Method and system for adjusting immersive sleep scenario based on rhythm planning analysis

Abstract

The present application relates to the technical field of sleep scene regulation, in particular to an immersive sleep scene regulation method and system based on rhythm planning analysis. Real-time physiological feedback data and immersive sleep state data of a target user are obtained to construct a real-time sleepiness index curve, and the sleepiness rhythm of the saturated state and the unsaturated state is mapped and planned in a sliding sub-window to obtain a real-time sleep rhythm phase. A preset creation control strategy of the immersive sleep scene is simulated by simulation software, and the response offset analysis scene of the rhythm phase is determined by combining electroencephalogram signal inversion calculation. When the analysis result is a rhythm confusion scene, the real-time rhythm phase under the sleep staging mode is used as a driving benchmark to calculate the phase time difference and design an amplitude compensation mechanism to adjust the parameters of the chaos scene factor. The present application can realize dynamic and personalized regulation and control of the immersive sleep scene, effectively improve the rhythm confusion or inertia problem, and improve the immersive sleep quality of the user.

Description

Technical Field

[0001] This invention relates to the field of sleep scene regulation technology, and in particular to a method and system for regulating immersive sleep scenes based on rhythm planning analysis. Background Technology

[0002] With the development of immersive technology, multimodal interaction has been widely applied in the field of sleep assistance. Existing methods for regulating immersive sleep scenarios typically create single or multiple combinations of external factors such as light, sound, smell, or ambient temperature to induce relaxation and improve sleep. However, human sleep is inherently influenced by a dual-process regulatory mechanism of circadian rhythm and sleep homeostasis, and users' sleep rhythm states are highly individualized and dynamically changing. Existing regulation technologies lack in-depth utilization of users' real-time physiological feedback and sleep state data, making it difficult to accurately track changes in users' sleepiness in real time, thus failing to accurately plan circadian phases that conform to users' sleep habits. Secondly, the lack of a dynamic mechanism for planning sleep rhythm phases based on real-time sleepiness concentration makes it impossible to accurately create circadian phase profiles that conform to users' sleep habits, resulting in lagging scenario regulation and an inability to effectively address circadian rhythm disorder or inertia. Furthermore, in scenarios with abnormal circadian rhythms, existing regulation methods lack refined regulation strategies for the two differentiated states of scenario guidance disorder and inertia, often resulting in coarse regulation with limited effects, which may even reduce users' sleep quality. Therefore, there is an urgent need for a method to regulate immersive sleep scenarios that can be combined with real-time rhythm planning and analysis, so as to achieve personalized, dynamic and refined scenario regulation interventions to improve users' experience and stability of immersive sleep. Summary of the Invention

[0003] This invention overcomes the shortcomings of the prior art and provides a method and system for regulating immersive sleep scenarios based on rhythm planning analysis.

[0004] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0005] The first aspect of this invention provides a method for regulating immersive sleep scenarios based on rhythmic programming analysis, comprising the following steps:

[0006] S102: Obtain real-time physiological feedback data and immersive sleep state data of the target user to track real-time sleepiness, obtain a real-time sleepiness index curve, and map and plan the sleepiness rhythm of saturated and unsaturated states within the sleep sliding sub-window based on the real-time sleepiness index curve to obtain the real-time sleep rhythm phase of the target user.

[0007] S104: Simulate the preset creation control strategy of the current immersive sleep scene using simulation software to shake the real-time sleep rhythm phase. Calculate the response shift of the current immersive sleep scene to the rhythm phase based on the simulation results and the actual EEG signal sequence. Analyze the current immersive sleep scene based on the phase response drift function between the generated current sleep rhythm phase and the real-time sleep rhythm phase to obtain the rhythm planning analysis results.

[0008] S106: If the rhythm planning analysis results show a rhythm disorder scenario, then by using the real-time sleep rhythm phase based on the sleep stage modality as the driving benchmark of the guiding event to align to the rhythm disorder scenario modality, the phase timing difference is obtained, and the amplitude compensation design based on the phase timing difference is used to adjust the creation parameters of different scenario factors of the rhythm disorder scenario to generate the first adjustment scheme.

[0009] S108: If the rhythm planning analysis results show a rhythm inertia scenario, then construct a multi-dimensional parameter-tuning polar coordinate model. Based on the inertia delay of the phase response drift in the current rhythm inertia scenario, analyze the adjustment strategy of the scenario factor on the multi-dimensional parameter-tuning polar coordinate model to adjust the creation parameters of the rhythm inertia scenario and obtain the second adjustment scheme.

[0010] Preferably, step S102 specifically includes the following steps:

[0011] Real-time physiological feedback data of target users is collected through the Internet of Things, and sleep logs are used to extract data on the immersive sleep state of target users when real-time physiological feedback data is generated in historical time periods.

[0012] A biological sleep dynamics state equation is established by introducing the biological sleep dynamics mechanism. Based on the sleep plasticity of real-time physiological feedback data of sleep homeostasis assessment, the real-time sleepiness of the target user is tracked and updated in the biological sleep dynamics state equation along the tangent direction of the immersive sleep state data, so as to obtain the real-time sleepiness index curve of sleep time series.

[0013] We construct a current circadian rhythm homeostasis sleep sliding window domain, formulate a variance weighting criterion for the sleep sliding window domain based on the short-term sleep driving mechanism, and divide the sleep sliding window domain into several sleep sliding sub-windows based on the sleep behavior information of the target user.

[0014] Based on the variance weighting criterion, the kurtosis weighted calculation of the slope steepness of the real-time sleepiness index curve within each sleep sliding window framework is performed to obtain the local sleepiness kurtosis decay value of each sleep sliding window.

[0015] If the local sleepiness kurtosis decay value is greater than the preset threshold, the peak-valley curve span corresponding to the sleep sliding sub-window is marked as the saturation cutoff segment; if it is less than the preset threshold, it is marked as the unsaturation cutoff segment.

[0016] The real-time sleepiness index of the saturation cutoff segment and the unsaturation cutoff segment is obtained by real-time sleepiness index curve. The center point of the saturation cutoff segment is located and labeled as the first proxy anchor point. The center point of the saturation cutoff segment is located and labeled as the second proxy anchor point.

[0017] At each first and second proxy anchor point, the average height of the rhythm function corresponding to each saturated and unsaturated cutoff segment of the real-time sleepiness index is calculated. The explicit Euler method is introduced to perform a time-series approximate integration of the average height of the rhythm function to generate the rhythm area under the real-time sleepiness index curve.

[0018] Obtain sleep staging rules, and perform phase mapping planning of sleep staging rules based on the surrogate interleaving distribution of sleepiness saturation integral - unsaturation integral in the area model under rhythm, to obtain the real-time sleep rhythm phase of the target user.

[0019] Preferably, the step of introducing a biological sleep dynamics mechanism to establish a biological sleep dynamics state equation, and tracking and updating the target user's real-time sleepiness in the biological sleep dynamics state equation along the tangent direction of the immersive sleep state data based on the sleep plasticity of real-time physiological feedback data of sleep homeostasis assessment, to obtain a real-time sleepiness index curve of sleep time sequence, specifically includes the following steps:

[0020] Based on big data retrieval, a rhythmic sleep index system is obtained. The biological sleep dynamics mechanism is introduced. The biological sleep dynamics mechanism is used to model the dynamic state control of real-time physiological feedback data with immersive sleep state data as the state variable input, and a nonlinear biological sleep dynamic state equation is generated.

[0021] Sleep homeostasis indicators are extracted through a circadian rhythm sleep index system. A series of sleep homeostasis indicators are used to evaluate real-time physiological feedback data in multiple dimensions to obtain the sleep plasticity coefficient of the target user. The tracking resolution step size is preset based on the sleep plasticity coefficient.

[0022] Based on the tangent direction of the immersive sleep state data as a linear approximation, the balance prediction value on the acquisition time sequence is calculated along the tangent direction based on the tracking resolution step size. The balance prediction value is substituted into the biological sleep dynamic state equation to predict the support position of the next dynamic equilibrium point, thus obtaining the dynamic iteration equilibrium fulcrum.

[0023] A Jacobian matrix is ​​calculated based on the dynamic change equilibrium pivot to track and update the tangent direction. The candidate tangent trajectory feature values ​​are output during the continuous update process of the Jacobian matrix. The real-time sleepiness index of the target user is determined based on the candidate tangent trajectory feature values, and a real-time sleepiness index curve on the sleep time series is constructed by fitting.

[0024] Preferably, step S104 specifically includes the following steps:

[0025] By retrieving the preset creation and control strategies of the current immersive sleep scene through the Internet of Things, a dynamic simulation model of the current immersive sleep scene is constructed using sleep scene simulation software.

[0026] A preset control strategy is used to inject jitter into a dynamic simulation model. The real-time sleep rhythm phase is simulated by the injected dynamic simulation model. During the simulation, the target user's EEG signal is recorded and the simulated EEG signal sequence is output.

[0027] The actual EEG signal sequence of the target user in the current immersive sleep scenario is obtained, a time-frequency domain registration space is constructed, and the time-frequency domain registration space is divided into N sub-registration blocks based on the predetermined stage of sleep stage.

[0028] A short-time Fourier transform algorithm is introduced to perform rhythmic response transformation calculation on the actual EEG signal sequence to obtain the actual response spectrum waveform. In the time-frequency domain registration space, the simulated EEG signal sequence and the actual EEG signal sequence are registered.

[0029] Only simulated response spectrum waveforms and corresponding sub-registration blocks that have a similarity greater than a preset similarity threshold to the actual response spectrum waveform in the simulated EEG signal sequence are extracted and marked as response registration blocks. The registration overlap rate between the simulated response spectrum waveform and the actual response spectrum waveform is obtained through the response registration blocks.

[0030] The response spectrum distortion coefficients of the target user's response to the real-time sleep rhythm phase under the current immersive sleep scene jitter are determined based on the registration and superposition rate, and the Hilbert instantaneous phase transformation matrix is ​​constructed based on the response spectrum distortion coefficients.

[0031] A phase oscillation equation is introduced, and the response shift of the real-time sleep rhythm phase is smoothed by observation and analysis of scattered points in the phase oscillation equation through the Hilbert instantaneous phase transformation matrix, and then inverted to generate a simulated sleep rhythm phase.

[0032] If the phase response drift function between the simulated sleep rhythm phase and the real-time sleep rhythm phase is positive, then the current immersive sleep scenario in which the real-time sleep rhythm phase is located is marked as a rhythmic chaotic scenario; if the phase response drift function between the simulated sleep rhythm phase and the real-time sleep rhythm phase is negative, then the current immersive sleep scenario is marked as a rhythmic inertial scenario, and the rhythm planning analysis results are obtained.

[0033] Preferably, step S106 specifically includes the following steps:

[0034] If the rhythm planning analysis results show a rhythmic chaos scenario, then the immersive sleep scenario creation system will be used to obtain the phased guidance decision information of different scenario factors in the rhythmic chaos scenario for the target user to enter the immersive sleep mode.

[0035] The architecture of using sleep staging rules as the baseline modal scale is embedded to construct a stage alignment grid for heterogeneous modalities. The real-time sleep rhythm phase is set as the guiding window event, and the trigger timestamp of the sleep staging rule when the guiding window event occurs is obtained.

[0036] Based on the phased guidance decision information, the rhythmic chaos scene is aligned to the phase alignment grid around the trigger timestamp as the alignment benchmark. At the same time, during the alignment process, the modal alignment range between sleep phase and rhythmic chaos scene is traversed to obtain the phase timing difference of rhythmic chaos scene.

[0037] Obtain the open-loop transfer function of the immersive sleep scene creation system to control the creation of a circadian chaotic scene, draw the Bode open-loop frequency diagram of the circadian chaotic scene based on the open-loop transfer function, and identify the current low-frequency gain of the system in creating the circadian chaotic scene by analyzing the Bode open-loop frequency diagram.

[0038] Based on the phase timing differential preset maximum low frequency gain requirement, if the current low frequency gain is insufficient for the maximum low frequency gain requirement, then an amplitude adjustment compensator is deployed and constructed in the immersive sleep scene creation system, and the boundary cutoff frequency of the amplitude adjustment compensator is constrained based on the maximum low frequency gain requirement.

[0039] Based on the current low-frequency gain, an amplitude adjustment compensator is used to construct the adjustment correction value for the rhythmic chaos scene. The current cutoff frequency is obtained, and the adjustment correction value is applied to the immersive sleep scene creation system to adjust the creation parameters of each scene factor in the rhythmic chaos scene until the current cutoff frequency approaches the boundary cutoff frequency, thus obtaining the first adjustment scheme.

[0040] Preferably, step S108 specifically includes the following steps:

[0041] If the rhythm planning analysis results show a rhythmic inertia scenario, then the immersive sleep scenario creation system is used to obtain several preparatory creation parameters for different scenario factors in the rhythmic inertia scenario, and the corresponding intervention function of each preparatory creation parameter to promote immersive sleep.

[0042] Based on different scenario factors, a parameter-tuning polar coordinate branch is established. According to the intervention function, a fixed-point coordinate network of the reserve construction parameters located in the parameter-tuning polar coordinate branch is woven. The K-means clustering algorithm is introduced to cluster each reserve construction parameter to a network node of the fixed-point coordinate network. An inert counter is assigned to each fixed-point coordinate to generate a multi-dimensional parameter-tuning polar coordinate model of the rhythmic inert scenario.

[0043] Obtain the real-time reserve construction parameters of the rhythmic inertial scenario under the current working condition, find the fixed point coordinates of each real-time reserve construction parameter in the multi-dimensional adjustment coordinate model and connect them to obtain the real-time parameter adjustment polar coordinate map.

[0044] By analyzing the inertial accumulation exponent of the inertial counter, the delay weight of the real-time parameter-tuned polar coordinate layout for the phase response drift function is analyzed. Based on the delay weight, sampling planning is carried out starting from the real-time parameter-tuned polar coordinate layout as the starting center to obtain the auxiliary sampling layout.

[0045] Extract the polar coordinate branch of the parameter adjustment covered by the auxiliary sampling map, the corresponding fixed point coordinates and the corresponding reserve construction parameters, and define them as candidate adjustment parameter solutions. At the same time, preset the jitter residual gradient threshold so that the phase response drift function approaches 0.

[0046] If the jitter residual gradient of the fixed-point coordinate does not exceed the jitter residual gradient threshold, the update of the candidate adjustment parameter solution for that fixed-point coordinate is skipped; if the jitter residual gradient of the fixed-point coordinate exceeds the jitter residual gradient threshold, the update of the corresponding candidate adjustment parameter solution is applied to obtain the adjustment strategy for each scene factor.

[0047] Based on the adjustment strategy, the creation parameters of each scene factor in the immersive sleep scene creation system are adjusted to obtain the second adjustment scheme.

[0048] A second aspect of the present invention provides an immersive sleep scene adjustment system based on rhythm planning analysis. The system includes: a memory, a processor, and a communication interface. The memory includes a program for an immersive sleep scene adjustment method based on rhythm planning analysis. The communication interface is used for data connection and communication between the memory and the processor. When the program for an immersive sleep scene adjustment method based on rhythm planning analysis is executed by the processor, it implements the steps of the immersive sleep scene adjustment method described in any one of the present invention.

[0049] This invention addresses the technical deficiencies in the prior art, and its beneficial technical effects are as follows:

[0050] This invention acquires real-time physiological feedback data and immersive sleep state data of target users, analyzes and tracks real-time sleepiness index curves, and performs rhythm phase mapping planning to accurately locate the user's sleep rhythm dynamics. By simulating the creation strategy of immersive sleep scenarios using simulation software and combining it with EEG signals for inversion calculations, the actual response offset of the immersive scenario to the rhythm phase can be effectively obtained, achieving precise planning and analysis of the rhythm phase. When rhythmic disorder scenarios occur, based on the driving benchmark alignment and amplitude compensation mechanism of real-time rhythm phase, the creation parameters of different scenario factors can be adjusted in a targeted manner to restore rhythm stability and reduce rhythmic orientation disorder caused by inappropriate adjustment of the immersive sleep scenario. When rhythmic inertia scenarios occur, the inertia delay is analyzed through a multi-dimensional parametric polar coordinate model, providing targeted adjustment strategies to effectively improve rhythmic response lag and reduce the low adaptability of real-time rhythm to the current immersive sleep scenario. This invention enables personalized, dynamic, and refined adjustment in immersive sleep scenarios, significantly improving the user's sleep rhythm coordination and immersive sleep quality. Attached Figure Description

[0051] 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 embodiments can be obtained from these drawings without creative effort.

[0052] Figure 1 A flowchart of the first method for regulating immersive sleep scenarios based on rhythmic planning analysis is shown;

[0053] Figure 2 A flowchart of the second method for regulating immersive sleep scenarios based on rhythmic planning analysis is shown;

[0054] Figure 3 A system framework diagram of an immersive sleep scene regulation system based on rhythm planning analysis is shown. Detailed Implementation

[0055] To better understand the above-mentioned objectives, features, and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.

[0056] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and therefore the scope of protection of the invention is not limited to the specific embodiments disclosed below.

[0057] The first aspect of this invention provides a method for regulating immersive sleep scenarios based on rhythmic programming analysis, such as... Figure 1 As shown, it includes the following steps:

[0058] S102: Obtain real-time physiological feedback data and immersive sleep state data of the target user to track real-time sleepiness, obtain a real-time sleepiness index curve, and map and plan the sleepiness rhythm of saturated and unsaturated states within the sleep sliding sub-window based on the real-time sleepiness index curve to obtain the real-time sleep rhythm phase of the target user.

[0059] S104: Simulate the preset creation control strategy of the current immersive sleep scene using simulation software to shake the real-time sleep rhythm phase. Calculate the response shift of the current immersive sleep scene to the rhythm phase based on the simulation results and the actual EEG signal sequence. Analyze the current immersive sleep scene based on the phase response drift function between the generated current sleep rhythm phase and the real-time sleep rhythm phase to obtain the rhythm planning analysis results.

[0060] S106: If the rhythm planning analysis results show a rhythm disorder scenario, then by using the real-time sleep rhythm phase based on the sleep stage modality as the driving benchmark of the guiding event to align to the rhythm disorder scenario modality, the phase timing difference is obtained, and the amplitude compensation design based on the phase timing difference is used to adjust the creation parameters of different scenario factors of the rhythm disorder scenario to generate the first adjustment scheme.

[0061] S108: If the rhythm planning analysis results show a rhythm inertia scenario, then construct a multi-dimensional parameter-tuning polar coordinate model. Based on the inertia delay of the phase response drift in the current rhythm inertia scenario, analyze the adjustment strategy of the scenario factor on the multi-dimensional parameter-tuning polar coordinate model to adjust the creation parameters of the rhythm inertia scenario and obtain the second adjustment scheme.

[0062] It should be noted that different scene factors include lighting (blue light), background sound (white noise, natural sound, and breathing guidance sound), temperature (air conditioner, air heater), humidity (humidifier), bed vibration (mattress, pillow), and the release of sleep aids (fragrance).

[0063] Preferably, step S102 specifically includes the following steps:

[0064] Real-time physiological feedback data of target users is collected through the Internet of Things, and sleep logs are used to extract data on the immersive sleep state of target users when real-time physiological feedback data is generated in historical time periods.

[0065] A biological sleep dynamics state equation is established by introducing the biological sleep dynamics mechanism. Based on the sleep plasticity of real-time physiological feedback data of sleep homeostasis assessment, the real-time sleepiness of the target user is tracked and updated in the biological sleep dynamics state equation along the tangent direction of the immersive sleep state data, so as to obtain the real-time sleepiness index curve of sleep time series.

[0066] We construct a current circadian rhythm homeostasis sleep sliding window domain, formulate a variance weighting criterion for the sleep sliding window domain based on the short-term sleep driving mechanism, and divide the sleep sliding window domain into several sleep sliding sub-windows based on the sleep behavior information of the target user.

[0067] Based on the variance weighting criterion, the kurtosis weighted calculation of the slope steepness of the real-time sleepiness index curve within each sleep sliding window framework is performed to obtain the local sleepiness kurtosis decay value of each sleep sliding window.

[0068] If the local sleepiness kurtosis decay value is greater than the preset threshold, the peak-valley curve span corresponding to the sleep sliding sub-window is marked as the saturation cutoff segment; if it is less than the preset threshold, it is marked as the unsaturation cutoff segment.

[0069] The real-time sleepiness index of the saturation cutoff segment and the unsaturation cutoff segment is obtained by real-time sleepiness index curve. The center point of the saturation cutoff segment is located and labeled as the first proxy anchor point. The center point of the saturation cutoff segment is located and labeled as the second proxy anchor point.

[0070] At each first and second proxy anchor point, the average height of the rhythm function corresponding to each saturated and unsaturated cutoff segment of the real-time sleepiness index is calculated. The explicit Euler method is introduced to perform a time-series approximate integration of the average height of the rhythm function to generate the rhythm area under the real-time sleepiness index curve.

[0071] Obtain sleep staging rules, and perform phase mapping planning of sleep staging rules based on the surrogate interleaving distribution of sleepiness saturation integral - unsaturation integral in the area model under rhythm, to obtain the real-time sleep rhythm phase of the target user.

[0072] It should be noted that real-time physiological feedback data includes heart rate, heart rate variability parameters, respiratory rate, adenosine levels, body movement parameters, and skin temperature parameters. Immersive sleep state data includes neural activity data, skin conductance data, electromyography (EMG) data, eye movement data, and body movement data. Since the sleep rhythm of the target user is not a fixed pattern but is determined by multiple external factors, such as short-term mental or physical exercise before sleep, the adenosine level rises rapidly from low to high, resulting in a sharp increase in actual sleepiness after exercise. Therefore, the user's real-time sleep rhythm may be advanced as a result. However, existing regulation methods struggle to reasonably predict and plan the rhythm in real time based on the user's sleepiness, leading to inaccurate regulation of the immersive sleep scenario for different users' individual rhythms. Therefore, this method first uses a biological sleep dynamics state equation established based on biological sleep dynamics mechanisms to conduct sleep plasticity analysis on the target user's real-time physiological feedback data and immersive sleep state data generated over historical time periods. This further tracks the target user's real-time sleepiness trend during the sleep sequence, i.e., the real-time sleepiness index curve. As the longer the awake time, the greater the brain's need for sleep, sleepiness pressure gradually accumulates. This pressure is gradually released upon entering the sleep stage, demonstrating that real-time sleepiness is driven by sleep homeostasis. Therefore, this method further develops a variance-weighted criterion for the sleep sliding window domain based on the short-term sleep driving mechanism. This criterion effectively removes local offsets in the kurtosis calculation of the real-time sleepiness index curve within each subsequent sleep sliding sub-window, improving the accuracy of the curve's representation of the sleepiness decay kurtosis distribution. Furthermore, this method, through kurtosis weighting of the real-time sleepiness index curve using micro-windowing, can quantify the tail thickness and outlier tendency of the window's sleepiness pressure concentration as the target user progresses with time gradients. This allows for more sensitive capture of the timing pulse of rhythm entry, significantly improving the accuracy of subsequent real-time planning of personalized sleep rhythms based on the user's sleepiness concentration gradient.

[0073] It should be noted that if the local sleepiness kurtosis decay value is greater than the preset threshold, it indicates that the peak-valley curve span depicted by the real-time sleepiness index curve at that moment is large. This suggests that the target user's sleepiness concentration at that moment has reached an extreme saturation state and is trending towards unsaturation decay, representing a relatively strong sleepiness. Therefore, the peak-valley curve span corresponding to this sleep sliding sub-window is designated as the saturation cutoff segment. Conversely, a large peak-valley curve span indicates that the target user has a low level of sleepiness at that moment, and is therefore designated as the unsaturation cutoff segment. The first and second proxy anchor points serve as representative base points for the saturation and unsaturation cutoff segments, respectively. Approximate estimation of the rhythm phase using these anchor points balances the integral values ​​at the left and right endpoints of the curve span, effectively offsetting some cumulative deviations at both ends of the interval and improving the reliability of rhythm planning based on sleepiness gradient changes. The average height of the rhythm function is the function height value that best represents the average sleepiness concentration of the saturation and unsaturation cutoff segments. It is important to note that the sleep rhythm planned by this method is based on the turning points of sleep stage rules, such as the timing of the transition from N1 light sleep to N2 light sleep in the sleep lead rhythm. This method can track a user's real-time sleepiness and use sleep homeostasis mechanisms as key rules to plan the sleep rhythm phases with high precision. This allows for the creation of unique rhythm positioning for users with different sleep habits, providing a personalized adjustment basis for subsequent immersive sleep scenarios and meeting the sleep assistance needs of different user groups.

[0074] Preferably, the step of introducing a biological sleep dynamics mechanism to establish a biological sleep dynamics state equation, and tracking and updating the target user's real-time sleepiness in the biological sleep dynamics state equation along the tangent direction of the immersive sleep state data based on the sleep plasticity of real-time physiological feedback data of sleep homeostasis assessment, to obtain a real-time sleepiness index curve of sleep time sequence, specifically includes the following steps:

[0075] Based on big data retrieval, a rhythmic sleep index system is obtained. The biological sleep dynamics mechanism is introduced. The biological sleep dynamics mechanism is used to model the dynamic state control of real-time physiological feedback data with immersive sleep state data as the state variable input, and a nonlinear biological sleep dynamic state equation is generated.

[0076] Sleep homeostasis indicators are extracted through a circadian rhythm sleep index system. A series of sleep homeostasis indicators are used to evaluate real-time physiological feedback data in multiple dimensions to obtain the sleep plasticity coefficient of the target user. The tracking resolution step size is preset based on the sleep plasticity coefficient.

[0077] Based on the tangent direction of the immersive sleep state data as a linear approximation, the balance prediction value on the acquisition time sequence is calculated along the tangent direction based on the tracking resolution step size. The balance prediction value is substituted into the biological sleep dynamic state equation to predict the support position of the next dynamic equilibrium point, thus obtaining the dynamic iteration equilibrium fulcrum.

[0078] A Jacobian matrix is ​​calculated based on the dynamic change equilibrium pivot to track and update the tangent direction. The candidate tangent trajectory feature values ​​are output during the continuous update process of the Jacobian matrix. The real-time sleepiness index of the target user is determined based on the candidate tangent trajectory feature values, and a real-time sleepiness index curve on the sleep time series is constructed by fitting.

[0079] It should be noted that for real-time sleepiness tracking of different target users, this method uses sleep homeostasis indicators to independently and jointly evaluate the real-time physiological feedback data generated by the target users based on different dimensional benchmarks. For example, it assesses the brain's sleepiness pressure tendency when the user has abnormal heart rate variability parameters, and the adenosine accumulation level when heart rate variability parameters and respiratory rate fluctuations occur simultaneously. This allows for plasticity analysis of the user's intention to fall asleep, and a sleep plasticity coefficient is used to quantify this. Based on the preset resolution step size, the real-time sleepiness tracking path can be effectively determined. Immersive sleep state data provides tracking guidance. Using this tracking guidance, the balance prediction value of the collected real-time data is calculated on the predetermined tracking path. This balance prediction value is a "candidate orientation" close to the actual sleep balance threshold. By substituting the balance prediction value back into the biological sleep dynamic state equation, the support point that drives the user to reach sleep homeostasis at different short moments can be predicted, namely the dynamic change balance fulcrum. This is the key point for locking in whether the user experiences a sleepiness jump. Subsequently, at the new equilibrium point, the Jacobian matrix is ​​calculated to track and update the tangent direction, determining the extension direction for the next short time step. This ensures the continuity of sleepiness perception, prevents sleepiness from deviating or diverging, and achieves a real-time, continuous sleepiness characterization effect. Finally, the candidate tangent trajectory feature values ​​recorded by the Jacobian matrix during the update process reveal the real-time sleepiness dynamic threshold at which the target user reaches sleep homeostasis. This method can track real-time sleepiness during the sleep process of different users, thus providing a personalized reference template for rhythm planning. This makes rhythm positioning more reliable for different users and improves the accuracy of subsequent immersive sleep scene guidance.

[0080] Preferably, S104, as Figure 2 As shown, the specific steps include:

[0081] By retrieving the preset creation and control strategies of the current immersive sleep scene through the Internet of Things, a dynamic simulation model of the current immersive sleep scene is constructed using sleep scene simulation software.

[0082] A preset control strategy is used to inject jitter into a dynamic simulation model. The real-time sleep rhythm phase is simulated by the injected dynamic simulation model. During the simulation, the target user's EEG signal is recorded and the simulated EEG signal sequence is output.

[0083] The actual EEG signal sequence of the target user in the current immersive sleep scenario is obtained, a time-frequency domain registration space is constructed, and the time-frequency domain registration space is divided into N sub-registration blocks based on the predetermined stage of sleep stage.

[0084] A short-time Fourier transform algorithm is introduced to perform rhythmic response transformation calculation on the actual EEG signal sequence to obtain the actual response spectrum waveform. In the time-frequency domain registration space, the simulated EEG signal sequence and the actual EEG signal sequence are registered.

[0085] Only simulated response spectrum waveforms and corresponding sub-registration blocks that have a similarity greater than a preset similarity threshold to the actual response spectrum waveform in the simulated EEG signal sequence are extracted and marked as response registration blocks. The registration overlap rate between the simulated response spectrum waveform and the actual response spectrum waveform is obtained through the response registration blocks.

[0086] The response spectrum distortion coefficients of the target user's response to the real-time sleep rhythm phase under the current immersive sleep scene jitter are determined based on the registration and superposition rate, and the Hilbert instantaneous phase transformation matrix is ​​constructed based on the response spectrum distortion coefficients.

[0087] A phase oscillation equation is introduced, and the response shift of the real-time sleep rhythm phase is smoothed by observation and analysis of scattered points in the phase oscillation equation through the Hilbert instantaneous phase transformation matrix, and then inverted to generate a simulated sleep rhythm phase.

[0088] If the phase response drift function between the simulated sleep rhythm phase and the real-time sleep rhythm phase is positive, then the current immersive sleep scenario in which the real-time sleep rhythm phase is located is marked as a rhythmic chaotic scenario; if the phase response drift function between the simulated sleep rhythm phase and the real-time sleep rhythm phase is negative, then the current immersive sleep scenario is marked as a rhythmic inertial scenario, and the rhythm planning analysis results are obtained.

[0089] It should be noted that evaluating the response performance of the target user to the planned real-time sleep rhythm phase under the jittery environment of the current immersive sleep scenario by constructing a dynamic simulation model can replace the tedious steps of manual on-site testing, saving a significant amount of manpower and resources in scenario adjustment costs and greatly improving adjustment efficiency. Specifically, the time-frequency domain registration space is divided into N sub-registration blocks based on predetermined sleep stages. This allows for more refined analysis of the rhythmic response spectrum between simulated and actual EEG signals following the sleep stage transition, avoiding omissions and misalignments that occur with traditional coarse spectrum matching, and significantly improving the accuracy of response node localization. The actual response spectrum waveform reveals the known trigger nodes of the target user's response rhythm phase during actual sleep in the current immersive sleep scenario. The response spectrum distortion coefficient quantifies the deviation of the simulated response characteristics triggering the real-time sleep rhythm phase under jittery conditions relative to known response characteristics. The sensitivity of the current immersive sleep scenario to the real-time sleep rhythm phase is visualized by constructing a Hilbert instantaneous phase transformation matrix, and finally, the simulated sleep rhythm phase is inverted through the phase oscillation equation. The phase response drift function reflects the magnitude of the phase drift of the simulated sleep rhythm phase (in the current immersive sleep scenario) compared to the real-time sleep rhythm phase (standard).

[0090] It should be noted that immersive sleep scenarios are generated based on real-time sleep rhythm phases, including a lead scenario (T0-T1, 30-60 min), a sleep onset scenario (T1-T2, 0-30 min), a maintenance scenario (T2-T3, during sleep), and a pre-wake scenario (T3-T4, 15-30 min before wake-up). If the phase response drift function between the simulated sleep rhythm phase and the real-time sleep rhythm phase is positive, it indicates that the simulated sleep rhythm phase is responding earlier than the real-time sleep rhythm phase. This suggests that the atmosphere created by the current immersive sleep scenario is misleading or interfering with the target user's staged sleep, causing confusion in the direction of the real-time sleep rhythm phase. For example, the target user may be in the critical sleep onset rhythm from N1 light sleep to N2 light sleep, but the high-frequency white noise and gradually brightening of the lights generated by the current immersive sleep scenario prevent the target user from entering the planned rhythm. Therefore, this scenario is a rhythmic chaos scenario. Conversely, if the phase response drift function is negative, it indicates that the current immersive sleep scenario cannot respond to the real-time sleep rhythm phase. In other words, the real-time sleep rhythm phase is not adapted to the current immersive sleep scenario, resulting in phase inertia. Therefore, this scenario is an unsuitable one that causes rhythm phase inertia. This method can calculate the response characteristics and phase shift of the real-time sleep rhythm phase by simulating the jitter of the current immersive sleep scenario, thereby analyzing the correctness and rationality of the current scenario's guidance of the planned rhythm phase and improving the accuracy of the immersive sleep scenario's adjustment for different user rhythms.

[0091] Preferably, step S106 specifically includes the following steps:

[0092] If the rhythm planning analysis results show a rhythmic chaos scenario, then the immersive sleep scenario creation system is used to obtain the phased guidance decision information of different scenario factors in the rhythmic chaos scenario for the target user to develop an immersive sleep mode.

[0093] The architecture of using sleep staging rules as the baseline modal scale is embedded to construct a stage alignment grid for heterogeneous modalities. The real-time sleep rhythm phase is set as the guiding window event, and the trigger timestamp of the sleep staging rule when the guiding window event occurs is obtained.

[0094] Based on the phased guidance decision information, the rhythmic chaos scene is aligned to the phase alignment grid around the trigger timestamp as the alignment benchmark. At the same time, during the alignment process, the modal alignment range between sleep phase and rhythmic chaos scene is traversed to obtain the phase timing difference of rhythmic chaos scene.

[0095] Obtain the open-loop transfer function of the immersive sleep scene creation system to control the creation of a circadian chaotic scene, draw the Bode open-loop frequency diagram of the circadian chaotic scene based on the open-loop transfer function, and identify the current low-frequency gain of the system in creating the circadian chaotic scene by analyzing the Bode open-loop frequency diagram.

[0096] Based on the phase timing differential preset maximum low frequency gain requirement, if the current low frequency gain is insufficient for the maximum low frequency gain requirement, then an amplitude adjustment compensator is deployed and constructed in the immersive sleep scene creation system, and the boundary cutoff frequency of the amplitude adjustment compensator is constrained based on the maximum low frequency gain requirement.

[0097] Based on the current low-frequency gain, an amplitude adjustment compensator is used to construct the adjustment correction value for the rhythmic chaos scene. The current cutoff frequency is obtained, and the adjustment correction value is applied to the immersive sleep scene creation system to adjust the creation parameters of each scene factor in the rhythmic chaos scene until the current cutoff frequency approaches the boundary cutoff frequency, thus obtaining the first adjustment scheme.

[0098] It should be noted that the staged guidance decision information formulates the switching decision for guiding the transition to immersive sleep mode in a circadian rhythm chaos scenario. For example, in the sleep scenario (T1~T2, 0~30min), low-frequency breathing sounds and low light waves of less than or equal to 10lx are introduced to guide the target user from the N2 light sleep stage to the N3 deep sleep stage. Since a circadian rhythm chaos scenario leads to disordered real-time sleep rhythm phase response, and the real-time sleep rhythm phase is based on sleep stage rules, there is a heterogeneous misalignment between the appropriate triggering time of the real-time sleep rhythm phase in the sleep stage modality and the actual triggering time of the stage modality in the circadian rhythm chaos scenario. This leads to the erroneous guidance of the real-time sleep rhythm phase by the circadian rhythm chaos scenario. Therefore, this method first constructs a stage alignment grid of heterogeneous modalities based on sleep stage rules. This stage alignment grid is used for interface-based alignment of the triggering time scale of the real-time sleep rhythm phase in the sleep stage and the stage of the circadian rhythm chaos scenario. Furthermore, by setting the real-time sleep rhythm phase as a guiding window event detection, its trigger timestamp within the sleep stage rule is determined. This trigger timestamp serves as an alignment reference point or anchor point on the time series, guiding the precise alignment of the stage modality in the circadian rhythm chaos scenario with the sleep stage modality. The degree of heterogeneous misalignment between the two, from the original trigger timing point to the aligned timing range, represents the phase timing difference. This phase timing difference visualizes the guidance error of the circadian rhythm chaos scenario, guiding the adjustment range and scope of the scenario creation parameters. It is a key adjustment prerequisite for correcting the real-time sleep rhythm phase guidance of the circadian rhythm chaos scenario.

[0099] It should be noted that a Bode open-loop frequency diagram is plotted on the open-loop transfer function of the immersive sleep scene creation system controlling the creation of a circadian rhythmic chaos scene. This diagram is used to analyze the phase margin and amplitude characteristics of the current system, thereby determining whether the low-frequency gain for compensating the phase timing difference is sufficient and intuitively indicating the necessary direction for chaos compensation. The maximum low-frequency gain requirement is the open-loop compensation amount needed to correct the phase timing difference caused by the circadian rhythmic chaos scene. If the current low-frequency gain is insufficient to meet the maximum low-frequency gain requirement, it indicates that the immersive sleep scene creation system cannot meet the adaptive compensation capability of the open-loop compensation amount. Therefore, this method uses an amplitude adjustment compensator to perform the compensation calculation effect on the open-loop compensation amount. The boundary cutoff frequency is the critical constraint that meets the maximum low-frequency gain requirement. By constructing and adjusting the correction value so that the current cutoff frequency is close to the boundary cutoff frequency, the current low-frequency gain can be effectively placed in the low-frequency range. This ensures that the compensator can improve the adjustment gain of different scene factors in the low-frequency band without significantly affecting the nearby rhythm phase margin. This improves the reasonable adjustment and accuracy of the corresponding creation parameters of each scene factor by heterogeneous misalignment compensation based on the triggering timing of rhythmic chaos scenes. This method can perform heterogeneous misalignment alignment analysis and correction compensation for unreasonable sleep guidance of real-time sleep rhythm phase caused by staged rhythmic chaos scenes, thereby achieving the adjustment of rhythmic chaos scenes, eliminating the phenomenon of rhythm phase trigger chaos, improving the immersive sleep quality of target users, and avoiding negative optimization of sleep guidance.

[0100] Preferably, step S108 specifically includes the following steps:

[0101] If the rhythm planning analysis results show a rhythmic inertia scenario, then the immersive sleep scenario creation system is used to obtain several preparatory creation parameters for different scenario factors in the rhythmic inertia scenario, and the corresponding intervention function of each preparatory creation parameter to promote immersive sleep.

[0102] Based on different scenario factors, a parameter-tuning polar coordinate branch is established. According to the intervention function, a fixed-point coordinate network of the reserve construction parameters located in the parameter-tuning polar coordinate branch is woven. The K-means clustering algorithm is introduced to cluster each reserve construction parameter to a network node of the fixed-point coordinate network. An inert counter is assigned to each fixed-point coordinate to generate a multi-dimensional parameter-tuning polar coordinate model of the rhythmic inert scenario.

[0103] Obtain the real-time reserve construction parameters of the rhythmic inertial scenario under the current working condition, find the fixed point coordinates of each real-time reserve construction parameter in the multi-dimensional adjustment coordinate model and connect them to obtain the real-time parameter adjustment polar coordinate map.

[0104] By analyzing the inertial accumulation exponent of the inertial counter, the delay weight of the real-time parameter-tuned polar coordinate layout for the phase response drift function is analyzed. Based on the delay weight, sampling planning is carried out starting from the real-time parameter-tuned polar coordinate layout as the starting center to obtain the auxiliary sampling layout.

[0105] Extract the polar coordinate branch of the parameter adjustment covered by the auxiliary sampling map, the corresponding fixed point coordinates and the corresponding reserve construction parameters, and define them as candidate adjustment parameter solutions. At the same time, preset the jitter residual gradient threshold so that the phase response drift function approaches 0.

[0106] If the jitter residual gradient of the fixed-point coordinate does not exceed the jitter residual gradient threshold, the update of the candidate adjustment parameter solution for that fixed-point coordinate is skipped; if the jitter residual gradient of the fixed-point coordinate exceeds the jitter residual gradient threshold, the update of the corresponding candidate adjustment parameter solution is applied to obtain the adjustment strategy for each scene factor.

[0107] Based on the adjustment strategy, the creation parameters of each scene factor in the immersive sleep scene creation system are adjusted to obtain the second adjustment scheme.

[0108] It should be noted that the preparatory creation parameters are reasonable preset adjustment parameters planned in advance by the immersive sleep scene creation system for different scene factors in rhythmic inertia scenarios. Since the accurate response of real-time sleep rhythm phase cannot adapt to rhythmic inertia scenarios, this method establishes a polar coordinate model, where the included angle dimension of the polar coordinate model is the parameter tuning polar coordinate branch, representing different scene factors, forming a multi-dimensional parameter tuning polar coordinate model. Compared with the traditional polar coordinate model, the multi-dimensional parameter tuning polar coordinate model of this method embeds a fixed-point coordinate network clustered by the corresponding preparatory creation parameters in the included angle domain of each scene factor, representing and storing the candidate parameters that can be tuned to adjust the rhythmic inertia scenario within the polar coordinate dimension of the scene factor, realizing synchronous parameter tuning of different scene factors. Compared with traditional adjustment methods, it reduces the step of parameter tuning calculation and debugging, and significantly improves the adjustment efficiency and accuracy of immersive sleep scenarios. Furthermore, an inertia counter is deployed at each fixed coordinate to weightedly measure the delay strength of the preparatory construction parameters in the current rhythmic inertia scenario. This provides a sampling benchmark for candidate parameter tuning to mitigate or eliminate the inertia caused by the scenario, resulting in an auxiliary sampling pattern. The real-time parameter tuning polar coordinate pattern is the polar coordinate radius pattern for the current rhythmic inertia scenario created by parameter tuning, where the polar coordinate radius represents the parameter tuning vector (magnitude and direction). The auxiliary sampling pattern is the polar coordinate radius pattern required to eliminate scenario inertia through parameter tuning scenario factors. The preparatory construction parameters covered in this polar coordinate radius pattern constitute the candidate adjustment parameter set that satisfies inertia mitigation. The jitter residual gradient reflects the gradient contribution of the preparatory construction parameters at the fixed coordinates to maximize the elimination of scenario inertia, revealing the optimality and suitability of each candidate adjustment parameter solution for rhythmic inertia scenario adjustment.

[0109] It should be noted that if the jitter residual gradient of the fixed-point coordinate does not exceed the jitter residual gradient threshold, it indicates that the amplitude of the jitter phase response drift function of the candidate adjustment parameter solution is small and its contribution is limited. Therefore, it is not considered as a suitable parameter for adjusting the rhythm inertia scenario, and the update of the candidate adjustment parameter solution for this fixed-point coordinate is skipped. Conversely, it indicates that the gradient contribution of the candidate adjustment parameter solution to eliminate scenario inertia is large, which can ensure that the rhythm inertia scenario accurately triggers the real-time sleep rhythm phase response. Therefore, the update of the candidate adjustment parameter solution on this fixed-point coordinate is applied, constituting the adjustment strategy for each scenario factor of the rhythm inertia scenario. This method can eliminate the phase trigger inertia phenomenon caused by the rhythm inertia scenario, ensure that the real-time sleep rhythm phase can have high adaptability in the current immersive sleep scenario stage, improve the guidance stability and reliability of immersive sleep, and improve the comfort and satisfaction of the target user in the immersive sleep scenario feedback.

[0110] The second aspect of this method provides an immersive sleep scene regulation system based on rhythmic planning analysis, such as... Figure 3 As shown, the system includes: a memory 301, a processor 302, and a communication interface 303. The memory 301 includes a program for adjusting an immersive sleep scene based on rhythm planning analysis. The communication interface 303 is used for data connection and communication between the memory 301 and the processor 302. When the program for adjusting an immersive sleep scene based on rhythm planning analysis is executed by the processor 302, it implements any of the steps of the immersive sleep scene adjustment method described above.

[0111] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for regulating immersive sleep scenarios based on rhythmic programming analysis, characterized in that, Includes the following steps: S102: Obtain real-time physiological feedback data and immersive sleep state data of the target user to track real-time sleepiness, obtain a real-time sleepiness index curve, and map and plan the sleepiness rhythm of saturated and unsaturated states within the sleep sliding sub-window based on the real-time sleepiness index curve to obtain the real-time sleep rhythm phase of the target user. S104: Simulate the preset creation control strategy of the current immersive sleep scene using simulation software to shake the real-time sleep rhythm phase. Calculate the response shift of the current immersive sleep scene to the rhythm phase based on the simulation results and the actual EEG signal sequence. Analyze the current immersive sleep scene based on the phase response drift function between the generated current sleep rhythm phase and the real-time sleep rhythm phase to obtain the rhythm planning analysis results. S106: If the rhythm planning analysis results show a rhythm disorder scenario, then by using the real-time sleep rhythm phase based on the sleep stage modality as the driving benchmark of the guiding event to align to the rhythm disorder scenario modality, the phase timing difference is obtained, and the amplitude compensation design based on the phase timing difference is used to adjust the creation parameters of different scenario factors of the rhythm disorder scenario to generate the first adjustment scheme. S108: If the rhythm planning analysis results show a rhythm inertia scenario, then construct a multi-dimensional parameter-tuning polar coordinate model. Based on the inertia delay of the phase response drift in the current rhythm inertia scenario, analyze the adjustment strategy of the scenario factor on the multi-dimensional parameter-tuning polar coordinate model to adjust the creation parameters of the rhythm inertia scenario and obtain the second adjustment scheme. Specifically, S104 includes the following steps: By retrieving the preset creation and control strategies of the current immersive sleep scene through the Internet of Things, a dynamic simulation model of the current immersive sleep scene is constructed using sleep scene simulation software. A preset control strategy is used to inject jitter into a dynamic simulation model. The real-time sleep rhythm phase is simulated by the injected dynamic simulation model. During the simulation, the target user's EEG signal is recorded and the simulated EEG signal sequence is output. The actual EEG signal sequence of the target user in the current immersive sleep scenario is obtained, a time-frequency domain registration space is constructed, and the time-frequency domain registration space is divided into N sub-registration blocks based on the predetermined stage of sleep stage. A short-time Fourier transform algorithm is introduced to perform rhythmic response transformation calculation on the actual EEG signal sequence to obtain the actual response spectrum waveform. In the time-frequency domain registration space, the simulated EEG signal sequence and the actual EEG signal sequence are registered. Only simulated response spectrum waveforms and corresponding sub-registration blocks that have a similarity greater than a preset similarity threshold to the actual response spectrum waveform in the simulated EEG signal sequence are extracted and marked as response registration blocks. The registration overlap rate between the simulated response spectrum waveform and the actual response spectrum waveform is obtained through the response registration blocks. The response spectrum distortion coefficients of the target user's response to the real-time sleep rhythm phase under the current immersive sleep scene jitter are determined based on the registration and superposition rate, and the Hilbert instantaneous phase transformation matrix is ​​constructed based on the response spectrum distortion coefficients. A phase oscillation equation is introduced, and the response shift of the real-time sleep rhythm phase is smoothed by observation and analysis of scattered points in the phase oscillation equation through the Hilbert instantaneous phase transformation matrix, and then inverted to generate a simulated sleep rhythm phase. If the phase response drift function between the simulated sleep rhythm phase and the real-time sleep rhythm phase is positive, then the current immersive sleep scenario in which the real-time sleep rhythm phase is located is marked as a rhythmic chaotic scenario; if the phase response drift function between the simulated sleep rhythm phase and the real-time sleep rhythm phase is negative, then the current immersive sleep scenario is marked as a rhythmic inertial scenario, and the rhythm planning analysis results are obtained.

2. The method for regulating immersive sleep scenarios based on rhythmic programming analysis according to claim 1, characterized in that, S102 specifically includes the following steps: Real-time physiological feedback data of target users is collected through the Internet of Things, and sleep logs are used to extract data on the immersive sleep state of target users when real-time physiological feedback data is generated in historical time periods. A biological sleep dynamics state equation is established by introducing the biological sleep dynamics mechanism. Based on the sleep plasticity of real-time physiological feedback data of sleep homeostasis assessment, the real-time sleepiness of the target user is tracked and updated in the biological sleep dynamics state equation along the tangent direction of the immersive sleep state data, so as to obtain the real-time sleepiness index curve of sleep time series. We construct a current circadian rhythm homeostasis sleep sliding window domain, formulate a variance weighting criterion for the sleep sliding window domain based on the short-term sleep driving mechanism, and divide the sleep sliding window domain into several sleep sliding sub-windows based on the sleep behavior information of the target user. Based on the variance weighting criterion, the kurtosis weighted calculation of the slope steepness of the real-time sleepiness index curve within each sleep sliding window framework is performed to obtain the local sleepiness kurtosis decay value of each sleep sliding window. If the local sleepiness kurtosis decay value is greater than the preset threshold, the peak-valley curve span corresponding to the sleep sliding sub-window is marked as the saturation cutoff segment; if it is less than the preset threshold, it is marked as the unsaturation cutoff segment. The real-time sleepiness index of the saturation cutoff segment and the unsaturation cutoff segment is obtained by real-time sleepiness index curve. The center point of the saturation cutoff segment is located and labeled as the first proxy anchor point. The center point of the saturation cutoff segment is located and labeled as the second proxy anchor point. At each first and second proxy anchor point, the average height of the rhythm function corresponding to each saturated and unsaturated cutoff segment of the real-time sleepiness index is calculated. The explicit Euler method is introduced to perform a time-series approximate integration of the average height of the rhythm function to generate the rhythm area under the real-time sleepiness index curve. Obtain sleep staging rules, and perform phase mapping planning of sleep staging rules based on the surrogate interleaving distribution of sleepiness saturation integral - unsaturation integral in the area model under rhythm, to obtain the real-time sleep rhythm phase of the target user.

3. The method for regulating immersive sleep scenarios based on rhythmic programming analysis according to claim 2, characterized in that, The process involves introducing biological sleep dynamics mechanisms to establish a biological sleep dynamic state equation. Based on the sleep plasticity of real-time physiological feedback data from sleep homeostasis assessment, the real-time sleepiness of the target user is tracked and updated in the biological sleep dynamic state equation along the tangent direction of the immersive sleep state data, resulting in a real-time sleepiness index curve for the sleep time series. The specific steps include: Based on big data retrieval, a rhythmic sleep index system is obtained. The biological sleep dynamics mechanism is introduced. The biological sleep dynamics mechanism is used to model the dynamic state control of real-time physiological feedback data with immersive sleep state data as the state variable input, and a nonlinear biological sleep dynamic state equation is generated. Sleep homeostasis indicators are extracted through a circadian rhythm sleep index system. A series of sleep homeostasis indicators are used to evaluate real-time physiological feedback data in multiple dimensions to obtain the sleep plasticity coefficient of the target user. The tracking resolution step size is preset based on the sleep plasticity coefficient. Based on the tangent direction of the immersive sleep state data as a linear approximation, the balance prediction value on the acquisition time sequence is calculated along the tangent direction based on the tracking resolution step size. The balance prediction value is substituted into the biological sleep dynamic state equation to predict the support position of the next dynamic equilibrium point, thus obtaining the dynamic iteration equilibrium fulcrum. A Jacobian matrix is ​​calculated based on the dynamic change equilibrium pivot to track and update the tangent direction. The candidate tangent trajectory feature values ​​are output during the continuous update process of the Jacobian matrix. The real-time sleepiness index of the target user is determined based on the candidate tangent trajectory feature values, and a real-time sleepiness index curve on the sleep time series is constructed by fitting.

4. The method for regulating immersive sleep scenarios based on rhythmic programming analysis according to claim 1, characterized in that, S106 specifically includes the following steps: If the rhythm planning analysis results show a rhythmic chaos scenario, then the immersive sleep scenario creation system will be used to obtain the phased guidance decision information of different scenario factors in the rhythmic chaos scenario for the target user to enter the immersive sleep mode. The architecture of using sleep staging rules as the baseline modal scale is embedded to construct a stage alignment grid for heterogeneous modalities. The real-time sleep rhythm phase is set as the guiding window event, and the trigger timestamp of the sleep staging rule when the guiding window event occurs is obtained. Based on the phased guidance decision information, the rhythmic chaos scene is aligned to the phase alignment grid around the trigger timestamp as the alignment benchmark. At the same time, during the alignment process, the modal alignment range between sleep phase and rhythmic chaos scene is traversed to obtain the phase timing difference of rhythmic chaos scene. Obtain the open-loop transfer function of the immersive sleep scene creation system to control the creation of a circadian chaotic scene, draw the Bode open-loop frequency diagram of the circadian chaotic scene based on the open-loop transfer function, and identify the current low-frequency gain of the system in creating the circadian chaotic scene by analyzing the Bode open-loop frequency diagram. Based on the phase timing differential preset maximum low frequency gain requirement, if the current low frequency gain is insufficient for the maximum low frequency gain requirement, then an amplitude adjustment compensator is deployed and constructed in the immersive sleep scene creation system, and the boundary cutoff frequency of the amplitude adjustment compensator is constrained based on the maximum low frequency gain requirement. Based on the current low-frequency gain, an amplitude adjustment compensator is used to construct the adjustment correction value for the rhythmic chaos scene. The current cutoff frequency is obtained, and the adjustment correction value is applied to the immersive sleep scene creation system to adjust the creation parameters of each scene factor in the rhythmic chaos scene until the current cutoff frequency approaches the boundary cutoff frequency, thus obtaining the first adjustment scheme.

5. The method for regulating immersive sleep scenarios based on rhythmic programming analysis according to claim 1, characterized in that, S108 specifically includes the following steps: If the rhythm planning analysis results show a rhythmic inertia scenario, then the immersive sleep scenario creation system is used to obtain several preparatory creation parameters for different scenario factors in the rhythmic inertia scenario, and the corresponding intervention function of each preparatory creation parameter to promote immersive sleep. Based on different scenario factors, a parameter-tuning polar coordinate branch is established. According to the intervention function, a fixed-point coordinate network of the reserve construction parameters located in the parameter-tuning polar coordinate branch is woven. The K-means clustering algorithm is introduced to cluster each reserve construction parameter to a network node of the fixed-point coordinate network. An inert counter is assigned to each fixed-point coordinate to generate a multi-dimensional parameter-tuning polar coordinate model of the rhythmic inert scenario. Obtain the real-time reserve construction parameters of the rhythmic inertial scenario under the current working condition, find the fixed point coordinates of each real-time reserve construction parameter in the multi-dimensional adjustment coordinate model and connect them to obtain the real-time parameter adjustment polar coordinate map. By analyzing the inertial accumulation exponent of the inertial counter, the delay weight of the real-time parameter-tuned polar coordinate layout for the phase response drift function is analyzed. Based on the delay weight, sampling planning is carried out starting from the real-time parameter-tuned polar coordinate layout as the starting center to obtain the auxiliary sampling layout. Extract the polar coordinate branch of the parameter adjustment covered by the auxiliary sampling map, the corresponding fixed point coordinates and the corresponding reserve construction parameters, and define them as candidate adjustment parameter solutions. At the same time, preset the jitter residual gradient threshold so that the phase response drift function approaches 0. If the jitter residual gradient of the fixed-point coordinate does not exceed the jitter residual gradient threshold, the update of the candidate adjustment parameter solution for that fixed-point coordinate is skipped; if the jitter residual gradient of the fixed-point coordinate exceeds the jitter residual gradient threshold, the update of the corresponding candidate adjustment parameter solution is applied to obtain the adjustment strategy for each scene factor. Based on the adjustment strategy, the creation parameters of each scene factor in the immersive sleep scene creation system are adjusted to obtain the second adjustment scheme.

6. An immersive sleep scene regulation system based on rhythmic programming analysis, characterized in that, The system includes: a memory, a processor, and a communication interface. The memory includes a program for adjusting an immersive sleep scene based on rhythm planning analysis. The communication interface is used for data connection and communication between the memory and the processor. When the program for adjusting an immersive sleep scene based on rhythm planning analysis is executed by the processor, it implements the steps of the immersive sleep scene adjustment method as described in any one of claims 1-5.

Images