An electric bed self-adaptive wake-up control method and system

Abstract

The application provides an electric bed self-adaptive wake-up control method and system, which comprises the following steps: S1, setting a target wake-up time T0 and a wake-up window ΔT on a user terminal, the wake-up window ΔT being a continuous time interval ending at the target wake-up time T0; S2, collecting human physiological signals and body movement signals in a sleep state at a starting time T0-ΔT of the wake-up window through a millimeter wave radar sensor; S3, performing signal noise reduction processing and feature extraction on the collected human physiological signals and body movement signals; and S4, transmitting the noise-reduced physiological signals and body movement signals to an embedded processor, the embedded processor being internally provided with a sleep staging algorithm and a wake-up reasoning algorithm, the sleep staging algorithm being used to determine a current sleep stage, and the wake-up reasoning algorithm being used to realize a multi-stage progressive wake-up strategy. The application has the advantages of non-inductive precise monitoring, scientific and comfortable wake-up, high wake-up reliability, self-adaptation and intelligence, high function integration degree and data value improvement.

Description

Technical Field

[0001] This invention belongs to the field of smart home and electric bed control technology, and more specifically, relates to an adaptive wake-up control method and system for an electric bed. Background Technology

[0002] With the increasing demand for healthy sleep, smart electric beds and smart wake-up technology are gradually merging, but existing technologies have the following core pain points:

[0003] 1. Defects in sleep monitoring methods: Traditional smart beds mostly use piezoelectric sensors, airbag pressure sensors, or wearable devices (such as smart eye masks and wristbands) to monitor sleep status. Piezoelectric / airbag sensors are easily affected by mattress deformation and environmental vibration, resulting in low monitoring accuracy; wearable devices are uncomfortable to wear, affecting sleep quality, and have poor privacy.

[0004] 2. Disconnect between sleep staging and wake-up strategies: Existing intelligent wake-up technologies mostly determine the wake-up time based solely on whether the user has fallen asleep, failing to accurately distinguish between deep sleep, light sleep, and REM sleep stages. This results in users waking up in a deep sleep state, leading to headaches, grogginess, and morning grumpiness. While some technologies attempt to incorporate sleep staging, they rely on contact-based data collection methods such as EEG signals, resulting in poor practicality.

[0005] 3. The wake-up function of electric beds is monotonous: The wake-up function of existing electric beds is mostly a vibration of fixed intensity or a headboard lifting. The execution parameters (such as lifting angle, speed, vibration frequency) are not adaptively adjusted according to sleep stages. The wake-up experience is stiff and cannot verify whether the user is truly awake. This can easily lead to the problem of "going back to sleep after turning off the alarm" and staying in bed.

[0006] 4. Insufficient adaptation to double-person scenarios: The wake-up control of the left and right bed areas of the double electric bed lacks independence, which can easily lead to one side being awakened and disturbing the sleep of the other side. It does not achieve closed-loop control of zone monitoring, zone reasoning, and zone wake-up. Summary of the Invention

[0007] Therefore, to solve the above-mentioned technical problems, this invention proposes an adaptive wake-up control method and system for an electric bed, comprising the following steps: S1, setting a target wake-up time T0 and a wake-up window ΔT on a user terminal, wherein the wake-up window ΔT is a continuous time interval ending at the target wake-up time T0; S2, real-time acquisition of human physiological signals and body movement signals during sleep using a millimeter-wave radar sensor during the start time T0-ΔT of the wake-up window, wherein the physiological signals include respiratory rate and heart rate, and the body movement signals include body movement amplitude and body movement frequency; S3, performing signal denoising and feature extraction on the acquired human physiological signals and body movement signals, and outputting the denoised physiological signals and body movement signals; S4, transmitting the denoised physiological signals and body movement signals to an embedded processor, wherein the embedded processor is equipped with a sleep staging algorithm and a wake-up inference algorithm, and the current sleep stage is determined by the sleep staging algorithm, wherein the current sleep stage includes at least a deep sleep stage. The system identifies sleep stages, light sleep stages, REM sleep stages, and the user's state of being out of bed. The wake-up inference algorithm implements a multi-level progressive wake-up strategy, which includes: continuously determining whether the user has entered the light sleep stage within the wake-up window ΔT; if so, triggering a level one wake-up; if no light sleep stage is detected within the wake-up window ΔT, triggering a level three wake-up at the target wake-up time T0; after each wake-up, verifying the user's wakefulness using a millimeter-wave radar sensor; if awake, ending the wake-up; if not awake, escalating to the next level of wake-up after a preset time T_interval, until the user is detected to be awake or out of bed; when the user's out-of-bed state is detected for more than the fourth threshold T4, ending the wake-up and recording the sleep stage data and wake-up process data; the wake-up inference algorithm also generates execution parameters including bedside lifting angle, lifting speed, vibration frequency, and wake-up duration based on the current sleep stage and user preferences. The advantages of this invention are: 1. Seamless and accurate monitoring: using FMCW. 1. Millimeter-wave radar enables non-contact sleep monitoring, requiring no wearable devices and causing no sleep disturbance; 2. Scientific and comfortable wake-up: An adaptive wake-up strategy based on sleep stages initiates wake-up during the light sleep stage, avoiding abrupt awakening during deep sleep, reducing morning grumpiness and grogginess, and improving wake-up comfort; 3. High wake-up reliability: Multi-level progressive stimulation combined with closed-loop wakefulness verification ensures that the user is eventually awakened, preventing missed wake-up times due to prolonged periods without entering light sleep; 4. Adaptive and intelligent: Wake-up parameters are dynamically adjusted according to sleep stages and support user preference learning; 5. High functional integration: A single millimeter-wave radar simultaneously performs sleep stage monitoring, wakefulness verification, and bed exit detection, requiring no additional sensors and keeping costs under control; 6. Enhanced data value: Automatically records sleep data and generates sleep reports, providing users with health references and increasing product added value.

[0008] An adaptive wake-up control method for an electric bed includes the following steps:

[0009] S1. By setting a target wake-up time T0 and a wake-up window ΔT on the user terminal, the wake-up window ΔT is a continuous time interval with the target wake-up time T0 as the end time;

[0010] S2. At the start time T0-ΔT of the wake-up window, the human physiological signals and body movement signals in the sleep state are collected in real time by a millimeter-wave radar sensor. The physiological signals include respiratory rate and heart rate, and the body movement signals include body movement amplitude and body movement frequency.

[0011] S3. Perform signal denoising and feature extraction on the collected human physiological signals and body movement signals, and output the denoised physiological signals and body movement signals.

[0012] S4. The noise-reduced physiological signals and body movement signals are transmitted to an embedded processor. The embedded processor has a sleep staging algorithm and a wake-up inference algorithm. The sleep staging algorithm determines the current sleep stage, which includes at least deep sleep, light sleep, REM sleep, and out-of-bed state. The wake-up inference algorithm implements a multi-level progressive wake-up strategy, which includes: continuously determining whether the user has entered the light sleep stage within the wake-up window ΔT; if the user has entered the light sleep stage, a first-level wake-up is triggered; if no light sleep stage is detected within the wake-up window ΔT... If the target wake-up time T0 is reached, a three-level wake-up is triggered. After each wake-up, the millimeter-wave radar sensor verifies whether the user is awake. If awake, the wake-up ends. If not awake, the wake-up is upgraded to the next level after a preset time interval T_interval, until the user is detected to be awake or has left the bed. When the user is detected to have left the bed for more than the fourth threshold T4, the wake-up ends, and the sleep stage data and wake-up process data are recorded. The wake-up inference algorithm also generates execution parameters including the headboard lifting angle, lifting speed, vibration frequency, and wake-up duration based on the current sleep stage and user preferences.

[0013] Furthermore, in S1, the user terminal includes a touch panel and a display screen, used to set the target wake-up time T0, wake-up window ΔT, wake-up intensity preference, and view sleep data.

[0014] Furthermore, in S2, the millimeter-wave radar sensor adopts a 60GHz FMCW frequency-modulated continuous wave radar with a detection range of 0.5-3m and a ranging accuracy of ±0.1m. It is installed above the head of the electric bed or inside the mattress to collect human physiological signals and body movement signals. For a double electric bed, the millimeter-wave radar sensor is installed in the left and right bed areas of the electric bed to collect human signals in the corresponding bed areas.

[0015] Furthermore, in S3, the signal noise reduction processing includes smoothing, filtering, and windowing of human physiological signals and body movement signals to remove environmental interference noise.

[0016] Furthermore, in S4, the basic state is determined based on the noise-reduced body movement signal. If there is no body movement amplitude and frequency and the respiratory rate and heart rate are stable within a continuously set time T1, it is determined to be a static state; if the body movement amplitude is greater than the first threshold A1, it is determined to be a body movement state; if there is no respiratory rate and heart rate and no body movement amplitude and frequency, it is determined to be an out-of-bed state.

[0017] Furthermore, in S4, based on the denoised physiological signals, the specific criteria for the sleep staging algorithm include:

[0018] Deep sleep stage: When the duration of rest is greater than the set time T2, the coefficient of variation of respiratory rate is less than the first coefficient C1, and the coefficient of variation of heart rate is less than the second coefficient C2;

[0019] Light sleep stage: The resting state lasts for 10-20 minutes, the coefficient of variation of respiratory rate is between 5% and 10%, the coefficient of variation of heart rate is between 8% and 15%, and there are occasional slight body movements;

[0020] REM sleep stage: The coefficients of variation of both respiratory rate and heart rate are greater than 10%, and there is no obvious body movement;

[0021] Out-of-bed status: No respiratory rate signal, no heart rate signal, and no body movement for a duration longer than the set time T3.

[0022] Furthermore, in S4, after triggering the first-level wake-up, the electric bed actuator is controlled to perform the corresponding mechanical action and / or environmental linkage; after the first-level wake-up, the user's wake-up status is verified by millimeter-wave radar signal. If the user is awake, the wake-up ends; if the user is not awake, the wake-up is upgraded to the second-level wake-up after a preset time T_interval, and the verification is performed again; if the user is still not awake after the second-level wake-up, the third-level wake-up is triggered when the target wake-up time T0 is reached; when it is detected that the user has been out of bed for more than the fourth threshold T4, the wake-up ends, and the sleep stage data and wake-up process data are recorded.

[0023] Furthermore, in S4, the multi-level progressive wake-up strategy specifically includes:

[0024] Level 1 wake-up: The headboard is raised to the first angle θ1 at the first speed V1, the vibration motor vibrates at the first frequency F1 at a low intensity, and the smart lights gradually brighten in conjunction with it;

[0025] Level 2 wake-up: The headboard is raised to the second angle θ2, the vibration motor vibrates at the second frequency F2, and the smart speaker plays audio in conjunction with it;

[0026] Level 3 wake-up: The head of the bed is raised to the maximum angle θ3, the vibration motor vibrates at the third frequency F3, the light brightness is increased to the maximum, and the music volume gradually increases;

[0027] Where θ1<θ2≤θ3, F1<F2<F3, and the interval T_interval is set between each wake-up level.

[0028] An adaptive wake-up control system for an electric bed includes:

[0029] Parameter setting module: used to set the target wake-up time T0, wake-up window ΔT, and wake-up preference parameters;

[0030] Signal acquisition module: includes at least one millimeter-wave radar sensor for acquiring human physiological signals and body movement signals during sleep. The physiological signals include respiratory rate and heart rate, and the body movement signals include body movement amplitude and body movement frequency.

[0031] Signal processing module: connected to the signal acquisition module, used to perform noise reduction processing on the acquired physiological signals and body movement signals;

[0032] Main control inference module: Connected to the signal processing module, it has a built-in sleep staging algorithm and wake-up inference algorithm. It performs sleep staging identification based on the noise-reduced physiological signals and body movement signals, and generates adaptive wake-up control commands according to the sleep staging results and preset wake-up parameters. The wake-up inference algorithm starts level one wake-up only when a light sleep stage is detected within the wake-up window set by the user. If no light sleep stage is detected within the window, wake-up is started at the target wake-up time. After wake-up, the wake-up level is verified based on millimeter-wave radar feedback. If the person is not awake, the wake-up level is upgraded.

[0033] Execution module: Signal-connected to the main control inference module, including at least one set of bed area drive components, each set of bed area drive components including a lifting motor and a vibration motor, used to receive the adaptive wake-up control command and execute multi-level progressive mechanical actions.

[0034] Furthermore, the main control inference module incorporates a sleep staging algorithm and a wake-up inference algorithm. The sleep staging algorithm divides sleep states into deep sleep, light sleep, REM sleep, and out-of-bed state based on the coefficient of variation of respiratory rate, coefficient of variation of heart rate, duration of body movement, and amplitude of body movement. The wake-up inference algorithm identifies the light sleep stage as the wake-up initiation time within the wake-up window set by the user, and generates execution parameters including the headboard lifting angle, lifting speed, vibration frequency, and wake-up duration based on the current sleep stage and user preferences.

[0035] The beneficial effects of this invention: This invention proposes an adaptive wake-up control method and system for an electric bed, comprising the following steps: S1, setting a target wake-up time T0 and a wake-up window ΔT on a user terminal, wherein the wake-up window ΔT is a continuous time interval ending at the target wake-up time T0; S2, real-time acquisition of human physiological signals and body movement signals during sleep using a millimeter-wave radar sensor during the start time T0-ΔT of the wake-up window, wherein the physiological signals include respiratory rate and heart rate, and the body movement signals include body movement amplitude and body movement frequency; S3, performing signal denoising and feature extraction on the acquired human physiological signals and body movement signals, and outputting the denoised physiological signals and body movement signals; S4, transmitting the denoised physiological signals and body movement signals to an embedded processor, wherein the embedded processor is equipped with a sleep staging algorithm and a wake-up inference algorithm, and using the sleep staging algorithm to determine the current sleep stage, wherein the current sleep stage includes at least a deep sleep stage and a light sleep stage. The sleep stage, REM sleep stage, and out-of-bed state are defined. The wake-up inference algorithm implements a multi-level progressive wake-up strategy, which includes: continuously determining whether the user has entered the light sleep stage within the wake-up window ΔT; if the user has entered the light sleep stage, a first-level wake-up is triggered; if no light sleep stage is detected within the wake-up window ΔT, a third-level wake-up is triggered at the target wake-up time T0; after each wake-up, the user is verified to be awake by a millimeter-wave radar sensor; if awake, the wake-up ends; if not awake, the wake-up is upgraded to the next level after a preset time T_interval, until the user is detected to be awake or out of bed; when the user is detected to be out of bed for more than the fourth threshold T4, the wake-up ends, and the sleep stage data and wake-up process data are recorded; the wake-up inference algorithm also generates execution parameters including the headboard lifting angle, lifting speed, vibration frequency, and wake-up duration based on the current sleep stage and user preferences. The advantages of this invention are: 1. Seamless and accurate monitoring: using FMCW 1. Millimeter-wave radar enables non-contact sleep monitoring, requiring no wearable devices and causing no sleep disturbance; 2. Scientific and comfortable wake-up: An adaptive wake-up strategy based on sleep stages initiates wake-up during the light sleep stage, avoiding abrupt awakening during deep sleep, reducing morning grumpiness and grogginess, and improving wake-up comfort; 3. High wake-up reliability: Multi-level progressive stimulation combined with closed-loop wakefulness verification ensures that the user is eventually awakened, preventing missed wake-up times due to prolonged periods without entering light sleep; 4. Adaptive and intelligent: Wake-up parameters are dynamically adjusted according to sleep stages and support user preference learning; 5. High functional integration: A single millimeter-wave radar simultaneously performs sleep stage monitoring, wakefulness verification, and bed exit detection, requiring no additional sensors and keeping costs under control; 6. Enhanced data value: Automatically records sleep data and generates sleep reports, providing users with health references and increasing product added value. Attached Figure Description

[0036] Figure 1This is a flowchart of an adaptive wake-up control method and system for an electric bed according to the present invention.

[0037] Figure 2 This is a flowchart of an adaptive wake-up control method and system for an electric bed according to the present invention.

[0038] The following detailed description, in conjunction with the accompanying drawings, will further illustrate the present invention. Detailed Implementation

[0039] The following embodiments are described to aid in understanding this application. These embodiments are not, and should not be, construed in any way as limiting the scope of protection of this application.

[0040] In the following description, those skilled in the art will recognize that, throughout this discussion, various components or portions thereof may be divided into individual components or may be integrated together (including integration within a single system or component).

[0041] Furthermore, the connection between components or systems is not intended to be limited to a direct connection; on the contrary, data between these components may be modified, reformatted, or otherwise altered by intermediate components. Additionally, other or fewer connections may be used. It should also be noted that the terms "connection," "link," or "input" should be understood to include direct connections, indirect connections via one or more intermediate devices, and wireless connections.

[0042] Example 1

[0043] like Figure 1 The diagram shown is a flowchart of an adaptive wake-up control method and system for an electric bed according to the present invention; Figure 2 The diagram shown is a flowchart of an adaptive wake-up control method and system for an electric bed according to the present invention.

[0044] An adaptive wake-up control method for an electric bed includes the following steps:

[0045] S1. By setting a target wake-up time T0 and a wake-up window ΔT on the user terminal, the wake-up window ΔT is a continuous time interval with the target wake-up time T0 as the end time;

[0046] S2. At the start time T0-ΔT of the wake-up window, the human physiological signals and body movement signals in the sleep state are collected in real time by a millimeter-wave radar sensor. The physiological signals include respiratory rate and heart rate, and the body movement signals include body movement amplitude and body movement frequency.

[0047] S3. Perform signal denoising and feature extraction on the collected human physiological signals and body movement signals, and output the denoised physiological signals and body movement signals.

[0048] S4. The noise-reduced physiological signals and body movement signals are transmitted to an embedded processor. The embedded processor has a sleep staging algorithm and a wake-up inference algorithm. The sleep staging algorithm determines the current sleep stage, which includes at least deep sleep, light sleep, REM sleep, and out-of-bed state. The wake-up inference algorithm implements a multi-level progressive wake-up strategy, which includes: continuously determining whether the user has entered the light sleep stage within the wake-up window ΔT; if the user has entered the light sleep stage, a first-level wake-up is triggered; if no light sleep stage is detected within the wake-up window ΔT... If the target wake-up time T0 is reached, a three-level wake-up is triggered. After each wake-up, the millimeter-wave radar sensor verifies whether the user is awake. If awake, the wake-up ends. If not awake, the wake-up is upgraded to the next level after a preset time interval T_interval, until the user is detected to be awake or has left the bed. When the user is detected to have left the bed for more than the fourth threshold T4, the wake-up ends, and the sleep stage data and wake-up process data are recorded. The wake-up inference algorithm also generates execution parameters including the headboard lifting angle, lifting speed, vibration frequency, and wake-up duration based on the current sleep stage and user preferences.

[0049] In S1, the user terminal includes a touch panel and a display screen, used to set the target wake-up time T0, wake-up window ΔT, wake-up intensity preference, and view sleep data.

[0050] In S2, the millimeter-wave radar sensor adopts a 60GHz FMCW frequency-modulated continuous wave radar with a detection range of 0.5-3m and a ranging accuracy of ±0.1m. It is installed above the head of the electric bed or inside the mattress to collect human physiological signals and body movement signals. For double electric beds, the millimeter-wave radar sensor is installed in the left and right bed areas of the electric bed to collect human signals in the corresponding bed areas.

[0051] In S3, the signal noise reduction processing includes smoothing, filtering, and windowing of human physiological signals and body movement signals to remove environmental interference noise.

[0052] In S4, the basic state is determined based on the noise-reduced body movement signal. If there is no body movement amplitude and frequency and the respiratory rate and heart rate are stable within a continuously set time T1, it is determined to be a static state. If the body movement amplitude is greater than the first threshold A1, it is determined to be a body movement state. If there is no respiratory rate and heart rate and no body movement amplitude and frequency, it is determined to be an out-of-bed state.

[0053] In S4, the specific criteria for the sleep staging algorithm, combined with the denoised physiological signals, include:

[0054] Deep sleep stage: When the duration of rest is greater than the set time T2, the coefficient of variation of respiratory rate is less than the first coefficient C1, and the coefficient of variation of heart rate is less than the second coefficient C2;

[0055] Light sleep stage: The resting state lasts for 10-20 minutes, the coefficient of variation of respiratory rate is between 5% and 10%, the coefficient of variation of heart rate is between 8% and 15%, and there are occasional slight body movements;

[0056] REM sleep stage: The coefficients of variation of both respiratory rate and heart rate are greater than 10%, and there is no obvious body movement;

[0057] Out-of-bed status: No respiratory rate signal, no heart rate signal, and no body movement for a duration longer than the set time T3.

[0058] In S4, after triggering the first-level wake-up, the electric bed actuator is controlled to perform the corresponding mechanical action and / or environmental linkage. After the first-level wake-up, the user's wake-up status is verified by millimeter-wave radar signal. If the user is awake, the wake-up ends. If the user is not awake, the wake-up is upgraded to the second-level wake-up after a preset time interval T_interval, and the verification is performed again. If the user is still not awake after the second-level wake-up, the third-level wake-up is triggered when the target wake-up time T0 is reached. When it is detected that the user has been out of bed for more than the fourth threshold T4, the wake-up ends, and the sleep stage data and wake-up process data are recorded.

[0059] In S4, the multi-level progressive wake-up strategy specifically includes:

[0060] Level 1 wake-up: The headboard is raised to the first angle θ1 at the first speed V1, the vibration motor vibrates at the first frequency F1 at a low intensity, and the smart lights gradually brighten in conjunction with it;

[0061] Level 2 wake-up: The headboard is raised to the second angle θ2, the vibration motor vibrates at the second frequency F2, and the smart speaker plays audio in conjunction with it;

[0062] Level 3 wake-up: The head of the bed is raised to the maximum angle θ3, the vibration motor vibrates at the third frequency F3, the light brightness is increased to the maximum, and the music volume gradually increases;

[0063] Where θ1<θ2≤θ3, F1<F2<F3, and the interval T_interval is set between each wake-up level.

[0064] In S4, the criterion for the awakening verification is: if the millimeter-wave radar sensor detects that the body movement amplitude is greater than the second threshold A2 and the heart rate rises by more than 10%, then the person is determined to be awake and the wake-up is terminated; otherwise, the person enters the next level of wake-up or remains awake until the target wake-up time T0 is reached.

[0065] The alertness verification also includes detecting changes in the waveform of respiratory rate and heart rate using a millimeter-wave radar sensor: when the respiratory rate waveform changes from stable to irregular, and the heart rate shows a momentary increase followed by a decrease, these changes, combined with the amplitude of body movement, constitute an auxiliary criterion for alertness.

[0066] For double electric beds, a dual-zone independent wake-up control step is also included: independent wake-up parameters are set for the left and right bed zones respectively. When either bed zone is activated for wake-up, the millimeter-wave radar sensor of the other bed zone detects the body movement amplitude of the user in that bed zone in real time. If the body movement amplitude exceeds the third threshold A3, the execution intensity of the wake-up side is reduced (the vibration frequency or lifting speed is reduced) until the interference disappears or the wake-up ends.

[0067] The embedded processor records the user's actual response to each wake-up level (awake time, time out of bed, manual intervention actions) during each wake-up process, and optimizes the execution parameters of subsequent wake-up levels (lifting angle, speed, vibration frequency, linkage intensity) through machine learning algorithms, so that the wake-up strategy gradually conforms to individual habits.

[0068] An adaptive wake-up control system for an electric bed includes:

[0069] Parameter setting module: used to set the target wake-up time T0, wake-up window ΔT, and wake-up preference parameters;

[0070] Signal acquisition module: includes at least one millimeter-wave radar sensor for acquiring human physiological signals and body movement signals during sleep. The physiological signals include respiratory rate and heart rate, and the body movement signals include body movement amplitude and body movement frequency.

[0071] Signal processing module: connected to the signal acquisition module, used to perform noise reduction processing on the acquired physiological signals and body movement signals;

[0072] Main control inference module: Connected to the signal processing module, it has a built-in sleep staging algorithm and wake-up inference algorithm. It performs sleep staging identification based on the noise-reduced physiological signals and body movement signals, and generates adaptive wake-up control commands according to the sleep staging results and preset wake-up parameters. The wake-up inference algorithm starts level one wake-up only when a light sleep stage is detected within the wake-up window set by the user. If no light sleep stage is detected within the window, wake-up is started at the target wake-up time. After wake-up, the wake-up level is verified based on millimeter-wave radar feedback. If the person is not awake, the wake-up level is upgraded.

[0073] Execution module: Signal-connected to the main control inference module, including at least one set of bed area drive components, each set of bed area drive components including a lifting motor and a vibration motor, used to receive the adaptive wake-up control command and execute multi-level progressive mechanical actions.

[0074] The main control inference module incorporates a sleep staging algorithm and a wake-up inference algorithm. The sleep staging algorithm divides sleep states into deep sleep, light sleep, REM sleep, and out-of-bed state based on the coefficient of variation of respiratory rate, coefficient of variation of heart rate, duration of body movement, and amplitude of body movement. The wake-up inference algorithm identifies the light sleep stage as the wake-up initiation time within the wake-up window set by the user, and generates execution parameters including the headboard lifting angle, lifting speed, vibration frequency, and wake-up duration based on the current sleep stage and user preferences.

[0075] The beneficial effects of this invention: This invention proposes an adaptive wake-up control method and system for an electric bed, comprising the following steps: S1, setting a target wake-up time T0 and a wake-up window ΔT on a user terminal, wherein the wake-up window ΔT is a continuous time interval ending at the target wake-up time T0; S2, real-time acquisition of human physiological signals and body movement signals during sleep using a millimeter-wave radar sensor during the start time T0-ΔT of the wake-up window, wherein the physiological signals include respiratory rate and heart rate, and the body movement signals include body movement amplitude and body movement frequency; S3, performing signal denoising and feature extraction on the acquired human physiological signals and body movement signals, and outputting the denoised physiological signals and body movement signals; S4, transmitting the denoised physiological signals and body movement signals to an embedded processor, wherein the embedded processor is equipped with a sleep staging algorithm and a wake-up inference algorithm, and using the sleep staging algorithm to determine the current sleep stage, wherein the current sleep stage includes at least a deep sleep stage and a light sleep stage. The sleep stage, REM sleep stage, and out-of-bed state are defined. The wake-up inference algorithm implements a multi-level progressive wake-up strategy, which includes: continuously determining whether the user has entered the light sleep stage within the wake-up window ΔT; if the user has entered the light sleep stage, a first-level wake-up is triggered; if no light sleep stage is detected within the wake-up window ΔT, a third-level wake-up is triggered at the target wake-up time T0; after each wake-up, the user is verified to be awake by a millimeter-wave radar sensor; if awake, the wake-up ends; if not awake, the wake-up is upgraded to the next level after a preset time T_interval, until the user is detected to be awake or out of bed; when the user is detected to be out of bed for more than the fourth threshold T4, the wake-up ends, and the sleep stage data and wake-up process data are recorded; the wake-up inference algorithm also generates execution parameters including the headboard lifting angle, lifting speed, vibration frequency, and wake-up duration based on the current sleep stage and user preferences. The advantages of this invention are: 1. Seamless and accurate monitoring: using FMCW 1. Millimeter-wave radar enables non-contact sleep monitoring, requiring no wearable devices and causing no sleep disturbance; 2. Scientific and comfortable wake-up: An adaptive wake-up strategy based on sleep stages initiates wake-up during the light sleep stage, avoiding abrupt awakening during deep sleep, reducing morning grumpiness and grogginess, and improving wake-up comfort; 3. High wake-up reliability: Multi-level progressive stimulation combined with closed-loop wakefulness verification ensures that the user is eventually awakened, preventing missed wake-up times due to prolonged periods without entering light sleep; 4. Adaptive and intelligent: Wake-up parameters are dynamically adjusted according to sleep stages and support user preference learning; 5. High functional integration: A single millimeter-wave radar simultaneously performs sleep stage monitoring, wakefulness verification, and bed exit detection, requiring no additional sensors and keeping costs under control; 6. Enhanced data value: Automatically records sleep data and generates sleep reports, providing users with health references and increasing product added value.

[0076] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. An adaptive wake-up control method for an electric bed, characterized in that, Includes the following steps: S1. By setting a target wake-up time T0 and a wake-up window ΔT on the user terminal, the wake-up window ΔT is a continuous time interval with the target wake-up time T0 as the end time; S2. At the start time T0-ΔT of the wake-up window, the human physiological signals and body movement signals in the sleep state are collected in real time by a millimeter-wave radar sensor. The physiological signals include respiratory rate and heart rate, and the body movement signals include body movement amplitude and body movement frequency. S3. Perform signal denoising and feature extraction on the collected human physiological signals and body movement signals, and output the denoised physiological signals and body movement signals. S4. The noise-reduced physiological signal and the body movement signal are transmitted to the embedded processor. The embedded processor is equipped with a sleep staging algorithm and a wake-up inference algorithm. The current sleep stage is determined by the sleep staging algorithm. The current sleep stage includes at least the deep sleep stage, the light sleep stage, the REM sleep stage, and the out-of-bed state. The wake-up inference algorithm implements a multi-level progressive wake-up strategy, which includes: continuously determining whether the user has entered the light sleep stage within the wake-up window ΔT; if the user has entered the light sleep stage, a first-level wake-up is triggered. If no light sleep stage is detected within the wake-up window ΔT, a three-level wake-up is triggered at the target wake-up time T0. After each wake-up, the user is verified to be awake by a millimeter-wave radar sensor. If awake, the wake-up ends. If not awake, the wake-up is upgraded to the next level after a preset time interval T_interval, until the user is detected to be awake or has left the bed. When the user is detected to have left the bed for more than the fourth threshold T4, the wake-up ends, and the sleep stage data and wake-up process data are recorded. The wake-up inference algorithm also generates execution parameters including the headboard lifting angle, lifting speed, vibration frequency, and wake-up duration based on the current sleep stage and user preferences.

2. The adaptive wake-up control method for an electric bed according to claim 1, characterized in that: In S1, the user terminal includes a touch panel and a display screen, used to set the target wake-up time T0, wake-up window ΔT, wake-up intensity preference, and view sleep data.

3. The adaptive wake-up control method for an electric bed according to claim 1, characterized in that: In S2, the millimeter-wave radar sensor adopts a 60GHz FMCW frequency-modulated continuous wave radar with a detection range of 0.5-3m and a ranging accuracy of ±0.1m. It is installed above the head of the electric bed or inside the mattress to collect human physiological signals and body movement signals. For double electric beds, the millimeter-wave radar sensor is installed in the left and right bed areas of the electric bed to collect human signals in the corresponding bed areas.

4. The adaptive wake-up control method for an electric bed according to claim 1, characterized in that: In S3, the signal noise reduction processing includes smoothing, filtering, and windowing of human physiological signals and body movement signals to remove environmental interference noise.

5. The adaptive wake-up control method for an electric bed according to claim 1, characterized in that: In S4, the basic state is determined based on the noise-reduced body movement signal. If there is no body movement amplitude and frequency and the respiratory rate and heart rate are stable within a continuously set time T1, it is determined to be a static state. If the body movement amplitude is greater than the first threshold A1, it is determined to be a body movement state. If there is no respiratory rate and heart rate and no body movement amplitude and frequency, it is determined to be an out-of-bed state.

6. The adaptive wake-up control method for an electric bed according to claim 5, characterized in that: In S4, the specific criteria for the sleep staging algorithm, combined with the denoised physiological signals, include: Deep sleep stage: When the duration of rest is greater than the set time T2, the coefficient of variation of respiratory rate is less than the first coefficient C1, and the coefficient of variation of heart rate is less than the second coefficient C2; Light sleep stage: The resting state lasts for 10-20 minutes, the coefficient of variation of respiratory rate is between 5% and 10%, the coefficient of variation of heart rate is between 8% and 15%, and there are occasional slight body movements; REM sleep stage: The coefficients of variation of both respiratory rate and heart rate are greater than 10%, and there is no obvious body movement; Out-of-bed status: No respiratory rate signal, no heart rate signal, and no body movement for a duration longer than the set time T3.

7. The adaptive wake-up control method for an electric bed according to claim 6, characterized in that: In S4, after triggering the first-level wake-up, the electric bed actuator is controlled to perform the corresponding mechanical action and / or environmental linkage. After the first-level wake-up, the user's wake-up status is verified by millimeter-wave radar signal. If the user is awake, the wake-up ends. If the user is not awake, the wake-up is upgraded to the second-level wake-up after a preset time interval T_interval, and the verification is performed again. If the user is still not awake after the second-level wake-up, the third-level wake-up is triggered when the target wake-up time T0 is reached. When it is detected that the user has been out of bed for more than the fourth threshold T4, the wake-up ends, and the sleep stage data and wake-up process data are recorded.

8. The adaptive wake-up control method for an electric bed according to claim 7, characterized in that: In S4, the multi-level progressive wake-up strategy specifically includes: Level 1 wake-up: The headboard is raised to the first angle θ1 at the first speed V1, the vibration motor vibrates at the first frequency F1 at a low intensity, and the smart lights gradually brighten in conjunction with it; Level 2 wake-up: The headboard is raised to the second angle θ2, the vibration motor vibrates at the second frequency F2, and the smart speaker plays audio in conjunction with it; Level 3 wake-up: The head of the bed is raised to the maximum angle θ3, the vibration motor vibrates at the third frequency F3, the light brightness is increased to the maximum, and the music volume gradually increases; Where θ1<θ2≤θ3, F1<F2<F3, and the interval T_interval is set between each wake-up level.

9. An adaptive wake-up control system for an electric bed, characterized in that, include: Parameter setting module: used to set the target wake-up time T0, wake-up window ΔT, and wake-up preference parameters; Signal acquisition module: includes at least one millimeter-wave radar sensor for acquiring human physiological signals and body movement signals during sleep. The physiological signals include respiratory rate and heart rate, and the body movement signals include body movement amplitude and body movement frequency. Signal processing module: connected to the signal acquisition module, used to perform noise reduction processing on the acquired physiological signals and body movement signals; Main control inference module: Connected to the signal processing module, it has a built-in sleep staging algorithm and wake-up inference algorithm. It performs sleep staging identification based on the noise-reduced physiological signals and body movement signals, and generates adaptive wake-up control commands according to the sleep staging results and preset wake-up parameters. The wake-up inference algorithm starts level one wake-up only when a light sleep stage is detected within the wake-up window set by the user. If no light sleep stage is detected within the window, wake-up is started at the target wake-up time. After wake-up, the wake-up level is verified based on millimeter-wave radar feedback. If the person is not awake, the wake-up level is upgraded. Execution module: Signal-connected to the main control inference module, including at least one set of bed area drive components, each set of bed area drive components including a lifting motor and a vibration motor, used to receive the adaptive wake-up control command and execute multi-level progressive mechanical actions.

10. The adaptive wake-up control system for an electric bed according to claim 9, characterized in that: The main control inference module incorporates a sleep staging algorithm and a wake-up inference algorithm. The sleep staging algorithm divides sleep states into deep sleep, light sleep, REM sleep, and out-of-bed state based on the coefficient of variation of respiratory rate, coefficient of variation of heart rate, duration of body movement, and amplitude of body movement. The wake-up inference algorithm identifies the light sleep stage as the wake-up initiation time within the wake-up window set by the user, and generates execution parameters including the headboard lifting angle, lifting speed, vibration frequency, and wake-up duration based on the current sleep stage and user preferences.

Images