as the mounting base for the equipment, providing excellent heat dissipation and structural strength to ensure stable operation of internal components. The outer frame 802, also made of aluminum alloy or other metal materials, wraps around the optical module and control module, forming the overall outline of the device and providing physical protection for the internal components. The protective cover 803 covers the light-emitting surface of the light source, possessing high light transmittance and scratch and wear resistance, ensuring uniform light transmission while also providing dust and water protection.

[0098] The optical module is the core unit for light generation and homogenization, used to output a soft and uniform wake-up light effect. From bottom to top, it includes: LED beads 804, reflector paper 805, light guide plate 806, and light homogenizing plate 807. The LED beads 804 can output four colors of light: red, green, blue, and white, allowing for flexible adjustment of color temperature and brightness to adapt to different wake-up scenarios. The reflector paper 805, located below the LED beads 804, reflects the downward-scattered light back to the light guide direction, improving light efficiency. The light guide plate 806 converts a point light source into a surface light source, ensuring uniform light diffusion. The light homogenizing plate 807 further softens the light passing through the light guide plate 806, eliminating light spots and differences in brightness, ensuring uniform and soft light output.

[0099] The control module is the core computing and signal interaction unit of the entire machine, mainly including the main control module 808, the driver module 809, the communication module 810, and the light signal analysis module 811. The main control module 808, as the core processor, is responsible for receiving the wake-up signal from the sleep monitoring device, scheduling the light effect output logic, and coordinating the adjustment of the lamp's brightness and color temperature. The driver module 809 provides stable power supply and dimming control for the LED beads 804, achieving precise adjustment of the light effect. The communication module 810 establishes a wireless connection with devices such as the intelligent wake-up device, receiving wake-up signals and control commands. The light signal analysis module 811 analyzes and processes the collected ambient light data and wake-up light effect parameters to achieve adaptive adjustment of the light effect.

[0100] In some embodiments, the wake-up execution terminal may further include an ambient light detection sensor 812 and a connecting wire 813. The ambient light detection sensor 812 is disposed on the side or front of the lamp to sense the ambient light intensity in real time, providing data support for light effect adjustment. The connecting wire 813 is used for power input and signal transmission, ensuring power supply and data interaction for the device.

[0101] In practical implementation, the main control module 808 determines the corresponding wake-up light parameters based on the wake-up signal and outputs the corresponding wake-up light. These wake-up light parameters include, but are not limited to, light duration, illuminance, and color temperature. These parameters play a crucial role in the secretion of melatonin and the regulation of circadian rhythms. Compared to simple brightness stimulation, this maximizes the regulatory effect of light on the biological clock, enhancing the "naturalness" of wake-up. This synchronizes the wake-up process with the body's internal physiological rhythms, creating a comfortable wake-up environment and improving the wake-up experience.

[0102] In some embodiments, the main control module 808 can adjust the color temperature based on the ambient light monitoring data collected by the ambient light detection sensor 812, so that the wake-up lighting color temperature is kept at a fixed value, such as 2500K.

[0103] In some embodiments, the main control module 808 can also adjust the color temperature by simultaneously combining the first ambient light monitoring data (i.e., the ambient light monitoring data around the smart wake-up device). Since the first ambient light monitoring data is closer to the user, it is closer to the user's actual ambient light. Therefore, adjusting the color temperature by combining the first ambient light monitoring data can make the adjusted color temperature value more accurate, thereby making the wake-up environment more natural.

[0104] In some embodiments, the main control module 808 may include a data acquisition unit and an adjustment unit. The data acquisition unit can acquire the user's wake-up state data, and the adjustment unit can use the wake-up state data to adjust the wake-up illumination parameters.

[0105] Specifically, the adjustment unit can have a built-in machine learning module. This module can dynamically adjust the illumination parameters for the next wake-up based on wake-up status data such as the user's heart rate recovery rate and bed-leaning reaction time after each wake-up. As usage frequency increases, this machine learning module can "understand the user better with use," thereby reducing the differences in wake-up effects between different individuals and enhancing overall adaptability.

[0106] In some embodiments, the pre-set wake-up lighting parameters can be quantified. For example, the pre-set start time can be T minutes before the user's preset wake-up time, where T ranges from 20 to 40 minutes (preferably 30 minutes), and the total wake-up time can be denoted as t, ranging from 15 to 30 minutes (preferably 20 minutes). The illuminance and color temperature curves should show a linear increase over time. This allows the lamp to accurately execute the wake-up program, ensuring a reproducible lighting environment that aligns with human physiological rhythms, improving the wake-up experience, and resolving the issues of ambiguous parameters and unstable experiences in existing technologies. This creates a deterministic and comfortable morning light environment for the user, enhancing the consistency and effectiveness of the wake-up experience.

[0107] Using the intelligent wake-up system in this embodiment of the invention, the intelligent wake-up device 60 can transmit the decision result of the optimal wake-up time to the independent wake-up execution terminal 80 through wireless communication. The wake-up execution terminal 80 then executes a non-contact, simulated sunrise light wake-up program to achieve a gentler and more instinctive wake-up stimulus.

[0108] This invention also provides a computer-readable storage medium having a computer program stored thereon, the computer program being executed by a processor to implement the steps of any of the above methods.

[0109] In specific implementations, the computer-readable storage medium may include ROM, RAM, disk, or optical disk, etc.

[0110] This invention also provides an electronic device, which includes a memory and a processor. The memory stores a computer program that can run on the processor, and when the processor runs the computer program, it performs the steps of any of the methods described above.

[0111] Regarding the modules / units included in the various devices and products described in the above embodiments, they can be software modules / units, hardware modules / units, or a combination of both. For example, for various devices and products applied to or integrated into a chip, all of their modules / units can be implemented using hardware methods such as circuits, or at least some modules / units can be implemented using software programs that run on a processor integrated within the chip, while the remaining (if any) modules / units can be implemented using hardware methods such as circuits; for various devices and products applied to or integrated into a chip module, all of their modules / units can be implemented using hardware methods such as circuits, and different modules / units can be located in the same component (e.g., chip, circuit module, etc.) or different components of the chip module, or at least some modules / units can be implemented using hardware methods such as circuits. The components can be implemented using software programs that run on the processor integrated within the chip module. The remaining (if any) modules / units can be implemented using hardware methods such as circuits. For various devices and products applied to or integrated into the terminal, each of its components / units can be implemented using hardware methods such as circuits. Different modules / units can be located in the same component (e.g., chip, circuit module, etc.) or in different components within the terminal. Alternatively, at least some modules / units can be implemented using software programs that run on the processor integrated within the terminal, while the remaining (if any) modules / units can be implemented using hardware methods such as circuits.

[0112] While the present invention has been disclosed above, it is not limited thereto. Any person skilled in the art can make various modifications and alterations without departing from the spirit and scope of the invention; therefore, the scope of protection of the present invention should be determined by the scope defined in the claims.

Claims

1. A smart wake-up method, characterized in that, include: Determine the non-deep sleep period closest to the preset wake-up time; A wake-up signal is generated during the defined non-deep sleep period for use in waking.

2. The intelligent wake-up method as described in claim 1, characterized in that, Determining the non-deep sleep period closest to the preset wake-up time includes: Real-time acquisition of monitoring data for wake-up parameters; Based on the collected wake-up parameter monitoring data, the current sleep stage is identified to determine the non-deep sleep stage closest to the preset wake-up time.

3. The intelligent wake-up method as described in claim 1, characterized in that, Determining the non-deep sleep period closest to the preset wake-up time includes: Obtain the phase prediction results of the circadian rhythm phase prediction model; Using the phase prediction results, the non-deep sleep period closest to the preset wake-up time is determined.

4. The intelligent wake-up method as described in claim 3, characterized in that, The step of using the phase prediction result to determine the non-deep sleep period closest to the preset wake-up time includes: Determine whether the phase prediction result satisfies the wake-up decision constraint; When the phase prediction result satisfies the wake-up decision constraint, the non-deep sleep period closest to the preset wake-up time is obtained based on the phase prediction result.

5. The intelligent wake-up method as described in claim 4, characterized in that, The step of determining the non-deep sleep period closest to the preset wake-up time based on the phase prediction result further includes: When the phase prediction result does not meet the wake-up decision constraints, the monitoring data based on the wake-up parameters identifies the non-deep sleep period closest to the preset wake-up time.

6. The intelligent wake-up method as described in claim 3, characterized in that, The wake-up decision constraints include: The matching error between the phase prediction result and the sleep period identification result obtained from the monitoring data based on the wake-up parameters is within a preset error threshold.

7. The intelligent wake-up method as described in claim 3, characterized in that, The step of using the phase prediction result to determine the non-deep sleep period closest to the preset wake-up time includes: Determine whether the actual application duration of the circadian rhythm phase prediction model has reached the preset duration; When the actual application duration of the circadian rhythm phase prediction model reaches the preset duration, the non-deep sleep period closest to the preset wake-up time is obtained based on the phase prediction results.

8. The intelligent wake-up method as described in claim 3, characterized in that, Also includes: Collect user wake-up status data; The model parameters of the circadian rhythm phase prediction model are optimized based on the collected wake-up state data.

9. The intelligent wake-up method as described in claim 2, 5, or 6, characterized in that, The wake-up parameters include: heart rate parameters and body movement parameters.

10. The intelligent wake-up method as described in claim 1, characterized in that, The non-deep sleep period is the REM sleep period or light sleep period.

11. The intelligent wake-up method as described in claim 1, characterized in that, Generating wake-up signals during the defined non-deep sleep period includes: When a light sleep period closest to the preset wake-up time is detected, a wake-up signal is generated within that light sleep period. When the light sleep period closest to the preset wake-up time is not detected, but the REM period closest to the preset wake-up time is detected, a wake-up signal is generated within the REM period closest to the preset wake-up time.

12. The intelligent wake-up method as described in claim 1, characterized in that, Also includes: Send the wake-up signal.

13. A smart wake-up device, characterized in that, include: The determination unit is suitable for identifying the non-deep sleep period closest to the preset wake-up time; A wake-up unit is adapted to generate a wake-up signal during a defined non-deep sleep period for use in waking up.

14. An intelligent wake-up system, characterized in that, include: The intelligent wake-up device according to claim 13; And wake up the execution terminal; The wake-up execution terminal is connected to the intelligent wake-up device and is used to receive the wake-up signal generated by the intelligent wake-up device and perform the wake-up operation.

15. The intelligent wake-up system as described in claim 14, characterized in that, The wake-up execution terminal includes a wake-up light, which is used to determine the corresponding wake-up illumination parameters according to the wake-up signal and output the corresponding wake-up light.

16. The intelligent wake-up system as described in claim 15, characterized in that, The wake-up light includes: The data acquisition unit is used to collect the user's wake-up status data; An adjustment unit is used to adjust the wake-up illumination parameters using the wake-up state data.

17. A computer-readable storage medium having a computer program stored thereon, characterized in that, The computer program is executed by a processor to implement the steps of the method according to any one of claims 1 to 12.

18. An electronic device comprising a memory and a processor, wherein the memory stores a computer program capable of running on the processor, characterized in that, When the processor runs the computer program, it performs the steps of the method according to any one of claims 1 to 12.

Images