A multi-modal circadian rhythm entrainment sleep intervention method and system

Abstract

The present application relates to sleep aid technology field, specifically relates to a kind of multi-modal physiological rhythm traction type sleep intervention method and system, and with heart rate data as benchmark dynamic determination target guiding heart rate, generate and contain inspiration stage and expiration stage and length adaptive adjustment with heart rate variation physiological guiding period, so that guiding rhythm and user actual heart rate variation keep synchronization;Generation respectively acts on visual channel, auditory channel and tactile channel synchronous control signal, through light, sound, touch three sensory channels in the same rhythm clock forced synchronization change, form cross-sensory rhythm traction field, actively induce user heart rate to target resting level approach;With real-time update heart rate data continuous adjustment physiological guiding period, form monitoring, intervention, evaluation, again intervention closed-loop control link, until user autonomic nervous state switches to sleep preparation state, to realize multi-modal synergistic effect under active physiological rhythm traction.

Description

Technical Field

[0001] This invention relates to the field of sleep aid technology, specifically to a multimodal physiological rhythm traction sleep intervention method and system. Background Technology

[0002] With the widespread use of electronic devices, it has become common for users to spend extended periods of time at night using screen devices such as mobile phones and computers. The short-wavelength blue light emitted by screens (wavelength 400-450nm) inhibits the secretion of melatonin by the pineal gland, leading to a prolonged sleep latency. The mental focus and information stimulation caused by screen content keep the brain in a state of beta wave excitation, manifested as increased resting heart rate and decreased heart rate variability. Even after turning off the screen, users still experience problems such as rumination, rapid heartbeat, and shallow breathing, making it difficult to quickly enter a sleep-prepared state.

[0003] Existing sleep aid products can be mainly categorized as follows: Sleep aid desk lamps create a dim environment by lowering color temperature and brightness, but this is only a passive environmental adjustment; Sleep monitoring bracelets record sleep data using accelerometers and heart rate sensors, but lack active intervention capabilities; White noise playback devices mask environmental noise with natural sounds, but the effect of single auditory stimulation on suppressing sympathetic nerve excitation behind the screen is limited, and it cannot actively counteract high-frequency human voices on the screen; Breathing guide lights prompt users to adjust their breathing through fixed-frequency changes in brightness, but this is an open-loop control that cannot detect whether the user is actually following, nor can it resolve frequency conflicts caused by the user's spontaneous heart rate being much higher than the guided frequency due to mental excitement.

[0004] In summary, existing sleep aids rely on a single sensory experience to guide sleep, resulting in limited intervention effects. These products are all open-loop controls, unable to detect whether the user is actually following the sleep aid guidance, and even less able to dynamically adjust intervention strategies based on the user's real-time physiological state. Furthermore, it is difficult to construct a physiological rhythm traction mechanism that uses the user's current heart rate as a benchmark to generate and apply an external guidance signal that is slightly slower than the current rhythm and includes a prolonged expiratory phase, thereby inducing the heart rhythm to approach the resting level. Summary of the Invention

[0005] The main objective of this invention is to provide a multimodal physiological rhythm-guided sleep intervention method and system to solve the technical problem that existing technologies guide sleep based on a single sense, which cannot sense the user's state for dynamic adjustment, thus making it difficult to actively guide the user into a sleep preparation state using physiological rhythms.

[0006] To achieve the above objectives, the present invention provides a multimodal circadian rhythm-induced sleep intervention method, comprising the following steps: S1. Collect the user's real-time heart rate data; S2. Determine the target guiding heart rate based on the real-time heart rate data, and generate a physiological guiding cycle that includes an inhalation phase and an exhalation phase; wherein, the duration of the physiological guiding cycle is dynamically adjusted according to the change of the target guiding heart rate; S3. Generate synchronous control signals corresponding to the visual channel, auditory channel and tactile channel respectively according to the physiological guidance cycle; S4. During the inhalation phase, the illumination brightness, audio rhythm pulse output, and tactile feedback intensity are synchronously increased according to the synchronization control signal; during the exhalation phase, the illumination brightness is synchronously decreased, the audio output resumes continuous pulse-free output, and the tactile feedback stops according to the synchronization control signal. S5. Repeat steps S1 to S4, dynamically adjust the physiological guidance cycle based on the real-time updated heart rate data to form a closed-loop control until the user enters a sleep preparation state.

[0007] Furthermore, step S1 specifically includes the following steps: S11. The wearable device's sensor collects the raw pulse wave signal at a first sampling frequency, and simultaneously collects triaxial acceleration data at the first sampling frequency through the wearable device's inertial measurement unit. S12. Calculate the synthetic acceleration amplitude based on the triaxial acceleration data, and use the synthetic acceleration amplitude as a reference noise source to filter and denoise the original pulse wave signal to obtain the filtered pulse wave signal. S13. Obtain the duration of the sliding window, perform peak detection on the filtered pulse wave signal based on the duration of the sliding window, determine the pulse wave peak in the filtered pulse wave signal, and calculate the time interval between adjacent pulse wave peaks. S14. Calculate the instantaneous heart rate value based on the time interval between adjacent pulse wave peaks, and determine the real-time heart rate data based on the instantaneous heart rate value.

[0008] More preferably, after step S14, the following steps are also included: Determine whether the number of pulse wave peaks within the sliding window duration is greater than or equal to a preset threshold. If not, mark the real-time heart rate data acquired at the current moment as low confidence and keep the previous effective heart rate value unchanged. The previous effective heart rate value refers to the most recent real-time heart rate data calculated based on pulse wave peaks that meet the preset threshold requirement before the current moment, which was not marked as low confidence.

[0009] Furthermore, step S2 specifically includes the following steps: S21. Use the real-time heart rate data as the target guiding heart rate for the current moment; S22. Obtain the heart rate update cycle and the preset target resting heart rate. Every heart rate update cycle, determine whether the target guided heart rate at the current moment is greater than the preset target resting heart rate. If so, calculate the first difference between the target guided heart rate at the current moment and the preset target resting heart rate. S23. Obtain a preset decrease coefficient, determine the heart rate decrease range in this round based on the product of the first difference and the preset decrease coefficient, and use the target guided heart rate at the current moment minus the heart rate decrease range in this round as the updated target guided heart rate. S24. Determine the guiding beat interval based on the updated target guiding heart rate; wherein, the guiding beat interval is equal to sixty divided by the updated target guiding heart rate; S25. Obtain a preset number of inspiratory cardiac cycles, and determine the duration of the inspiratory phase based on the guide beat interval and the preset number of inspiratory cardiac cycles; wherein, the duration of the inspiratory phase is equal to the preset number of inspiratory cardiac cycles multiplied by the guide beat interval. S26. Obtain a preset number of expiratory cardiac cycles, and determine the duration of the expiratory phase based on the guide beat interval and the preset number of expiratory cardiac cycles; wherein, the duration of the expiratory phase is equal to the preset number of expiratory cardiac cycles multiplied by the guide beat interval, and the preset number of expiratory cardiac cycles is at least 1.5 times the preset number of inspiratory cardiac cycles. S27. The sum of the duration of the inhalation phase and the duration of the exhalation phase is taken as the duration of the physiological guidance cycle to obtain the updated physiological guidance cycle.

[0010] Furthermore, step S3 specifically includes the following steps: The synchronization control signals for the visual, auditory, and tactile channels are all under the same clock during the physiological guidance cycle: S31. Taking the start time of the physiological guidance cycle as the zero point of timing, obtain the cumulative running time of the current time relative to the zero point of timing; S32. The cumulative running time is modulo the duration of the physiological guidance cycle to obtain the original normalized phase within the physiological guidance cycle at the current moment; wherein the value of the original normalized phase ranges from zero to one. S33. Obtain a preset rhythm traction coefficient, determine the second difference between the real-time heart rate data and the target guided heart rate, divide the second difference by the real-time heart rate data and multiply by the preset rhythm traction coefficient to obtain the phase offset. S34. Add the original normalized phase to the phase offset, take the modulo of one, and then determine the adjusted normalized phase; wherein the value range of the adjusted normalized phase is from zero to one; S35. Calculate the first ratio of the duration of the inhalation phase to the duration of the physiological guidance cycle, and use the first ratio as the proportion of the inhalation phase. S36. Determine whether the adjusted normalized phase is less than the proportion of the inhalation phase; if so, generate a visual channel control signal, an auditory channel control signal, and a tactile channel control signal corresponding to the inhalation phase. S37. If the adjusted normalized phase is greater than or equal to the proportion of the inhalation phase, then a visual channel control signal corresponding to the exhalation phase, an auditory channel control signal corresponding to the exhalation phase, and a tactile channel control signal corresponding to the exhalation phase are generated.

[0011] More preferably, step S4, which involves synchronously increasing the illumination brightness, outputting audio rhythmic pulses, and increasing the tactile feedback intensity according to the synchronization control signal during the inhalation phase, specifically includes the following steps: The system receives a visual channel control signal corresponding to the inhalation phase, obtains the quotient obtained by dividing the adjusted normalized phase by the proportion of the inhalation phase, calculates and determines the target value of the illumination brightness during the inhalation phase using a first incrementing function, and then adjusts the output brightness of the illumination device to the target value of the illumination brightness during the inhalation phase; wherein, the first incrementing function is a quadratic function of the quotient, so that the target value of the illumination brightness during the inhalation phase increases rapidly as the adjusted normalized phase increases; The system receives an auditory channel control signal corresponding to the inhalation phase, acquires the guide beat interval of the physiological guidance cycle, determines the output time of the rhythmic pulse audio based on the guide beat interval, and controls the audio output module to play a pulse audio of a preset duration at the start time of each guide beat interval; wherein the output time of the rhythmic pulse audio is aligned with the start time of the cardiac cycle in the physiological guidance cycle. The system receives a tactile channel control signal corresponding to the inhalation phase, obtains the quotient obtained by dividing the adjusted normalized phase by the proportion of the inhalation phase, calculates and determines the target value of the tactile feedback intensity during the inhalation phase using a second incrementing function, and then adjusts the output intensity of the tactile feedback device to the target value of the tactile feedback intensity during the inhalation phase; wherein, the second incrementing function is a quadratic function of the quotient, so that the target value of the tactile feedback intensity during the inhalation phase increases rapidly as the adjusted normalized phase increases.

[0012] More preferably, step S4, which involves synchronously executing the decrease in illumination brightness, the restoration of continuous pulse-free audio output, and the cessation of tactile feedback during the exhalation phase according to the synchronization control signal, specifically includes the following steps: The system receives a visual channel control signal corresponding to the exhalation phase, calculates a first intermediate value and a second intermediate value, determines a second ratio of the first intermediate value to the second intermediate value, calculates a target value for illumination brightness during the exhalation phase using a first decreasing function, and then adjusts the output brightness of the illumination device to the target value for illumination brightness during the exhalation phase. The first intermediate value is the difference between the adjusted normalized phase and the proportion of the inhalation phase; the second intermediate value is the difference between 1 and the proportion of the inhalation phase; the first decreasing function is the square root function of the second ratio, causing the target value for illumination brightness during the exhalation phase to decrease at a slower rate as the adjusted normalized phase increases. It receives the auditory channel control signal corresponding to the exhalation phase, controls the audio output module to stop playing rhythmic pulse audio, and resumes playing continuous background audio without pulses; Receive the tactile channel control signal corresponding to the exhalation phase, adjust the output intensity of the tactile feedback device to zero, and stop the tactile feedback output.

[0013] Furthermore, the sleep preparation state described in step S5 is specifically determined by the following steps: Obtain updated real-time heart rate data and preset target resting heart rate, and determine the preset resting heart rate tolerance based on the preset target resting heart rate; Determine whether the real-time heart rate data is less than or equal to the preset tolerance of the resting heart rate; if so, determine that the user has entered a sleep preparation state.

[0014] This invention also provides a multimodal circadian rhythm traction sleep intervention system, which applies the multimodal circadian rhythm traction sleep intervention method described above, including: The data acquisition module is used to collect the user's real-time heart rate data; The calculation module, connected to the heart rate data acquisition module, is used to determine the target guiding heart rate based on the real-time heart rate data and generate a physiological guiding cycle including an inhalation phase and an exhalation phase, wherein the duration of the physiological guiding cycle is dynamically adjusted according to the change of the target guiding heart rate. A synchronization control module, connected to the computing module, is used to generate synchronization control signals corresponding to the visual channel, auditory channel and tactile channel respectively according to the physiological guidance cycle; An execution module, connected to the synchronization control module, is used to synchronously execute increasing illumination brightness, outputting audio rhythmic pulses, and increasing tactile feedback intensity according to the synchronization control signal during the inhalation phase; and synchronously execute decreasing illumination brightness, restoring continuous no-pulse audio output, and stopping tactile feedback according to the synchronization control signal during the exhalation phase. The closed-loop control module is connected to the data acquisition module and the calculation module respectively, and is used to control the data acquisition module, the calculation module, the synchronization control module and the execution module to repeatedly execute, dynamically adjust the physiological guidance cycle according to the real-time updated heart rate data, form a closed-loop control, until the user enters the sleep preparation state.

[0015] Furthermore, the execution module includes a desk lamp and a wristband. The desk lamp includes a lamp head with a light source and a lamp holder with a speaker. The desk lamp and the wristband are communicatively connected. The lamp head is rotatably connected to the lamp holder. The wristband collects the user's heart rate data in real time. The light source performs increasing and decreasing illumination brightness. The speaker performs audio rhythm pulse output. The wristband has a built-in vibration motor, and the vibration motor performs tactile feedback.

[0016] Compared with the prior art, the present invention has the following beneficial effects: This invention collects real-time heart rate data from users to perceive their current physiological state and dynamically determines a target guiding heart rate based on this data. It generates a physiological guidance cycle that includes inhalation and exhalation phases, with the duration adaptively adjusting to changes in heart rate. This ensures the guiding rhythm is synchronized with the user's actual heart rate changes, avoiding rhythmic conflicts between fixed-frequency guidance and the user's spontaneous heart rate. Based on the same physiological guidance cycle, it generates synchronous control signals acting on the visual, auditory, and tactile channels respectively. During inhalation, it synchronously drives an increase in lighting brightness, an increase in audio rhythmic pulse output, and an increase in tactile feedback intensity; during exhalation, it synchronously drives the lighting... The decreasing brightness, the resumption of continuous pulse-free audio output, and the cessation of tactile feedback, through the forced synchronous changes of the three sensory channels of light, sound, and touch under the same rhythmic clock, form a cross-sensory rhythmic traction field. By utilizing the physiological mechanism of respiratory sinus arrhythmia, the parasympathetic nervous system's regulatory tension on heart rate is enhanced, actively inducing the user's heart rate to approach the target resting level. By cyclically executing the above steps, the physiological guidance cycle is continuously adjusted with real-time updated heart rate data, forming a closed-loop control link of monitoring, intervention, evaluation, and re-intervention, until the user's autonomic nervous system switches to a sleep preparation state, thereby achieving active physiological rhythmic traction under multimodal synergy. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0018] Figure 1This is a flowchart illustrating a multimodal physiological rhythm traction sleep intervention method according to an embodiment of the present invention; Figure 2 This is a schematic diagram of a sleep intervention system including a desk lamp and a wristband according to an embodiment of the present invention; Figure 3 This is a schematic diagram of the main light source and auxiliary light source of a desk lamp in one embodiment of the present invention; Figure 4 This is a schematic diagram of the built-in module of the table lamp holder in one embodiment of the present invention; Figure 5 This is a schematic diagram of the internal module of the wristband in one embodiment of the present invention.

[0019] In the picture: 1. Lamp head; 2. Lamp holder; 3. Wristband; 4. Microphone array; 5. Speaker; 6. Main light source; 7. Secondary light source; 8. Ambient blue light sensor; 9. Green LED; 10. MCU; 11. Bluetooth communication module; 12. Vibration motor; 13. Three-axis gyroscope; 14. Three-axis accelerometer; 15. Skin contact impedance detection electrode; 16. Wristband microcontroller.

[0020] The objectives, features, and advantages of this invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0021] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0022] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0023] It should be noted that all directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of the present invention are only used to explain the relative positional relationship and movement of each component in a certain specific posture (as shown in the figure). If the specific posture changes, the directional indication will also change accordingly.

[0024] Furthermore, the use of terms such as "first" and "second" in this invention is for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined with "first" or "second" may explicitly or implicitly include at least one of that feature. Additionally, the technical solutions of the various embodiments can be combined with each other, but only on the basis of being achievable by those skilled in the art. When the combination of technical solutions is contradictory or impossible to implement, such a combination of technical solutions should be considered non-existent and not within the scope of protection claimed by this invention.

[0025] Please see Figures 1 to 5 This embodiment uses a sleep intervention system comprising a desk lamp and a wristband 3 as its hardware carrier. The desk lamp integrates a main light source 6, a secondary light source 7, an ambient blue light sensor 8, a microphone array 4, a speaker 5, a Bluetooth communication module 11, a synchronization control module, and a closed-loop control module. The wristband 3 integrates a PPG heart rate sensor, a vibration motor 12, a six-axis inertial measurement unit, skin contact impedance detection electrodes 15, and a Bluetooth communication module 11. The synchronization control module and the closed-loop control module employ an ARM Cortex-M series microcontroller with a main frequency of 72MHz, responsible for executing the steps of the multimodal physiological rhythm traction sleep intervention method in this embodiment.

[0026] This embodiment provides a multimodal physiological rhythm-based sleep intervention method, including the following steps: S1. Collect the user's real-time heart rate data; S2. Determine the target guiding heart rate based on the real-time heart rate data, and generate a physiological guiding cycle that includes an inhalation phase and an exhalation phase; wherein, the duration of the physiological guiding cycle is dynamically adjusted according to the change of the target guiding heart rate; S3. Generate synchronous control signals corresponding to the visual channel, auditory channel and tactile channel respectively according to the physiological guidance cycle; S4. During the inhalation phase, the illumination brightness, audio rhythm pulse output, and tactile feedback intensity are synchronously increased according to the synchronization control signal; during the exhalation phase, the illumination brightness is synchronously decreased, the audio output resumes continuous pulse-free output, and the tactile feedback stops according to the synchronization control signal. S5. Repeat steps S1 to S4, dynamically adjust the physiological guidance cycle based on the real-time updated heart rate data to form a closed-loop control until the user enters a sleep preparation state.

[0027] This embodiment collects real-time heart rate data to sense the user's current physiological state and dynamically determines the target guidance heart rate based on this heart rate data. It generates a physiological guidance cycle that includes inhalation and exhalation phases, with the duration adaptively adjusted according to heart rate changes. This ensures the guidance rhythm is synchronized with the user's actual heart rate changes, avoiding rhythmic conflicts between fixed-frequency guidance and the user's spontaneous heart rate. Based on the same physiological guidance cycle, it generates synchronous control signals acting on the visual, auditory, and tactile channels respectively. During the inhalation phase, it synchronously drives the increase in lighting brightness, audio rhythm pulse output, and tactile feedback intensity; during the exhalation phase, it synchronously drives the increase in lighting brightness... The decrease in intensity, the restoration of continuous pulse-free audio output, and the cessation of tactile feedback, through the forced synchronous changes of the three sensory channels of light, sound, and touch under the same rhythmic clock, form a cross-sensory rhythmic traction field. By utilizing the physiological mechanism of respiratory sinus arrhythmia, the parasympathetic nervous system's regulatory tension on heart rate is enhanced, actively inducing the user's heart rate to approach the target resting level. By cyclically executing the above steps, the physiological guidance cycle is continuously adjusted with real-time updated heart rate data, forming a closed-loop control link of monitoring, intervention, evaluation, and re-intervention, until the user's autonomic nervous system switches to a sleep preparation state, thereby achieving active physiological rhythmic traction under multimodal synergy.

[0028] As a further step in this embodiment, step S1 specifically includes the following steps: S11. The wearable device's sensor collects the raw pulse wave signal at a first sampling frequency, and simultaneously collects triaxial acceleration data at the first sampling frequency through the wearable device's inertial measurement unit. S12. Calculate the synthetic acceleration amplitude based on the triaxial acceleration data, and use the synthetic acceleration amplitude as a reference noise source to filter and denoise the original pulse wave signal to obtain the filtered pulse wave signal. S13. Obtain the duration of the sliding window, perform peak detection on the filtered pulse wave signal based on the duration of the sliding window, determine the pulse wave peak in the filtered pulse wave signal, and calculate the time interval between adjacent pulse wave peaks. S14. Calculate the instantaneous heart rate value based on the time interval between adjacent pulse wave peaks, and determine the real-time heart rate data based on the instantaneous heart rate value.

[0029] Determine whether the number of pulse wave peaks within the sliding window duration is greater than or equal to a preset threshold. If not, mark the real-time heart rate data acquired at the current moment as low confidence and keep the previous effective heart rate value unchanged. The previous effective heart rate value refers to the most recent real-time heart rate data calculated based on pulse wave peaks that meet the preset threshold requirement before the current moment, which was not marked as low confidence.

[0030] Specifically, in this embodiment, the PPG heart rate sensor built into the wearable bracelet 3 collects the raw pulse wave signal from the user's wrist at a first sampling frequency of 100Hz. Simultaneously, the six-axis inertial measurement unit within the bracelet 3 synchronously collects three-axis acceleration data at the same 100Hz sampling frequency.

[0031] Because users may make slight hand movements during sleep preparation, these movements can introduce motion artifacts, interfering with the accurate extraction of pulse wave signals. To eliminate this interference, the wristband microcontroller 16 first calculates the synthetic acceleration amplitude based on triaxial acceleration data. The calculation formula is as follows: ; in: This represents the magnitude of the resultant acceleration at time t. , and These represent the acceleration data along the X, Y, and Z axes, respectively.

[0032] Subsequently, a minimum mean square adaptive filter is used to synthesize the acceleration amplitude. As a reference noise source, the original pulse wave signal For motion noise suppression, the filter weight coefficient update formula is: ; in, This represents the weighting coefficient at time t. The initial value can be set to 0 or a preset small value. This is the step size factor, which is set to 0.001 in this embodiment. This represents the filtered pulse wave signal.

[0033] This allows the filter to dynamically adjust its ability to suppress motion noise, enabling the adaptive filter to track changes in hand movements in real time, thereby effectively eliminating motion artifacts and improving the accuracy of extracting heart rate from the original pulse wave signal.

[0034] Next, regarding Perform peak detection within a 5-second sliding window to identify pulse wave peaks and calculate the time interval between adjacent peaks. , and by Obtain the instantaneous heart rate value.

[0035] If fewer than three valid pulse wave peaks are detected within the sliding window duration, the current signal quality is determined to be low. The heart rate data at this time is marked as low confidence, and the previous valid heart rate value remains unchanged.

[0036] The instantaneous heart rate value is smoothed using a first-order low-pass filter, with a smoothing coefficient of [missing value]. Set it to 0.2 to obtain the real-time heart rate data at the current moment. The real-time heart rate data is uploaded to the desk lamp in the form of data packets via the Bluetooth communication module 11 of the wristband 3.

[0037] This embodiment effectively suppresses the interference of minor hand movements on the pulse wave signal through adaptive filtering, improving the accuracy of heart rate data in non-resting states; through low-confidence judgment and smoothing filtering, it avoids sudden changes in the guiding rhythm caused by instantaneous signal fluctuations, providing a stable and reliable physiological feedback source for closed-loop control.

[0038] In one embodiment, step S2 further includes the following steps: S21. Use the real-time heart rate data as the target guiding heart rate for the current moment; S22. Obtain the heart rate update cycle and the preset target resting heart rate. Every heart rate update cycle, determine whether the target guided heart rate at the current moment is greater than the preset target resting heart rate. If so, calculate the first difference between the target guided heart rate at the current moment and the preset target resting heart rate. S23. Obtain a preset decrease coefficient, determine the heart rate decrease range in this round based on the product of the first difference and the preset decrease coefficient, and use the target guided heart rate at the current moment minus the heart rate decrease range in this round as the updated target guided heart rate. S24. Determine the guiding beat interval based on the updated target guiding heart rate; wherein, the guiding beat interval is equal to sixty divided by the updated target guiding heart rate; S25. Obtain a preset number of inspiratory cardiac cycles, and determine the duration of the inspiratory phase based on the guide beat interval and the preset number of inspiratory cardiac cycles; wherein, the duration of the inspiratory phase is equal to the preset number of inspiratory cardiac cycles multiplied by the guide beat interval. S26. Obtain a preset number of expiratory cardiac cycles, and determine the duration of the expiratory phase based on the guide beat interval and the preset number of expiratory cardiac cycles; wherein, the duration of the expiratory phase is equal to the preset number of expiratory cardiac cycles multiplied by the guide beat interval, and the preset number of expiratory cardiac cycles is at least 1.5 times the preset number of inspiratory cardiac cycles. S27. The sum of the duration of the inhalation phase and the duration of the exhalation phase is taken as the duration of the physiological guidance cycle to obtain the updated physiological guidance cycle.

[0039] In this embodiment, after receiving real-time heart rate data, the desk lamp performs the following sub-steps to generate an adaptive physiological guidance cycle: First, determine the target heart rate. At the initial moment of sleep guidance mode activation, the first message received will be... Directly use as the initial target heart rate .

[0040] Every preset heart rate update cycle An update is performed every 60 seconds in this embodiment. The update rule is: compare the current... Compared to the preset target resting heart rate The size relationship in this embodiment Take 60 BPM, which corresponds to the ideal heart rate range for healthy adults during the sleep preparation period. Then calculate the first difference using the following formula: .

[0041] The heart rate decrease range for this round is determined by multiplying the first difference by a preset decrease coefficient. To avoid causing user discomfort due to excessively rapid changes in the guided rhythm, a preset step size limit of 3 BPM / minute is imposed on the heart rate decrease range for this round. The current... The updated target guided heart rate is obtained by subtracting the decrease in heart rate during this round. .like Then keep constant.

[0042] The "preset decrease coefficient" refers to a dimensionless scaling factor used to control the rate at which the target guided heart rate approaches the preset target resting heart rate. Its value is a real number greater than 0 and less than or equal to 1. Multiplying the preset decrease coefficient by the first difference between the current target guided heart rate and the preset target resting heart rate yields the heart rate decrease magnitude for this round. The specific value of the preset decrease coefficient can be set according to the user's sensitivity to the guidance rhythm or different guidance stages; in this embodiment, it is set to 0.1.

[0043] Based on the updated target guided heart rate Generate the physiological guidance cycle. Calculate the guidance beat interval. The calculation formula is: That is, the duration of each cardiac cycle.

[0044] The duration of the inspiratory phase is calculated based on the preset number of inspiratory cardiac cycles (4 cycles in this embodiment) and the preset number of expiratory cardiac cycles (8 cycles in this embodiment). Duration of the exhalation phase The duration of a complete physiological guidance cycle In this embodiment, the number of expiratory cardiac cycles is twice the number of inspiratory cardiac cycles. This ratio is designed to utilize the physiological mechanism of respiratory sinus arrhythmia—when the expiratory phase is prolonged, the parasympathetic nervous system's control over the sinoatrial node is enhanced, and the heart rate naturally decreases, thereby actively promoting the switching of the autonomic nervous system to sleep preparation mode.

[0045] In this embodiment, the target guided heart rate smoothly approaches 60 BPM at a rate not exceeding 3 BPM / minute, ensuring that the amplitude of the guided rhythm change remains below the human body's perception threshold for heart rate changes, thus avoiding abruptness during the guidance process; the duration of the physiological guidance cycle varies with... The duration gradually increases as the heart rate decreases, achieving dynamic synchronization between the guided rhythm and the user's actual heart rate changes.

[0046] In one embodiment, step S3 further includes the following steps: The synchronization control signals for the visual, auditory, and tactile channels are all under the same clock during the physiological guidance cycle: S31. Taking the start time of the physiological guidance cycle as the zero point of timing, obtain the cumulative running time of the current time relative to the zero point of timing; S32. The cumulative running time is modulo the duration of the physiological guidance cycle to obtain the original normalized phase within the physiological guidance cycle at the current moment; wherein the value of the original normalized phase ranges from zero to one. S33. Obtain a preset rhythm traction coefficient, determine the second difference between the real-time heart rate data and the target guided heart rate, divide the second difference by the real-time heart rate data and multiply by the preset rhythm traction coefficient to obtain the phase offset. S34. Add the original normalized phase to the phase offset, take the modulo of one, and then determine the adjusted normalized phase; wherein the value range of the adjusted normalized phase is from zero to one; S35. Calculate the first ratio of the duration of the inhalation phase to the duration of the physiological guidance cycle, and use the first ratio as the proportion of the inhalation phase. S36. Determine whether the adjusted normalized phase is less than the proportion of the inhalation phase; if so, generate a visual channel control signal, an auditory channel control signal, and a tactile channel control signal corresponding to the inhalation phase. S37. If the adjusted normalized phase is greater than or equal to the proportion of the inhalation phase, then a visual channel control signal corresponding to the exhalation phase, an auditory channel control signal corresponding to the exhalation phase, and a tactile channel control signal corresponding to the exhalation phase are generated.

[0047] In this embodiment, the desk lamp uses a physiological guidance cycle. The starting time is the zero point of the timing, the current cumulative running time t is obtained, and the following sub-steps are executed to generate the multi-modal synchronization control signal.

[0048] The original normalized phase is calculated using the following formula: ; Where mod represents the modulo operation, such that It monotonically increases from 0 to 1 during each physiological guidance cycle, repeating the change periodically.

[0049] Introducing a rhythmic traction mechanism: calculating real-time heart rate data Heart rate guided by current target The second difference, divide the second difference by Then multiply by a preset rhythm traction coefficient λ, which is 0.1 in this embodiment, to obtain the phase offset. The calculation formula is: .

[0050] This phase offset Its function is to: when the user's current heart rate is higher than the guided heart rate, A positive value causes the adjusted phase to lag slightly, resulting in the device output rhythm being slightly slower than the user's own heart rate, creating a guiding effect.

[0051] Will and After adding them and taking the modulus of 1, we obtain the adjusted normalized phase, expressed as: , The value range is between 0 and 1.

[0052] Calculate the percentage of the inspiratory phase: In this embodiment =4 / 12=1 / 3.

[0053] when At that time, visual channel control signals, auditory channel control signals, and tactile channel control signals corresponding to the inhalation phase are generated; when At that time, visual channel control signals, auditory channel control signals, and tactile channel control signals corresponding to the exhalation phase are generated.

[0054] This embodiment establishes a normalized phase clock that is strictly synchronized with the guidance cycle through modulo operation, ensuring precise alignment of multimodal signals in the time dimension; by using rhythm-induced phase offset, the device outputs a rhythm that is slightly slower than the user's current heart rate, and utilizes the induction effect of external rhythm on the heart's natural pacing to achieve active traction of the heart rate.

[0055] As a further preferred embodiment, step S4, which involves synchronously increasing the illumination brightness, outputting audio rhythmic pulses, and increasing the tactile feedback intensity according to the synchronization control signal during the inhalation phase, specifically includes the following steps: The system receives a visual channel control signal corresponding to the inhalation phase, obtains the quotient obtained by dividing the adjusted normalized phase by the proportion of the inhalation phase, calculates and determines the target value of the illumination brightness during the inhalation phase using a first incrementing function, and then adjusts the output brightness of the illumination device to the target value of the illumination brightness during the inhalation phase; wherein, the first incrementing function is a quadratic function of the quotient, so that the target value of the illumination brightness during the inhalation phase increases rapidly as the adjusted normalized phase increases; The system receives an auditory channel control signal corresponding to the inhalation phase, acquires the guide beat interval of the physiological guidance cycle, determines the output time of the rhythmic pulse audio based on the guide beat interval, and controls the audio output module to play a pulse audio of a preset duration at the start time of each guide beat interval; wherein the output time of the rhythmic pulse audio is aligned with the start time of the cardiac cycle in the physiological guidance cycle. The system receives a tactile channel control signal corresponding to the inhalation phase, obtains the quotient obtained by dividing the adjusted normalized phase by the proportion of the inhalation phase, calculates and determines the target value of the tactile feedback intensity during the inhalation phase using a second incrementing function, and then adjusts the output intensity of the tactile feedback device to the target value of the tactile feedback intensity during the inhalation phase; wherein, the second incrementing function is a quadratic function of the quotient, so that the target value of the tactile feedback intensity during the inhalation phase increases rapidly as the adjusted normalized phase increases.

[0056] More preferably, step S4, which involves synchronously executing the decrease in illumination brightness, the restoration of continuous pulse-free audio output, and the cessation of tactile feedback during the exhalation phase according to the synchronization control signal, specifically includes the following steps: The system receives a visual channel control signal corresponding to the exhalation phase, calculates a first intermediate value and a second intermediate value, determines a second ratio of the first intermediate value to the second intermediate value, calculates a target value for illumination brightness during the exhalation phase using a first decreasing function, and then adjusts the output brightness of the illumination device to the target value for illumination brightness during the exhalation phase. The first intermediate value is the difference between the adjusted normalized phase and the proportion of the inhalation phase; the second intermediate value is the difference between 1 and the proportion of the inhalation phase; the first decreasing function is the square root function of the second ratio, causing the target value for illumination brightness during the exhalation phase to decrease at a slower rate as the adjusted normalized phase increases. It receives the auditory channel control signal corresponding to the exhalation phase, controls the audio output module to stop playing rhythmic pulse audio, and resumes playing continuous background audio without pulses; Receive the tactile channel control signal corresponding to the exhalation phase, adjust the output intensity of the tactile feedback device to zero, and stop the tactile feedback output.

[0057] In this embodiment, the execution module drives the visual, auditory, and tactile channels to achieve synchronous output of multimodal signals according to the synchronization control signal generated in step S3.

[0058] For the visual channel: During the inhalation phase, the ratio of the adjusted normalized phase to the proportion of the inhalation phase is obtained. The target illumination value for the inhalation phase is calculated according to a first increasing function, which is a quadratic function of this ratio, and its expression is: ; in, This represents the minimum brightness value, taken as 1 lux. This represents the maximum brightness value, taken as 8 lux. This indicates the target value for lighting brightness.

[0059] This function simulates the physiological sensation of gradually increasing muscle tension during prolonged contraction, causing the brightness to rise slowly in the early stages of inhalation and then accelerate in the later stages, giving the user the visual cue that inhalation is gradually deepening. The execution module adjusts the brightness of the main light source 6 to the target illumination value via the PWM dimming interface.

[0060] During the exhalation phase, the first intermediate value is calculated. The second median value is Obtain the ratio of the two values, and calculate the target brightness value for the exhalation phase using a first decreasing function, which is the square root function of the ratio, expressed as: ; This function causes the brightness to decrease rapidly in the early stage of exhalation and then slowly approach its lowest value in the later stage, simulating the relaxed sensation of the body gradually sinking into the support surface.

[0061] For the auditory channel: During the inhalation phase, speaker 5 is controlled to play rhythmic pulse audio synchronized with the guiding beat. Specifically, according to the interval of the guiding beat... The pulse output time is determined, and a pre-set monosyllabic beat sound, such as a simulated heartbeat "thump," is played at the beginning of each cardiac cycle, lasting approximately 200ms. This beat sound creates a loudness swell at the beginning of each beat, providing the user with a clear rhythmic anchor. During the exhalation phase, the speaker 5 stops playing the rhythmic pulse audio and resumes playing continuous, pulse-free background audio. In this embodiment, theta wave music with a characteristic frequency envelope of 4-7Hz or continuous pink noise is selected to enhance the relaxation cues during the exhalation phase.

[0062] For the tactile channel: During the inhalation phase, obtain the quotient obtained by dividing the adjusted normalized phase by the proportion of the inhalation phase, and calculate the target value of the tactile feedback intensity during the inhalation phase according to the second increasing function. The second increasing function is a quadratic function of the quotient, and its expression is: ; in, Vibration intensity corresponding to 80% PWM duty cycle.

[0063] The synchronization control module sends the target value of tactile feedback intensity and the corresponding timestamp to the wristband 3 via the Bluetooth communication module 11. The wristband microcontroller 16 drives the vibration motor 12 to vibrate according to the target value of tactile feedback intensity at a specified time. During the exhalation phase, the target value of tactile feedback intensity is instantaneously set to zero, and the vibration immediately stops, forming a tactile silence period. To compensate for the total delay Δt between Bluetooth transmission and the response of the wristband 3, in this embodiment, the total delay Δt is taken as 80 ms. The target phase corresponding to the sending command is calculated using the following formula: mod 1; in, For transmitting phase; This embodiment ensures that the time difference between the perception of light effect changes and tactile vibrations is less than 100 ms, which is in line with the time window for the fusion of human ear and tactile perception.

[0064] In this embodiment, during the inhalation phase, the lighting brightness increases exponentially with a quadratic function, the audio is output in a rhythmic pulse, and the tactile vibration is simultaneously enhanced. These three elements work together to apply a multi-sensory synergistic signal that deepens inhalation and accelerates the rhythm, guiding the user to actively deepen their inhalation. During the exhalation phase, the lighting brightness decreases decelerates with a square root function, the audio returns to continuous pulselessness, and the tactile vibration immediately stops. These three elements work together to create a relaxing and calming atmosphere, utilizing the physiological mechanism of respiratory sinus arrhythmia to enhance the parasympathetic nervous system's control over heart rate, thus promoting a decrease in heart rate. The forced synchronous changes of the three sensory channels under the same phase clock form a cross-sensory rhythmic traction field, maximizing the capture of the user's attention resources and forcibly guiding the user from the beta-wave excitation state induced by the screen content to a relaxation state dominated by alpha waves or even theta waves.

[0065] In one embodiment, the sleep preparation state described in step S5 is specifically determined by the following steps: Obtain updated real-time heart rate data and preset target resting heart rate, and determine the preset resting heart rate tolerance based on the preset target resting heart rate; Determine whether the real-time heart rate data is less than or equal to the preset tolerance of the resting heart rate; if so, determine that the user has entered a sleep preparation state.

[0066] In this embodiment, a preset heart rate update cycle is used. For 60 seconds and multimodal signal generation period It is 100 milliseconds, of which Greater than Every 60 seconds, step S1 is repeated to retrieve the latest real-time heart rate data uploaded by the wristband 3. ; Perform step S2, based on the updated Calculate the updated target guided heart rate and the updated physiological guidance cycle Within a 60-second interval between two consecutive heart rate updates, step S3 is executed cyclically every 100 milliseconds, based on the most recently determined... and Generate synchronization control signals and drive multi-mode output.

[0067] Continuously compare updated real-time heart rate data Compared to the preset target resting heart rate The relationship between the sum of (60 BPM in this embodiment) and the preset tolerance (2 BPM in this embodiment). When When the system determines that the user's heart rate has stabilized and is approaching the target resting level, and the autonomic nervous system has switched from the screen-excited mode to the sleep preparation mode, the system stops executing heart rate updates and multimodal output cycles, the brightness of the main light source 6 is reduced to 1 lux low light mode, the audio gradually weakens until it stops, the vibration of the bracelet 3 stops, and the system enters a low-power standby state.

[0068] This embodiment constructs a complete closed-loop control circuit of "monitoring, intervention, evaluation, and re-intervention" through nested cycles of heart rate update and signal generation. The dynamic adjustment of the guidance cycle is always based on the user's actual heart rate as the feedback benchmark, making the entire guidance process smooth: the guidance rhythm starts from the user's current heart rate and slowly pulls towards the target resting heart rate, without any rhythm abrupt changes or frequency conflicts throughout the process.

[0069] In one embodiment, this embodiment can reduce the stimulation of retinal cells by short-wavelength blue light from the screen to visual cells and reduce its inhibitory signal on melatonin secretion from the pineal gland by dynamic green light compensation and blue light filtering from the main light source 6 at the visual level; and reduce the semantic processing load of the language center and prefrontal cortex by real-time identification and spectrum masking of high-frequency human voices on the screen at the auditory level, thereby blocking the continuous arousal of the brain by video content.

[0070] This embodiment uses a secondary light source 7 to compensate for green light in the environment. In this embodiment, an ambient blue light sensor 8 monitors the proportion of blue light energy from the direction of the mobile phone screen in real time. When the proportion of energy in the 400-450nm band exceeds a preset threshold, the secondary light source 7 is controlled to emit green light with a peak wavelength of 500-550nm for compensation, while simultaneously suppressing the blue light output of the main light source 6. By increasing the proportion of long-wavelength components in the ambient light, the relative sensitivity of S-cone cells to short-wavelength blue light is reduced, thus mitigating the inhibitory effect of blue light on melatonin secretion from a physiological and optical perspective.

[0071] In this embodiment, ambient sound is collected in real time using the microphone array 4 built into the desk lamp. When a screen voice with characteristics of a human voice emitted from a screen device such as a mobile phone is detected and its loudness exceeds a threshold, pink noise with spectral matching is automatically generated for dynamic masking. The amplitude envelope of the masking noise is synchronously modulated with the heart rate guidance rhythm, and the relative gain of the masking noise is gradually increased during the guidance process, achieving a smooth transfer of auditory attention from the video content to the guidance rhythm. In this embodiment, the screen voice characteristic frequencies are concentrated in the 300Hz-3.4kHz range, have a formant structure, and the envelope changes are related to semantic rhythm.

[0072] In one embodiment, the RGB three-channel values ​​of ambient light detected by the ambient blue light sensor 8 are read and denoted as... , and The calculation of the blue light energy percentage is performed; it is then determined whether the blue light energy percentage in the ambient light exceeds a preset threshold. In this embodiment, the preset threshold is set to 30%, meaning that when the blue light energy percentage in the ambient light exceeds 30%, the screen is deemed to be emitting excessive blue light.

[0073] Calculate the proportion of green light flux that needs to be compensated: ;in, This indicates the proportion of green light flux that needs to be compensated. The percentage of blue light in the current ambient light. To achieve the ideal blue light percentage, we use 15% in this embodiment.

[0074] Calculate the PWM duty cycle of the secondary light source: 100%; of which, The duty cycle of the secondary light source PWM. To compensate for the gain coefficient, this embodiment uses 1.2. If Then clamp to 100%.

[0075] The auxiliary light source 7 emits light according to its PWM duty cycle. Simultaneously, the PWM duty cycle of the blue channel of the main light source 6 is set to 0%, which filters out the 400-450nm blue light emitted by the main light source 6 itself, thus avoiding secondary blue light pollution.

[0076] Through the algorithm described above, when a user is facing a screen, the green light component in the environment can be dynamically increased, causing the total spectrum entering the human eye to shift towards the longer wavelength band. This reduces the relative sensitivity of S-cone cells to short-wavelength blue light, thus mitigating the inhibition of melatonin secretion by blue light from a physiological and optical perspective.

[0077] In one embodiment, preferably, it further includes a screen high-frequency human voice recognition and active acoustic adversarial algorithm, as detailed below: Audio Acquisition and Preprocessing: The lamp uses its built-in microphone array 4 to continuously collect ambient audio signals at a 16kHz sampling rate and 16-bit quantization precision. .

[0078] The acquired signal is processed in frames with a frame length of 20ms, 320 sampling points are set, the frame shift is 10ms, and a Hamming window is added to each frame.

[0079] Screen voice feature recognition: The following feature parameters are extracted from each frame of signal: Centroid of the spectrum: Reflects the brightness of the sound. The centroid of human voices on the screen is usually located between 800Hz and 2kHz.

[0080] Spectral flux: the rate of change of the spectrum between adjacent frames. Video voices have a high flux variance due to semantic switching.

[0081] The first 12 dimensions of the Mel cepstral coefficients and their first difference are used to characterize the formant structure of human voice.

[0082] Zero crossing rate: reflects the proportion of high-frequency components in a signal.

[0083] The aforementioned feature vectors are input into a pre-trained lightweight classifier. In this embodiment, the lightweight classifier uses a support vector machine or a small fully connected neural network to output the probability that the current frame belongs to "screen voice". .

[0084] Calculate the loudness level of the current human voice: ; in, Indicates the current loudness of the human voice. This represents the effective value only for the human voice frequency band (300Hz-3.4kHz after bandpass filtering). Indicates a reference value. This is the correction factor.

[0085] When both of the following conditions are met simultaneously, it is determined to be a high-wake-up screen voice and active confrontation is triggered: Condition one: If the value is greater than 0.7, it is considered a high-confidence human voice.

[0086] Condition two: If the noise level is >45dB, it is considered a significant human voice that is above the ambient noise level.

[0087] Dynamic masking signal generation: When active adversarial actions are triggered, masking noise is generated. The basic type of masking noise is pink noise. In this embodiment, the pink noise has a 1 / f spectrum, which is consistent with the spectrum distribution of the natural environment. Its advantage is that it is more efficient at masking human voice frequencies (especially 300Hz-3.4kHz) than white noise.

[0088] Spectral matching enhancement. Analyze the spectral envelope of the current voice frame. Calculate the gain coefficient of the masking noise in this frequency band. : ; in, The normalized amplitude of the human voice spectrum. To enhance the effect, this embodiment uses a value of 0.8. This operation automatically boosts the energy of the masking noise in the frequency band where human voice energy is concentrated, achieving precise masking.

[0089] Rhythm modulation: During the heart rate-guided inspiratory phase, amplitude envelope modulation is applied to the masking noise. Modulation function for: ; in, To generate a unipolar pulse synchronized with the heart rate beat, each beat produces a half-sine wave bulge with a width of approximately 200 ms. In this embodiment, the modulation depth is set to 0.3. The modulated masking noise provides masking while simultaneously transmitting rhythmic guidance signals to the user.

[0090] Progressive blending and output: The hybrid weight is calculated based on the boot process. Let the total boot time be... The current time is The gain coefficient of the original video audio Gain coefficient for masking noise They are respectively: ; ; Initially, =0, original sound accounts for 70%, and masked sound accounts for 30%; at the end. The original sound accounts for only 30%, while the masked sound accounts for 70%. This achieves a smooth transfer of auditory attention.

[0091] Final output audio signal : ; in This is an ambient sound that retains the vocal component after bandpass filtering. This is masked noise that has been modulated by rhythm.

[0092] The audio signal is output to speaker 5.

[0093] This invention also provides a multimodal circadian rhythm traction sleep intervention system, which applies the multimodal circadian rhythm traction sleep intervention method described above, including: The data acquisition module is used to collect the user's real-time heart rate data; The calculation module, connected to the heart rate data acquisition module, is used to determine the target guiding heart rate based on the real-time heart rate data and generate a physiological guiding cycle including an inhalation phase and an exhalation phase, wherein the duration of the physiological guiding cycle is dynamically adjusted according to the change of the target guiding heart rate. A synchronization control module, connected to the computing module, is used to generate synchronization control signals corresponding to the visual channel, auditory channel and tactile channel respectively according to the physiological guidance cycle; An execution module, connected to the synchronization control module, is used to synchronously execute increasing illumination brightness, outputting audio rhythmic pulses, and increasing tactile feedback intensity according to the synchronization control signal during the inhalation phase; and synchronously execute decreasing illumination brightness, restoring continuous no-pulse audio output, and stopping tactile feedback according to the synchronization control signal during the exhalation phase. The closed-loop control module is connected to the data acquisition module and the calculation module respectively, and is used to control the data acquisition module, the calculation module, the synchronization control module and the execution module to repeatedly execute, dynamically adjust the physiological guidance cycle according to the real-time updated heart rate data, form a closed-loop control, until the user enters the sleep preparation state.

[0094] Furthermore, the execution module includes a desk lamp and a wristband 3. The desk lamp includes a lamp head 1 with a light source and a lamp holder 2 with a speaker 5. The desk lamp and the wristband 3 are communicatively connected. The lamp head 1 is rotatably connected to the lamp holder 2. The wristband 3 collects the user's heart rate data in real time. The light source performs increasing and decreasing illumination brightness. The speaker 5 performs audio rhythm pulse output. The wristband 3 has a built-in vibration motor 12, which performs tactile feedback.

[0095] Preferably, in this embodiment, the main light source 6 of the desk lamp is composed of multiple LED beads with adjustable color temperature. The color temperature adjustment range of the LED beads is 1800K to 3000K, and the color rendering index (CRI) is ≥95. The surface of the LED beads is covered with three layers of soft light diffusion film, namely a microprism layer, a matte layer, and an anti-reflection layer, to achieve a diffuse reflection light emission effect with an illuminance uniformity of ≥90%.

[0096] Sub-light source 7: Composed of narrowband LEDs with peak wavelength of 500-550nm and half-width ≤30nm, it is independent of the main light source 6 circuit and can accept PWM dimming control.

[0097] Ambient blue light sensor 8: Installed on the side of the desk lamp head 1 facing the user's screen, the sensing wavelength covers 400-450nm, and the sampling frequency is ≥10Hz. Ambient blue light sensor 8 can output RGB three-channel digital values.

[0098] Microphone Array 4: The desk lamp has at least two built-in digital microphones for directional amplification of sound sources from the direction of the screen device. The sampling rate is 16kHz, and the quantization precision is 16bit.

[0099] Speaker 5: Built-in audio codec chip, storing a first audio library, a second audio library and a beat sound library. The characteristic frequency envelope of the first audio library is 8-13Hz, such as alpha wave music and soothing instrumental music. The characteristic frequency envelope of the second audio library is 4-7Hz, such as theta wave music and low-frequency pink noise. The beat sound library stores monotone beat sounds like simulated heartbeats.

[0100] The MCU10 integrates a computing module, a synchronous control module, and a closed-loop control module: It adopts an ARM Cortex-M series microcontroller with a main frequency of ≥72MHz, and has built-in timers, PWM output channels, I2C interfaces, and UART interfaces.

[0101] Bluetooth communication module 11: Supports Bluetooth 5.0 and above protocols, supports low power broadcast mode and connection mode, and is used for time synchronization and command transmission with the wristband 3.

[0102] In this embodiment, the wristband 3 includes the following components: PPG heart rate sensor: Uses green LED9 and photodiode to collect pulse wave signals at the wrist at a sampling rate of 100Hz.

[0103] Vibration motor 12: adopts a linear resonant motor or an eccentric rotary motor, and supports continuously adjustable intensity.

[0104] Six-axis inertial measurement unit: integrates a three-axis accelerometer 14 and a three-axis gyroscope 13, used to detect the spatial attitude and motion state of the wristband 3.

[0105] Skin contact impedance detection electrode 15: arranged on the inner side of the wristband of the wristband 3 and the skin contact surface. The contact impedance value is calculated by injecting a weak alternating current into the skin and measuring the voltage drop.

[0106] Bluetooth communication module 11: Pairs with the Bluetooth communication module 11 of the desk lamp, receives instructions and time synchronization beacons from the desk lamp, and uploads real-time heart rate data.

[0107] The wristband microcontroller 16 is responsible for local sensor data acquisition, PPG signal preprocessing, heart rate calculation, vibration motor 12 driving, and Bluetooth communication module 11 operation.

[0108] Preferably, in this embodiment, the wristband 3 performs data fusion through the skin contact impedance detection electrode 15 and the six-axis inertial measurement unit to determine whether it is currently in a pressure state. The pressure state is, for example, when lying on its side, the wristband 3 is pressed between the pillow and the wrist. When it is determined to be in a pressure state, the vibration output intensity of the vibration motor 12 is automatically reduced to 30% of the normal level to avoid discomfort and the risk of waking up caused by tactile overload.

[0109] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A multimodal physiological rhythm-induced sleep intervention method, characterized in that, Includes the following steps: S1. Collect the user's real-time heart rate data; S2. Determine the target guiding heart rate based on the real-time heart rate data, and generate a physiological guiding cycle that includes an inhalation phase and an exhalation phase; wherein, the duration of the physiological guiding cycle is dynamically adjusted according to the change of the target guiding heart rate; S3. Generate synchronous control signals corresponding to the visual channel, auditory channel and tactile channel respectively according to the physiological guidance cycle; S4. During the inhalation phase, the illumination brightness, audio rhythm pulse output, and tactile feedback intensity are synchronously increased according to the synchronization control signal; during the exhalation phase, the illumination brightness is synchronously decreased, the audio output resumes continuous pulse-free output, and the tactile feedback stops according to the synchronization control signal. S5. Repeat steps S1 to S4, dynamically adjust the physiological guidance cycle based on the real-time updated heart rate data to form a closed-loop control until the user enters the sleep preparation state. Step S2 specifically includes the following steps: S21. Use the real-time heart rate data as the target guiding heart rate for the current moment; S22. Obtain the heart rate update cycle and the preset target resting heart rate. Every heart rate update cycle, determine whether the target guided heart rate at the current moment is greater than the preset target resting heart rate. If so, calculate the first difference between the target guided heart rate at the current moment and the preset target resting heart rate. S23. Obtain a preset decrease coefficient, determine the heart rate decrease range in this round based on the product of the first difference and the preset decrease coefficient, and use the target guided heart rate at the current moment minus the heart rate decrease range in this round as the updated target guided heart rate. S24. Determine the guiding beat interval based on the updated target guiding heart rate; wherein, the guiding beat interval is equal to sixty divided by the updated target guiding heart rate; S25. Obtain a preset number of inspiratory cardiac cycles, and determine the duration of the inspiratory phase based on the guide beat interval and the preset number of inspiratory cardiac cycles; wherein, the duration of the inspiratory phase is equal to the preset number of inspiratory cardiac cycles multiplied by the guide beat interval. S26. Obtain a preset number of expiratory cardiac cycles, and determine the duration of the expiratory phase based on the guide beat interval and the preset number of expiratory cardiac cycles; wherein, the duration of the expiratory phase is equal to the preset number of expiratory cardiac cycles multiplied by the guide beat interval, and the preset number of expiratory cardiac cycles is at least 1.5 times the preset number of inspiratory cardiac cycles. S27. The sum of the duration of the inhalation phase and the duration of the exhalation phase is taken as the duration of the physiological guidance cycle to obtain the updated physiological guidance cycle.

2. The multimodal physiological rhythm traction sleep intervention method according to claim 1, characterized in that, Step S1 specifically includes the following steps: S11. The wearable device's sensor collects the raw pulse wave signal at a first sampling frequency, and simultaneously collects triaxial acceleration data at the first sampling frequency through the wearable device's inertial measurement unit. S12. Calculate the synthetic acceleration amplitude based on the triaxial acceleration data, and use the synthetic acceleration amplitude as a reference noise source to filter and denoise the original pulse wave signal to obtain the filtered pulse wave signal. S13. Obtain the duration of the sliding window, perform peak detection on the filtered pulse wave signal based on the duration of the sliding window, determine the pulse wave peak in the filtered pulse wave signal, and calculate the time interval between adjacent pulse wave peaks. S14. Calculate the instantaneous heart rate value based on the time interval between adjacent pulse wave peaks, and determine the real-time heart rate data based on the instantaneous heart rate value.

3. The multimodal physiological rhythm traction sleep intervention method according to claim 2, characterized in that, Step S14 is followed by the following steps: Determine whether the number of pulse wave peaks within the sliding window duration is greater than or equal to a preset threshold. If not, mark the real-time heart rate data acquired at the current moment as low confidence and keep the previous effective heart rate value unchanged. The previous effective heart rate value refers to the most recent real-time heart rate data calculated based on pulse wave peaks that meet the preset threshold requirement before the current moment, which was not marked as low confidence.

4. The multimodal physiological rhythm traction sleep intervention method according to claim 1, characterized in that, Step S3 specifically includes the following steps: The synchronization control signals for the visual, auditory, and tactile channels are all under the same clock during the physiological guidance cycle: S31. Taking the start time of the physiological guidance cycle as the zero point of timing, obtain the cumulative running time of the current time relative to the zero point of timing; S32. The cumulative running time is modulo the duration of the physiological guidance cycle to obtain the original normalized phase within the physiological guidance cycle at the current moment; wherein the value of the original normalized phase ranges from zero to one. S33. Obtain a preset rhythm traction coefficient, determine the second difference between the real-time heart rate data and the target guided heart rate, divide the second difference by the real-time heart rate data and multiply by the preset rhythm traction coefficient to obtain the phase offset. S34. Add the original normalized phase to the phase offset, take the modulo of one, and then determine the adjusted normalized phase; wherein the value range of the adjusted normalized phase is from zero to one; S35. Calculate the first ratio of the duration of the inhalation phase to the duration of the physiological guidance cycle, and use the first ratio as the proportion of the inhalation phase. S36. Determine whether the adjusted normalized phase is less than the proportion of the inhalation phase; if so, generate a visual channel control signal, an auditory channel control signal, and a tactile channel control signal corresponding to the inhalation phase. S37. If the adjusted normalized phase is greater than or equal to the proportion of the inhalation phase, then a visual channel control signal corresponding to the exhalation phase, an auditory channel control signal corresponding to the exhalation phase, and a tactile channel control signal corresponding to the exhalation phase are generated.

5. The multimodal physiological rhythm traction sleep intervention method according to claim 4, characterized in that, Step S4, which describes the synchronous execution of increasing illumination brightness, outputting audio rhythmic pulses, and increasing tactile feedback intensity during the inhalation phase according to the synchronization control signal, specifically includes the following steps: The system receives a visual channel control signal corresponding to the inhalation phase, obtains the quotient obtained by dividing the adjusted normalized phase by the proportion of the inhalation phase, calculates and determines the target value of the illumination brightness during the inhalation phase using a first incrementing function, and then adjusts the output brightness of the illumination device to the target value of the illumination brightness during the inhalation phase; wherein, the first incrementing function is a quadratic function of the quotient, so that the target value of the illumination brightness during the inhalation phase increases rapidly as the adjusted normalized phase increases; The system receives an auditory channel control signal corresponding to the inhalation phase, acquires the guide beat interval of the physiological guidance cycle, determines the output time of the rhythmic pulse audio based on the guide beat interval, and controls the audio output module to play a pulse audio of a preset duration at the start time of each guide beat interval; wherein the output time of the rhythmic pulse audio is aligned with the start time of the cardiac cycle in the physiological guidance cycle. The system receives a tactile channel control signal corresponding to the inhalation phase, obtains the quotient obtained by dividing the adjusted normalized phase by the proportion of the inhalation phase, calculates and determines the target value of the tactile feedback intensity during the inhalation phase using a second incrementing function, and then adjusts the output intensity of the tactile feedback device to the target value of the tactile feedback intensity during the inhalation phase; wherein, the second incrementing function is a quadratic function of the quotient, so that the target value of the tactile feedback intensity during the inhalation phase increases rapidly as the adjusted normalized phase increases.

6. The multimodal physiological rhythm traction sleep intervention method according to claim 4, characterized in that, Step S4, which describes the synchronous execution of decreasing illumination brightness, restoring continuous pulse-free audio output, and stopping tactile feedback during the exhalation phase according to the synchronization control signal, specifically includes the following steps: The system receives a visual channel control signal corresponding to the exhalation phase, calculates a first intermediate value and a second intermediate value, determines a second ratio of the first intermediate value to the second intermediate value, calculates a target value for illumination brightness during the exhalation phase using a first decreasing function, and then adjusts the output brightness of the illumination device to the target value for illumination brightness during the exhalation phase. The first intermediate value is the difference between the adjusted normalized phase and the proportion of the inhalation phase; the second intermediate value is the difference between 1 and the proportion of the inhalation phase; the first decreasing function is the square root function of the second ratio, causing the target value for illumination brightness during the exhalation phase to decrease at a slower rate as the adjusted normalized phase increases. It receives the auditory channel control signal corresponding to the exhalation phase, controls the audio output module to stop playing rhythmic pulse audio, and resumes playing continuous background audio without pulses; Receive the tactile channel control signal corresponding to the exhalation phase, adjust the output intensity of the tactile feedback device to zero, and stop the tactile feedback output.

7. The multimodal physiological rhythm traction sleep intervention method according to claim 1, characterized in that, The sleep preparation state described in step S5 is specifically determined by the following steps: Obtain updated real-time heart rate data and preset target resting heart rate, and determine the preset resting heart rate tolerance based on the preset target resting heart rate; Determine whether the real-time heart rate data is less than or equal to the preset tolerance of the resting heart rate; If so, the user is determined to be in a sleep preparation state.

8. A multimodal circadian rhythm traction sleep intervention system, employing the multimodal circadian rhythm traction sleep intervention method as described in any one of claims 1-7, characterized in that, include: The data acquisition module is used to collect the user's real-time heart rate data; The calculation module, connected to the data acquisition module, is used to determine the target guiding heart rate based on the real-time heart rate data and generate a physiological guiding cycle that includes an inhalation phase and an exhalation phase, wherein the duration of the physiological guiding cycle is dynamically adjusted according to the change of the target guiding heart rate. A synchronization control module, connected to the computing module, is used to generate synchronization control signals corresponding to the visual channel, auditory channel and tactile channel respectively according to the physiological guidance cycle; An execution module, connected to the synchronization control module, is used to synchronously execute increasing illumination brightness, outputting audio rhythmic pulses, and increasing tactile feedback intensity according to the synchronization control signal during the inhalation phase; and synchronously execute decreasing illumination brightness, restoring continuous no-pulse audio output, and stopping tactile feedback according to the synchronization control signal during the exhalation phase. The closed-loop control module is connected to the data acquisition module and the calculation module respectively, and is used to control the data acquisition module, the calculation module, the synchronization control module and the execution module to repeatedly execute, dynamically adjust the physiological guidance cycle according to the real-time updated heart rate data, form a closed-loop control, until the user enters the sleep preparation state.

9. The multimodal physiological rhythm traction sleep intervention system according to claim 8, characterized in that: The execution module includes a desk lamp and a wristband. The desk lamp includes a lamp head with a light source and a lamp holder with a speaker. The desk lamp and the wristband are communicatively connected. The lamp head is rotatably connected to the lamp holder. The wristband collects the user's heart rate data in real time. The light source performs increasing and decreasing illumination brightness. The speaker performs audio rhythm pulse output. The wristband has a built-in vibration motor that performs tactile feedback.

Images