Intelligent bracelet earthquake early warning service system and method

Abstract

The application discloses an intelligent bracelet earthquake early warning service system and method, which collects user physiological parameters and earthquake early warning signals, establishes a coupling relationship between the magnitude intensity parameter and the user physiological characteristics, forms an individual disaster bearing benchmark, combines the user geographic location and the building seismic grade for gravity sensing detection, analyzes the spatial relationship between the building micro-inclination angle and the user position, generates the gravity field risk characteristics, analyzes the group movement mode through the relative position change among multiple bracelets, extracts the movement coordination characteristics and the direction coordination coefficient, forms the displacement coordination early warning sequence, generates the disaster emergency coefficient based on the user movement coordination efficiency and the protection response behavior, constructs the sound wave positioning network of the multi-device linkage triggering area, realizes the collaborative alarm and the personalized evacuation guidance, can provide differentiated earthquake early warning services according to the individual physiological characteristics and the environmental conditions, and improves the pertinence and effectiveness of the early warning.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of disaster warning and personal safety protection, in particular to an intelligent bracelet earthquake early warning service system and method. BACKGROUND

[0002] Earthquake disaster has the characteristics of strong suddenness and great destructive power. The traditional earthquake early warning technology mainly adopts a unified early warning mode, which lacks consideration of individual differences and is difficult to provide personalized early warning services according to the physiological characteristics, physical conditions and environmental conditions of different users. The existing early warning method often ignores the difference in individual bearing capacity and cannot accurately reflect the real risk level and emergency needs of different users in earthquake disasters.

[0003] The rapid development of intelligent wearable devices provides a new technical path for individualized disaster warning, but there are still many technical bottlenecks in the current earthquake early warning research based on wearable devices. The existing method lacks effective coupling technology between user physiological parameters and earthquake threat intensity, and also has deficiencies in multi-device collaborative warning, group movement coordination and emergency capacity assessment, making it difficult to realize truly intelligent individual warning. Therefore, a method is needed to solve at least one of the above problems. SUMMARY

[0004] The present application discloses an intelligent bracelet earthquake early warning service system and method, aiming to establish an individual disaster bearing benchmark by obtaining user physiological parameters and earthquake early warning signals, form a personalized risk assessment combined with building seismic grade and gravity field risk analysis, realize group movement coordination and displacement warning through multi-bracelet collaborative analysis, and construct a multi-device linkage warning network based on movement coordination efficiency and disaster emergency coefficient, ultimately forming an intelligent earthquake early warning service that adapts to individual differences.

[0005] The present application discloses an intelligent bracelet earthquake early warning service system and method, aiming to establish an individual disaster bearing benchmark by obtaining user physiological parameters and earthquake early warning signals, form a personalized risk assessment combined with building seismic grade and gravity field risk analysis, realize group movement coordination and displacement warning through multi-bracelet collaborative analysis, and construct a multi-device linkage warning network based on movement coordination efficiency and disaster emergency coefficient, ultimately forming an intelligent earthquake early warning service that adapts to individual differences.

[0006] Obtain the earthquake early warning signal sent by the earthquake monitoring network and the user physiological parameter, extract the magnitude intensity parameter through the earthquake early warning signal, and establish an individual disaster bearing benchmark based on the magnitude intensity parameter and the user physiological parameter;

[0007] Identify the earthquake high-risk period based on the individual disaster bearing benchmark, detect the phase difference between bracelet vibration and ground vibration according to the earthquake high-risk period, generate earthquake disaster assessment data using the phase difference, and determine an individual disaster risk table according to the earthquake disaster assessment data;

[0008] Obtain the real-time geographical position of the user and the building seismic resistance level, generate the building micro-inclination angle through gravity sensing detection according to the building seismic resistance level, and couple the real-time geographical position of the user and the building micro-inclination angle to form a gravity field risk feature;

[0009] Identify the disaster early warning urgency based on the individual disaster risk table and the gravity field risk feature, determine the relative position change among multiple wristbands based on the disaster early warning urgency, and generate a displacement cooperative early warning sequence by using the relative position change among multiple wristbands;

[0010] Generate active escape response data based on the protection classification based on the displacement cooperative early warning sequence, analyze the inertia swing amplitude of the active escape response data to extract a motion coordination efficiency coefficient, and generate a disaster emergency coefficient based on the motion coordination efficiency coefficient;

[0011] Identify a multi-device linkage triggering area based on the disaster emergency coefficient, construct an acoustic positioning network for the multi-device linkage triggering area, generate acoustic network cooperative alarm data based on the acoustic positioning network, and complete the intelligent wristband earthquake early warning service.

[0012] The second aspect of the present application proposes an intelligent wristband earthquake early warning service system, comprising:

[0013] A signal acquisition module is configured to obtain an earthquake early warning signal sent by an earthquake monitoring network and a user physiological parameter, extract a magnitude intensity parameter through the earthquake early warning signal, and establish an individual disaster bearing benchmark based on the magnitude intensity parameter and the user physiological parameter;

[0014] A phase detection module is configured to identify a high-risk period of earthquake based on the individual disaster bearing benchmark, detect a phase difference between wristband vibration and ground vibration according to the high-risk period of earthquake, generate earthquake damage evaluation data by using the phase difference, and determine an individual disaster risk table according to the earthquake damage evaluation data;

[0015] A gravity sensing module is configured to obtain the real-time geographical position of the user and the building seismic resistance level, generate the building micro-inclination angle through gravity sensing detection according to the building seismic resistance level, and couple the real-time geographical position of the user and the building micro-inclination angle to form a gravity field risk feature;

[0016] A cooperative positioning module is configured to identify the disaster early warning urgency based on the individual disaster risk table and the gravity field risk feature, determine the relative position change among multiple wristbands based on the disaster early warning urgency, and generate a displacement cooperative early warning sequence by using the relative position change among multiple wristbands;

[0017] The motion analysis module is used for generating active escape response data based on the displacement coordination warning sequence for protection classification, performing inertia swing amplitude analysis on the active escape response data to extract a motion coordination efficiency coefficient, and generating a disaster emergency coefficient based on the motion coordination efficiency coefficient;

[0018] The warning output module is used for identifying a multi-device linkage triggering area based on the disaster emergency coefficient, constructing an acoustic positioning network for the multi-device linkage triggering area, generating acoustic network coordination alarm data based on the acoustic positioning network, and completing the smart bracelet earthquake warning service.

[0019] The beneficial effects of the present application are embodied in the following points: first, through the individual disaster bearing benchmark establishment and gravity field risk characteristic analysis technology, the personalized adaptation of earthquake warning is realized. The physiological parameters of the user can be coupled with the earthquake magnitude intensity for coupled analysis to establish an individualized bearing capacity threshold, and meanwhile, the building seismic grade and gravity sensing detection result are combined to form a risk assessment for specific users and specific environments, so as to ensure the precise matching of the warning content and the actual needs of the individual. Secondly, the multi-bracelet coordination analysis and displacement coordination warning technology are adopted to improve the organization and safety of group evacuation. By analyzing the relative position change and movement mode characteristics among multiple bracelets, the coordination degree and abnormal conditions of group movement can be identified, and timely coordination warning and evacuation guidance can be issued to avoid confusion and congestion in group movement and improve the overall evacuation efficiency. Finally, the motion coordination efficiency analysis and acoustic positioning network technology are combined to enhance the intelligent level and coordination effect of emergency response. The individualized disaster emergency coefficient can be generated by analyzing the motion coordination ability and emergency response behavior of the user, and a multi-device linkage acoustic positioning network can be constructed based on the coefficient to realize accurate position positioning and coordination alarm, thereby providing timely and accurate personalized earthquake warning service for the user. BRIEF DESCRIPTION OF DRAWINGS

[0020] The drawings herein show specific examples of the technical solutions described in the present application and constitute part of the specification together with the specific embodiments, which are used to explain the technical solutions, principles and effects of the present application.

[0021] Unless specifically stated or defined otherwise, the same reference signs in different drawings represent the same or similar technical features. Different reference signs may also be used to represent the same or similar technical features.

[0022] Figure 1 is a flowchart of a smart bracelet earthquake warning service method of the present application.

[0023] Figure 2 is a structural block diagram of a smart bracelet earthquake warning service system of the present application. DETAILED DESCRIPTION

[0024] In the following description, for purposes of explanation and not limitation, specific details are set forth such as particular architectures, techniques, etc. in order to provide a thorough understanding of the embodiments of the present application. However, it will be apparent to those skilled in the art that the present application can be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known methods, devices, circuits, and

[0025] It is to be understood that the terminology "including", "comprising", "consisting" and "consisting essentially of" used in the specification and the appended claims, are used in the sense of open ended inclusion, that is, to include, but not limited to, the recited features, integers, steps, operations, elements, and / or components.

[0026] Reference throughout this specification to "one embodiment", "an embodiment", or "a specific embodiment", means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the present application. Thus, the appearances of the phrases "in one embodiment", "in an embodiment", "in some embodiments", "in other embodiments", "in additional embodiments", and so on, in various places throughout this specification are not necessarily all referring to the same embodiment, unless otherwise specifically so stated. The terms "including", "comprising", "having" and variations thereof, mean "including but not limited to", unless expressly specified otherwise.

[0027] The technical solutions of the embodiments of the present application are introduced as follows.

[0028] As shown in Figure 1 The method for providing earthquake early warning service by the smart bracelet according to the embodiments of the present application comprises the following steps S110-S160:

[0029] In step S110, the earthquake early warning signal sent by the earthquake monitoring network and the user physiological parameter are acquired, the magnitude intensity parameter is extracted from the earthquake early warning signal, and the individual disaster bearing benchmark is established based on the magnitude intensity parameter and the user physiological parameter.

[0030] Specifically, the earthquake early warning signal sent by the earthquake monitoring network and the user physiological parameters are acquired. The earthquake early warning information transmitted by the China Earthquake Early Warning Network is acquired in real time through a standardized interface. The early warning information includes key parameters such as epicenter location, epicenter latitude and longitude, time of earthquake occurrence, estimated magnitude, focal depth, and epicenter intensity. The information receiving adopts a multi-channel parallel receiving mode. The main channel receives through the dedicated earthquake early warning network, and the standby channel receives through the public communication network, ensuring the reliability of information reception. The receiving module is configured with an automatic reconnection mechanism and a data integrity verification function to ensure the timeliness and accuracy of the early warning information. The user physiological parameters are collected in real time by wearable devices, including heart rate, blood pressure, body temperature, respiratory rate, and activity intensity. The physiological parameter acquisition adopts non-invasive sensing technology, and is continuously monitored through smart bracelets, smart watches, or chest patch sensors. The data acquisition frequency is adjusted according to the characteristics of the physiological indicators. The heart rate acquisition frequency is 1 Hz, the blood pressure acquisition frequency is 0.1 Hz, and the body temperature acquisition frequency is 0.05 Hz. The transmission of physiological parameters adopts low-power Bluetooth or WiFi technology to transmit the collected data to personal smart terminals or cloud servers in real time. Data preprocessing includes steps such as outlier detection, noise filtering, and data format standardization.

[0031] The magnitude intensity parameter is extracted from the earthquake early warning signal. Based on the estimated magnitude, epicenter intensity, focal depth, time of earthquake occurrence, and other basic data in the early warning information, the complete magnitude intensity parameter is further analyzed. The magnitude intensity parameter includes physical quantities such as earthquake magnitude, peak ground acceleration, vibration duration, frequency component, and attenuation characteristics. The magnitude extraction adopts both the Richter magnitude and the intensity level double standard. The magnitude reflects the energy size of the earthquake, and the intensity reflects the impact of the earthquake on a specific location. The peak ground acceleration is calculated according to the epicenter intensity and the focal depth through the ground motion intensity relationship, representing the maximum acceleration value of the ground motion. The vibration duration is estimated based on the empirical formula of the magnitude size and the focal depth, and is divided into strong earthquake segment duration and total duration. The strong earthquake segment is defined as the time period when the peak acceleration exceeds a certain threshold. The frequency component analysis is predicted according to the focal depth and the magnitude characteristics to identify the dominant frequency and the spectral distribution characteristics. The attenuation characteristics describe the attenuation law of the amplitude of the seismic wave with time and the propagation process, reflecting the release mode of the earthquake energy. The time series characteristics of the magnitude intensity parameter reflect the dynamic process of the earthquake development, including the magnitude rising stage, the peak stage, and the attenuation stage.

[0032] In some embodiments, the individual disaster bearing benchmark is established based on the magnitude intensity parameter and the user physiological parameter, including: generating a magnitude threat time series graph through the magnitude intensity parameter; forming a physiological-magnitude coupling curve by combining the user physiological parameter and the magnitude threat time series graph; extracting a safe bearing interval from the physiological-magnitude coupling curve; and setting a threshold value of the safe bearing interval as the individual disaster bearing benchmark.

[0033] A magnitude threat time series is generated by the magnitude intensity parameter. The horizontal axis of the magnitude threat time series represents the time dimension, and the vertical axis represents the threat intensity value. The time evolution of the threat intensity is shown in the form of a curve or a bar chart. The calculation of the threat intensity takes into account multiple factors such as magnitude size, duration, frequency characteristics, and propagation distance. The time resolution of the time series is determined according to the time scale of the earthquake development, usually using a time interval of seconds or minutes. The standardization of the threat intensity unifies parameters of different dimensions into the standard interval of 0-1. The smoothing of the time series uses moving average or low-pass filtering techniques to eliminate short-term fluctuations and highlight long-term trends. The feature extraction of the threat time series includes peak position, rising slope, falling slope, and fluctuation amplitude. The segmentation analysis of the time series divides the entire time process into the precursor stage, main shock stage, and aftershock stage. The data format of the threat time series is stored in the form of key-value pairs of time stamp and threat intensity value.

[0034] For example, the combination of the user physiological parameter and the magnitude threat time series forms a physiological-magnitude coupling curve, which includes: determining an analysis window according to the magnitude threat time series, the sensitive interval of the magnitude includes the rising gradient of the magnitude, the duration and the decay rate; tracking the physiological change process along the analysis window to form a physiological response map; extracting the time coordinates of each response point in the physiological response map; and arranging the response intensity according to the time coordinates to form a physiological-magnitude coupling curve.

[0035] According to the magnitude threat time series, the sensitive interval of the magnitude is identified to determine the analysis window. In the generated magnitude threat time series, the sensitive interval is identified to determine the important interval of the magnitude change that has an important influence on the physiological response, which is the key window for subsequent analysis. The sensitive interval of the magnitude is identified by gradient analysis and change point detection methods of the time series. The rising gradient of the magnitude reflects the growth rate of the threat intensity, the greater the gradient, the faster the threat rises, and the stronger the impact on the physiological system. The duration refers to the length of time that the magnitude threat remains at a high level, the longer the duration, the greater the cumulative impact on the physiological system. The decay rate describes the speed of the magnitude threat from the peak, fast decay is beneficial to physiological recovery, and slow decay may cause sustained stress. The identification criteria of the sensitive interval include gradient threshold, duration threshold, and change amplitude threshold. The gradient threshold is determined by the first-order difference calculation of the time series, and the interval exceeding the threshold is identified as the rising sensitive interval. The duration threshold is set according to the time characteristics of physiological adaptation, generally set to several minutes to several tens of minutes. The change amplitude threshold is determined by the change range of the threat intensity, and the interval whose change amplitude exceeds a certain multiple of the baseline level is identified as the sensitive interval. The setting of the window boundary considers the delay effect and inertia characteristics of the physiological response, and appropriate time buffer is added before and after the sensitive interval.

[0036] The physiological response graph is formed by tracking the physiological change process along the analysis window. Within the determined analysis window time range, the user's various physiological parameters are continuously monitored and tracked for analysis, and the physiological state trajectory over time is recorded. The multi-parameter time series graph is used to represent the physiological response graph, which simultaneously displays the change curves of heart rate, blood pressure, body temperature, respiratory rate, and other physiological indicators. For example, when the seismic threat rapidly rises from level 3 to level 6, the user's heart rate may rise from 72 beats per minute in a resting state to 110 beats per minute in a stress state, and the blood pressure rises from 120 / 80 mmHg to 140 / 95 mmHg. These change processes are clearly presented in the form of continuous curves in the physiological response graph. The tracking process uses high-frequency sampling to ensure that rapid changes and transient responses of physiological parameters can be captured. The time axis of the graph is synchronized with the seismic threat time series graph, facilitating comparative analysis. The amplitude of physiological changes is represented by the relative change rate, i.e., the ratio or difference between the current value and the baseline value. The smoothing of the response graph uses moving average filtering technology to remove measurement noise and retain the main change trend. Standardization of the graph unifies different physiological parameters into the same numerical range, facilitating comparison and analysis.

[0037] The time coordinates of each response point are extracted within the physiological response graph. Feature point recognition and coordinate extraction are performed on the constructed physiological response graph to determine key response positions that are important for coupling analysis. Response points are defined as time nodes where physiological parameters undergo significant changes or reach specific states, including peak points, valley points, inflection points, and plateau points. Time coordinate extraction uses a combination of numerical differentiation and threshold detection methods to identify feature point positions by calculating the first and second derivatives of the graph curve. Peak points correspond to the time position where the physiological parameter reaches a local maximum, reflecting the strongest response time of the physiological system. Valley points correspond to the time position where the physiological parameter reaches a local minimum, possibly reflecting physiological inhibition or recovery processes. Inflection points are determined by zero point detection of the second derivative, indicating the transition time of the physiological response trend. Plateau points identify time periods where the first derivative is close to zero, reflecting the relative stability period of the physiological parameter. The accuracy of the feature points is improved through interpolation and fitting techniques, and continuous processing is performed between discrete sampling points. The importance of response points is sorted by indicators such as change amplitude, duration, and statistical significance. The time coordinate format uses high-precision timestamps, with an accuracy of milliseconds.

[0038] The response intensity arranged according to the time coordinates forms a physiological-magnitude coupling curve. Based on the extracted response point time coordinates, the response intensity values corresponding to each time coordinate are calculated: the physiological parameter value at the corresponding moment in the physiological response map is found as the physiological response intensity through the time coordinate, and the magnitude threat intensity value corresponding to the same time coordinate in the magnitude threat time sequence map is found. The quantification of the response intensity uses the change amplitude of the physiological parameter relative to the baseline value, and the response intensity R = (P(t)-P_baseline) / P_baseline, where P(t) is the physiological parameter value at time t, and P_baseline is the baseline physiological parameter value. The time coordinates of all response points are calculated in this way to obtain the data pairs of magnitude threat intensity and physiological response intensity, and arranged in order according to the magnitude threat intensity from small to large to form an ordered data sequence. The generation of the coupling curve uses the scatter plot fitting technique to draw a scatter plot of the arranged data pairs in a two-dimensional coordinate system, with the horizontal axis representing the magnitude threat intensity and the vertical axis representing the physiological response intensity. The fitting method uses the least squares method for polynomial fitting to generate a continuous coupling relationship curve. The mathematical expression of the curve describes the functional relationship between the physiological response intensity and the magnitude threat intensity, and the coupling strength is quantitatively evaluated by the correlation coefficient of the fitting curve.

[0039] Extract the safe bearing interval from the physiological-magnitude coupling curve. The safe region identification and boundary determination are performed on the constructed physiological-magnitude coupling curve to extract the range of magnitude threats that the user can safely bear. The safe bearing interval is the range of magnitude threats corresponding to the physiological response remaining within the normal range. The normal range of physiological response is determined by the baseline physiological parameter of the user and the medical safety standard. The interval extraction method includes threshold detection, gradient analysis, and inflection point identification techniques. Threshold detection determines the critical value of magnitude threat by setting the upper limit of safe physiological response. Gradient analysis identifies the region of rapid response change by calculating the first derivative of the coupling curve. Inflection point identification determines the turning position of the curve by zero point detection of the second derivative. The lower boundary of the safe bearing interval corresponds to the magnitude threat value at which the physiological response deviates from the baseline. The upper boundary corresponds to the magnitude threat value at which the physiological response approaches the safety threshold. The width of the interval reflects the range of individual bearing capacity, and the greater the width, the stronger the bearing capacity. The extracted safe bearing interval is stored in an interval marker format, including attribute information such as lower boundary value, upper boundary value, and confidence.

[0040] The threshold of the safety bearing interval is set as the individual disaster bearing reference. The upper boundary threshold of the safety bearing interval is used as the basic value of the individual disaster bearing reference, ensuring that the user remains in a physiological safety state below this threshold. Individual adjustment of the reference value adopts a conservative principle, introducing a safety margin coefficient based on the upper boundary threshold to avoid approaching the physiological safety critical point. Age correction is adjusted by a pre-set age segmentation coefficient, with a standard coefficient of 1.0 for young users, a coefficient of 0.9 for middle-aged users, and a coefficient of 0.8 for old users, reflecting the physiological bearing capacity differences of different age groups. Individual difference correction is evaluated by analyzing the stability of the user's baseline physiological parameters, and users with greater baseline fluctuations are set with more conservative reference settings. The formula for calculating the modified individual reference value is S_baseline = S_max x K_age x K_safety, where S_max is the upper boundary of the safety bearing interval, K_age is the age correction coefficient, and K_safety is the safety margin coefficient. The individual disaster bearing reference is represented by a single modified value, serving as an individualized disaster response capability evaluation standard. The data format of the reference value includes numerical value, unit, applicable conditions, and other attribute information.

[0041] In step S120, the high-risk period of the earthquake is identified based on the individual disaster bearing reference. The phase difference between the wristband vibration and the ground vibration is detected according to the high-risk period of the earthquake, and the phase difference is used to generate seismic damage assessment data. The individual disaster risk table is determined according to the seismic damage assessment data.

[0042] Specifically, the high-risk period of the earthquake is identified based on the individual disaster bearing reference. A sliding window analysis method is used to divide the continuous time series into fixed-length time windows for analysis. The length of the time window is determined according to the time characteristics of seismic activity, generally set to a time interval of 5-15 minutes. The determination standard of the high-risk period uses the threshold comparison of the individual disaster bearing reference. When the intensity of the earthquake magnitude in the window exceeds the bearing reference, it is marked as a high-risk period. For example, when the individual disaster bearing reference of the user is set to 5.2, the seismic monitoring data shows that the magnitude of a certain period reaches 5.8 and lasts for 3 minutes, and this period is identified as a high-risk period of the earthquake. The identification process considers the duration and cumulative effect of the magnitude intensity, and both short-time high-intensity and long-time medium-intensity can constitute a high-risk threat. The spatial range of the high-risk period is determined by the epicenter distance and the influence radius, and the closer the area to the epicenter, the higher the risk level. The time period is marked with start time, end time, threat level, and influence range, etc.

[0043] The phase difference between the bracelet vibration and the ground vibration is detected during the high-risk period of earthquakes. During the identified high-risk period of earthquakes, the vibration signals of the smart bracelet worn by the user and the vibration signals of the ground vibration sensor are monitored synchronously, and the phase relationship and time difference between the two are analyzed. The bracelet vibration detection records the vibration acceleration of the bracelet in three spatial directions in real time through the built-in three-axis acceleration sensor. The ground vibration detection is realized through the earthquake monitoring sensor arranged around the user, and the vertical and horizontal vibration data of the ground are collected. The phase difference detection uses cross-correlation analysis method to calculate the time delay and phase shift of the bracelet vibration signal and the ground vibration signal. The preprocessing of the vibration signal includes denoising filter, baseline correction and frequency band selection, etc. to ensure the signal quality and analysis accuracy. The calculation formula of the phase difference is Δφ = 2πfΔt, where Δφ is the phase difference, f is the vibration frequency, and Δt is the time delay. The phase difference between the bracelet and the ground vibration reflects the attenuation, scattering and reflection effects in the vibration propagation process. The frequency characteristics of the phase difference are obtained through frequency domain analysis, and different frequency components may have different phase delay characteristics. The detection accuracy is guaranteed by high-frequency sampling and synchronous triggering technology, and the sampling frequency is set to more than 10 times the main frequency of the seismic wave.

[0044] In some embodiments, the generating the seismic damage assessment data using the phase difference comprises: decomposing the phase difference into a fast wave phase and a slow wave phase; projecting a phase probe sequence from the fast wave phase to the slow wave phase; recording the positions of amplitude enhancement points on the phase probe sequence to form an enhancement point set; and marking the point with the strongest intensity in the enhancement point set to generate the seismic damage assessment data.

[0045] The phase difference is decomposed into a fast wave phase and a slow wave phase. The detected phase difference between the bracelet vibration and the ground vibration is decomposed into frequency components, and the composite phase difference signal is separated into two components with different propagation velocities, fast wave and slow wave. The phase difference decomposition extracts the fast wave phase component through a high-pass filter and extracts the slow wave phase component through a low-pass filter. The fast wave phase corresponds to the high-frequency component of the compression wave and shear wave in the seismic wave, has a fast propagation speed, and mainly reflects the structural response characteristics of the near field. The slow wave phase corresponds to the surface wave and long-period component in the seismic wave, has a slow propagation speed, and mainly reflects the propagation effect of the far field and the deep medium characteristics. The frequency demarcation point of decomposition is determined through seismic wave spectrum analysis, and is generally set in the frequency range of 1-5 Hz. The characteristics of the fast wave phase include higher frequency, faster attenuation and stronger spatial locality. The characteristics of the slow wave phase include lower frequency, longer propagation distance and richer information carrying. The mathematical basis of phase decomposition is Fourier transform and frequency domain filtering theory, and frequency separation of time domain signals is realized through spectrum operation. The amplitude normalization processing of the fast wave and slow wave phases ensures that the two components are comparable in numerical value.

[0046] The phase detection sequence is projected from the fast wave phase to the slow wave phase. The fast wave phase obtained by decomposition is used as the feature source domain, and a specially designed phase analysis sequence is projected to the slow wave phase target domain. By analyzing the characteristics of the projection process, the damage information of the medium and structure is obtained. The phase detection sequence is generated by a digital signal processing algorithm, which contains a preset phase jump pattern and amplitude modulation sequence, and the sequence length is set to 100-200 data points. The projection of the detection sequence uses feature mapping technology, which maps the encoding sequence into the analysis framework of the slow wave phase according to the mathematical transformation rule, and the mapping depth is controlled within 10%-20%. During the projection process, the phase characteristics and amplitude distribution of the detection sequence will change characteristically when it passes through different damage modes. Under normal mode, the sequence maintains the original encoding characteristics, and under damage mode, the sequence appears phase distortion and amplitude attenuation. The slow wave phase is used as the projection target domain, which receives the mapped detection sequence and compares it with the original projection sequence. The calculation time of the sequence is accurately measured by the projection start marker and completion detection. The normal calculation time is 0.05-0.2 seconds, and complex damage modes will cause processing delay to increase. The projection effect is quantitatively evaluated by feature fidelity and correlation coefficient. High fidelity indicates that the structure is complete, and low fidelity indicates that there is damage.

[0047] The positions of the amplitude enhancement points on the phase detection sequence are recorded to form an enhancement point set. In the transmitted phase detection sequence, the spatial positions and time nodes where the amplitude appears significant enhancement are identified and recorded. These enhancement points reflect the special medium characteristics or structural features on the propagation path. The identification of amplitude enhancement points is achieved through threshold detection and peak search, with the judgment threshold of amplitude enhancement set to 1.5-2 times the average amplitude. The formation mechanism of enhancement points mainly includes resonance amplification, constructive interference of waves, and medium impedance mutation, etc. Resonance amplification occurs when the detection frequency matches the natural frequency of the medium or structure, resulting in a significant resonance peak in the amplitude. Constructive interference occurs when the in-phase superposition of multiple propagation detection signals at a certain point causes the synthesized amplitude to exceed the amplitude of a single signal. Impedance mutation occurs at the interface between different media, causing changes in the amplitude of reflected and transmitted signals. The position record of the enhancement point contains information in three dimensions: spatial coordinates, time coordinates, and amplitude values. Spatial coordinates are calculated through the geometric configuration of the sensor array and the time difference of signal arrival. Time coordinates reflect the time when the enhancement phenomenon occurs, corresponding to the time sequence of the earthquake development process. The amplitude value records the specific strength of the enhancement point, providing a basis for subsequent strength sorting and selection. The data structure of the enhancement point set uses a point set format, and each point contains position, time, amplitude, and type attribute fields.

[0048] The point with the maximum intensity in the enhanced point set is marked to generate the earthquake damage assessment data. The intensity comparison and sorting are performed from the recorded enhanced point set, the enhanced point with the maximum amplitude intensity is selected as the key indicator of earthquake damage assessment, and the quantitative damage assessment data is generated. The selection of the maximum intensity point adopts a global search method, which traverses all points in the enhanced point set to find the point with the maximum amplitude value. The physical meaning of the maximum intensity point is to identify the position with the most serious damage or the most significant impact on the propagation path, and the characteristic parameters of the point can represent the overall damage level. The generation of earthquake damage assessment data adopts the multi-dimensional characteristic parameters of the maximum intensity point, including amplitude intensity, position coordinates, occurrence time and frequency characteristics, etc. The amplitude intensity is converted into damage grade through standardization processing, and the larger the amplitude, the higher the damage grade. The position coordinate information indicates the spatial position of the damage, providing spatial reference for damage positioning and repair. The occurrence time information reflects the evolution time sequence of the damage in the earthquake process, helping to understand the development process of the damage. The frequency characteristic information reveals the correlation between the damage and the specific frequency component, providing frequency domain information for damage mechanism analysis. The quantification of the evaluation data adopts a scoring system of 0-100, and the higher the score, the more serious the earthquake damage.

[0049] An individual disaster risk table is determined according to the earthquake damage assessment data. Based on the generated earthquake damage assessment data, the individual characteristics and environmental conditions of the user are combined to construct a personalized disaster risk assessment table. The individual disaster risk table is organized in matrix form, with rows corresponding to different risk factors and columns corresponding to different risk levels or time periods. The risk factors include structural safety risk, personal safety risk, property loss risk and evacuation difficulty risk, etc. The structural safety risk is determined based on the building damage information in the earthquake damage assessment data, and the higher the damage degree, the higher the risk level. The personal safety risk considers individual factors such as the user's physical condition, age characteristics and health level. The property loss risk is quantitatively analyzed according to the user's property distribution and value assessment. The evacuation difficulty risk considers the user's geographical location, traffic conditions and the degree of unobstructedness of the evacuation path. The numerical value of the risk table adopts a scoring standard of 0-10, and the higher the score, the greater the risk. The risk level is divided by color coding, with green representing low risk, yellow representing medium risk and red representing high risk.

[0050] In step S130, the real-time geographical position of the user and the building seismic grade are obtained, the building micro-tilt angle is generated by gravity sensing detection according to the building seismic grade, and the real-time geographical position of the user and the building micro-tilt angle are coupled to analyze the gravity field risk characteristics.

[0051] Specifically, the real-time geographical location of the user and the building seismic grade are obtained. The user's location is accurately positioned through the GPS module, Beidou navigation chip and base station positioning technology built-in the smart bracelet, with a positioning accuracy of meter level. The geographical location information includes longitude and latitude coordinates, altitude, positioning timestamp and positioning accuracy parameters. The building seismic grade information is obtained by querying the building engineering database, which contains detailed information such as the designed seismic fortification intensity, structure type, construction year and seismic reinforcement of the building. The seismic grade is divided into four levels according to the national building seismic design specification, namely, A, B, C and D, with A being the highest seismic standard and D being the lowest seismic standard. The matching of the location and the building is realized through geographical information query and spatial geometric analysis to determine the specific information of the building where the user is currently located. The real-time updating frequency of the positioning data is set to 1-5 times per second to ensure that the user's moving track and location changes can be tracked. The coordinate format of the geographical location adopts the WGS84 standard, which is consistent with the international geographical coordinate system.

[0052] The building micro-tilt angle is generated according to the building seismic grade for gravity sensing detection. Based on the obtained building seismic grade information, the verticality and tilt state of the building are analyzed using the corresponding gravity sensing detection scheme of the building seismic grade. The detection scheme of the A-grade seismic building adopts a high-precision three-axis gravity sensor array combined with an external tilt meter, with a sensor arrangement density of one detection point per 50 square meters. The detection scheme of the B-grade seismic building adopts a standard three-axis gravity sensor, with a layout density of one detection point per 100 square meters. The detection scheme of the C and D grade seismic buildings adopts a simplified detection scheme of the built-in gravity sensor of the mobile phone, focusing on the key load-bearing parts. Different levels of detection schemes have different sampling frequencies and filtering parameters: the A-grade building adopts a 100Hz sampling frequency and a 0.1Hz low-pass filter, the B-grade building adopts a 50Hz sampling frequency and a 0.2Hz filter, and the C and D grade buildings adopt a 10Hz sampling frequency and a 0.5Hz filter. The gravity sensor detects the components of the gravity acceleration in three spatial directions, and the micro-tilt angle of the building is calculated by analyzing the direction change of the gravity vector. The calculation formula of the micro-tilt angle is where θ is the tilt angle, ax and ay are the horizontal gravity components, and az is the vertical gravity component.

[0053] In some embodiments, the coupling analysis of the real-time geographical location of the user and the building micro-tilt angle forms a gravity field risk feature, including: performing gravity field strength analysis on the building micro-tilt angle to obtain tilt parameters; performing risk matching on the real-time geographical location of the user through the tilt parameters to form a location risk coefficient; generating a risk level distribution map through spatial interpolation of the location risk coefficient; and forming a gravity field risk feature according to the risk level distribution map.

[0054] The gravity field intensity analysis is used to obtain the tilt parameters of the building. The tilt angle data of the building is converted into the spatial distribution parameters of the gravity field intensity, and the influence of the tilt state on the gravity field is quantified. The gravity field intensity analysis is based on the law of universal gravitation and the theory of gravity field. The tilt of the building will cause the non-uniform distribution of the gravity field in space. The tilt parameters include the tilt direction angle, the tilt amplitude, the gravity gradient, and the field intensity distribution. The tilt direction angle is determined by the direction of the horizontal component of the gravity vector, and represents the spatial orientation of the building tilt. The tilt amplitude is represented by the tangent of the tilt angle, and reflects the severity of the tilt. The gravity gradient describes the rate of change of the gravity field intensity in space, and the calculation formula is where is the gradient vector of the gravity field intensity, and g is the gravity field intensity scalar, are the partial derivatives of the gravity field intensity in the x, y, and z directions, respectively, and i, j, and k are the unit vectors in the x, y, and z coordinate directions. The field intensity distribution is represented by the isosurface method, which connects the spatial points with the same gravity field intensity into continuous isosurfaces. The calculation of the tilt parameters considers the mass distribution and the geometric shape of the building. The tilt of the high-rise building has a more significant impact on the gravity field than the tilt of the low-rise building. The spatial resolution of the parameters is determined according to the analysis accuracy requirements, and a 1 meter x 1 meter grid division is generally used. The time characteristics of the tilt parameters reflect the dynamic process of the gravity field change with the building tilt.

[0055] The location risk coefficient is formed by matching the real-time geographic location of the user with the tilt parameters. The risk matching is based on the analysis method of spatial superposition, and the user's location is geometrically superimposed and numerically calculated with the spatial distribution of the tilt parameters. The calculation of the location risk coefficient considers the spatial relationship factors such as the distance between the user and the tilt center, the relative orientation, and the elevation difference. The distance factor is represented by an inverse proportional function, and the closer to the tilt center, the higher the risk coefficient. The orientation factor considers the angle relationship of the user's location relative to the tilt direction, and the location in the downstream direction of the tilt has a higher risk coefficient. The elevation factor reflects the relationship between the user's floor and the tilt influence, and the tilt influence of the high floor is more intense than that of the low floor. The mathematical expression of the location risk coefficient is R=f(d, θ, h), where R is the risk coefficient, d is the distance factor, θ is the orientation factor, and h is the elevation factor. The numerical range of the risk coefficient is standardized to the interval of 0-1, 0 represents no risk, and 1 represents extremely high risk. The matching accuracy is improved by high-resolution geographic grid and interpolation technology, ensuring that each location has a corresponding risk coefficient.

[0056] The risk level distribution map is generated by spatial interpolation of the position risk coefficient. The spatial interpolation adopts the Kriging interpolation method, which can consider the spatial correlation and distance decay effect to generate smooth and continuous interpolation results. The interpolation process first analyzes the spatial variation characteristics of the risk coefficient, calculates the variogram and spatial correlation distance. The variogram describes the variation law of the risk coefficient with the spatial distance, and the correlation distance determines the influence range of interpolation. The interpolation calculation is realized by weighted average method, and the weight coefficient is determined according to the distance and spatial correlation. The risk level distribution map is represented by color coding, with green representing low-risk areas, yellow representing medium-risk areas, and red representing high-risk areas. The spatial resolution of the distribution map is set to 5 meters x 5 meters, meeting the needs of refined risk analysis. The boundary treatment of interpolation adopts the extrapolation method to ensure the integrity and continuity of the distribution map. The classification method adopts the natural breakpoint classification method, and the classification threshold is determined according to the statistical distribution characteristics of the risk coefficient. The output format of the risk level distribution map adopts the raster data format, which supports the reading and analysis of GIS software.

[0057] The risk characteristics of the gravity field are formed according to the risk level distribution map. Based on the generated risk level distribution map, the characteristic parameters reflecting the spatial distribution law of the risk of the gravity field are extracted and summarized to form a comprehensive feature description for risk analysis and decision support. For example, in a risk level distribution map of a campus area, if there is a large area of red high-risk area around teaching building A, and the library is mainly surrounded by green low-risk area, then teaching building A is identified as the risk center, and the risk intensity within 50 meters around it is marked as level 9 (full score 10). The risk characteristics of the gravity field include risk center, risk intensity, distribution pattern and change trend. The risk center points to the area with the highest risk coefficient and the largest influence range, usually corresponding to the place with the most serious building tilt. The risk intensity is represented by the maximum risk coefficient value, and the higher the value, the greater the danger of the area. The distribution pattern describes the geometric shape characteristics of the risk area, and the circular distribution represents uniform risk diffusion, and the elliptical distribution represents obvious directionality. The change trend reflects the development and change of the risk of the gravity field over time, and the upward trend indicates that the risk is intensifying, and the downward trend indicates that the risk is mitigating. Feature extraction is realized by spatial analysis and statistical analysis methods to identify the main mode and key area of risk distribution. The expression of the risk characteristics of the gravity field adopts the combination of hierarchical identification and textual description, which is convenient for emergency management personnel to quickly understand and apply.

[0058] In step S140, the disaster warning urgency is identified based on the individual disaster risk table and the risk characteristics of the gravity field, the relative position change between multiple wristbands is determined based on the disaster warning urgency, and the displacement cooperative warning sequence is generated by using the relative position change between multiple wristbands.

[0059] Specifically, the disaster warning urgency is identified based on the individual disaster risk table and the gravity field risk characteristics. A multi-factor weighting method is used to associate and calculate each risk indicator in the risk table with the gravity field risk characteristics. The individual disaster risk table provides the user's bearing capacity benchmark and personalized risk parameters, and the gravity field risk characteristics provide the dangerous distribution information of the spatial environment. The urgency identification considers multiple dimensions such as risk intensity, impact range, development trend, and user vulnerability. For example, when the individual disaster risk table shows that a user's bearing benchmark is 6.2 levels, and the gravity field risk characteristics show that the current environmental risk reaches 7.5 levels, the risk exceeds the bearing capacity by 1.3 levels, and the corresponding disaster warning urgency is marked as level 3 (a total of 5 levels). The quantification of the urgency uses a numerical scoring system of 0-10, and the higher the score, the higher the urgency. Different urgency levels correspond to different response strategies and warning modes, with low urgency using a reminder mode and high urgency using a forced evacuation mode.

[0060] Based on the disaster warning urgency, the relative position changes between multiple wristbands are determined. According to the identified disaster warning urgency level, the monitoring scheme and analysis strategy of the relative position changes between multiple wristbands are determined. When the urgency is low (0-3 levels), a standard monitoring scheme is used, with a position sampling frequency of 1 per second and a relative position relationship within 50 meters. When the urgency is medium (4-6 levels), an enhanced monitoring scheme is used, with a sampling frequency of 3 per second and an expanded attention range of 100 meters, and joint analysis of relative distance and direction is enabled. When the urgency is high (7-10 levels), a dense monitoring scheme is used, with a sampling frequency of 5 per second and an expanded attention range of 200 meters, and multi-wristband cooperative positioning and trajectory prediction functions are activated. Different urgency levels determine different position change detection thresholds: when the urgency is low, a position change exceeding 5 meters triggers a record, when the urgency is medium, a position change exceeding 3 meters triggers, and when the urgency is high, a position change exceeding 1 meter triggers. The data format and storage strategy of the relative position changes are also adjusted according to the urgency, and real-time transmission and redundant storage are used for high urgency to ensure reliable acquisition of critical position information.

[0061] In some embodiments, the generation of the displacement coordination warning sequence based on the relative position changes between the multiple wristbands includes: performing mobile pattern analysis on the relative position changes between the multiple wristbands to obtain group movement characteristics; performing speed anomaly detection based on the group movement characteristics to form a movement speed deviation value; analyzing the consistency of the movement direction through the movement speed deviation value to generate a direction coordination coefficient; and generating a displacement coordination warning sequence according to the direction coordination coefficient.

[0062] The relative position changes between multiple wristbands are analyzed to obtain group movement characteristics. The relative position change data of multiple monitored wristbands are pattern recognized and feature extracted to identify typical patterns and abnormal behaviors in group movement. Group movement characteristics include collective characteristic parameters such as movement speed distribution, direction consistency, spatial aggregation degree, and trajectory similarity. Movement speed distribution reflects the speed difference of different individuals in the group, and the mean, variance, and distribution form of the speed are obtained through statistical analysis. Direction consistency describes the degree of uniformity of the group's moving direction, and high consistency indicates that the group is moving in the same direction, while low consistency indicates that the moving direction is dispersed. Spatial aggregation quantifies the degree of aggregation of the group in space, and high aggregation indicates that group members are close to each other, while low aggregation indicates that members are distributed dispersedly. Trajectory similarity is obtained by comparing the movement paths of different wristbands, and high similarity indicates that the group adopts similar movement routes. The time window for feature extraction is determined according to the time scale of movement, and is generally set to a time interval of 30 seconds to 5 minutes. The quantification of group movement characteristics uses standardized numerical values to facilitate comparison and comprehensive analysis between different characteristics.

[0063] Rate anomaly detection is performed based on group movement characteristics to form a movement rate deviation value. The extracted group movement characteristics are used as a reference standard to detect anomalies and analyze deviations in the movement rate of individual wristbands in the group. Rate anomaly detection uses statistical anomaly detection methods to set a normal rate range and an anomaly determination threshold. The movement rate deviation value is calculated by the difference between the individual rate and the group average rate, and the larger the deviation value, the more abnormal the individual's movement state. Anomaly detection considers rate characteristics under different circumstances, and the group rate should remain relatively consistent during normal evacuation, while individual rates may be too fast or too slow under abnormal circumstances. For example, during an evacuation process, the group average movement speed is 2.5 meters / second, but a wristband shows that the user's movement speed is only 0.8 meters / second, and the calculated rate deviation value is -1.7 meters / second, indicating that the user may encounter difficulty in moving and need help. The formula for calculating the deviation value is Δv = vi-v_avg, where Δv is the rate deviation value, vi is the individual movement rate, and v_avg is the group average rate. The setting of the anomaly threshold considers factors such as the age structure, physical condition, and environmental conditions of the group. Statistical analysis of the rate deviation value provides overall characteristics of the deviation distribution and identification results of individual anomalies.

[0064] A direction coordination coefficient is generated by analyzing the consistency of moving direction through the moving speed bias value. Based on the calculated moving speed bias value, the coordination degree and consistency level of the group moving direction are further analyzed to generate a quantitative direction coordination evaluation index. The calculation formula of the direction coordination coefficient is: direction coordination coefficient =∑(wi×cos(|θi-θ_avg|)) / N, where wi is the weight coefficient based on the speed bias, θi is the moving direction angle of the ith bracelet, θ_avg is the average moving direction angle of the group, and N is the total number of bracelets. For example, in a campus evacuation, 15 users move in the northeast direction (angle 45 degrees), 3 in the southeast direction (angle 135 degrees), and 2 in the northwest direction (angle 315 degrees). The calculated direction coordination coefficient is 0.75, indicating a good coordination degree. If the moving directions of 20 users are completely random and scattered, the coordination coefficient will be close to 0.1, indicating chaotic movement that needs to be guided. The consistency analysis considers the influence of speed bias on direction coordination, and individuals with abnormal moving speed are given a lower weight to avoid the influence of individual abnormal users on the overall coordination evaluation. The time variation of the coordination coefficient is obtained by analyzing the sliding window of 5 consecutive minutes. A high and continuous coordination coefficient indicates that the group evacuation is well organized. When the direction coordination coefficient is less than 0.5, it indicates that the group movement is chaotic and uncoordinated, and a coordination warning of "please move uniformly towards the safe exit direction" needs to be issued.

[0065] For example, the displacement coordination warning sequence is generated based on the direction coordination coefficient, including: performing moving trajectory analysis based on the direction coordination coefficient to obtain trajectory variation parameters; performing safety distance evaluation using the trajectory variation parameters to form a distance control index; analyzing group distribution characteristics based on the distance control index to generate a group gathering density index; and generating a displacement coordination warning sequence based on the group gathering density index.

[0066] The trajectory variation parameters are obtained by performing moving trajectory analysis based on the direction coordination coefficient. Moving trajectory analysis extracts geometric and dynamic features of the trajectory from a sequence of consecutive position points. The trajectory variation parameters include kinematic parameters such as path curvature, direction change rate, speed change rate, and trajectory deviation degree. Path curvature reflects the degree of curvature of the trajectory, with high curvature indicating a clear trajectory and low curvature indicating a relatively straight trajectory. The direction change rate describes the speed of change of the moving direction over time, with a high change rate indicating frequent direction changes and a low change rate indicating relatively stable directions. The speed change rate reflects the degree of fluctuation in moving speed, with high fluctuation indicating unstable movement and low fluctuation indicating stable movement. The trajectory deviation degree is quantified by the deviation of the individual trajectory from the ideal evacuation path, with a high deviation degree indicating a large deviation from the evacuation path. The time resolution for parameter extraction is determined according to the accuracy requirements of trajectory analysis, typically using a time interval of 1-5 seconds. The statistical features of the trajectory variation parameters provide overall variation patterns and individual difference information for the group trajectory.

[0067] The distance control index is formed by evaluating the safety distance with trajectory variation parameters. Four trajectory variation parameters are used for comprehensive safety distance analysis: path curvature is used to judge the influence of trajectory bending on safety distance, and the trajectory area with large curvature needs to increase safety distance to cope with direction change. Direction change rate is used to evaluate the influence of movement stability on distance requirement, and individuals with frequent direction changes need larger safety buffer space. Speed change rate is used to analyze movement consistency, and groups with large speed fluctuations need to set more relaxed safety distance standards. Trajectory deviation degree is used to determine the coordination degree of individuals and groups, and individuals with high deviation degree need special distance control strategy. The distance control index is obtained by statistical analysis of the distance distribution characteristics between adjacent individuals in the group: the minimum distance reflects the most crowded condition in the trajectory complex area, the average distance describes the overall evacuation density considering the trajectory factors, the distance variance reflects the degree of distance distribution unevenness caused by trajectory difference, and the dangerous distance proportion is the proportion of individuals with distance less than the safety threshold in the trajectory complex situation. The distance control index finally formed includes statistical parameters such as minimum distance, average distance, distance variance and dangerous distance proportion.

[0068] The group gathering density index is generated by analyzing the group distribution characteristics through the distance control index. Based on the distance control index, the group distribution characteristics are analyzed in depth: the minimum distance index is used to identify the crowded hot spot area in the group, and the position with minimum distance less than the safety threshold is marked as a high-risk gathering point. The average distance index is used to evaluate the overall distribution density level of the group, and the smaller the average distance, the higher the group gathering density. The distance variance index is used to analyze the uniformity of group distribution, and the area with large variance has uneven density distribution and needs to be focused on. The dangerous distance proportion index is used to quantify the overall safety risk level, and a high proportion indicates that the group has a general problem of over-concentration. The calculation of the group gathering density index uses the index fusion method: the local density heat map is generated based on the minimum distance index to identify the most serious gathering space position; the overall density level is calculated using the average distance index to provide basis for global evacuation strategy; the density gradient distribution is identified through the distance variance index to determine the area that needs to be diverted and adjusted; and the group safety state is evaluated by combining the dangerous distance proportion index, and an emergency dispersion warning is triggered when the proportion exceeds the threshold. The finally generated group gathering density index is a quantitative density level value, which is used to guide the collaborative warning decision-making in the evacuation process.

[0069] The displacement coordination warning sequence is generated according to the crowd gathering density index. The displacement coordination warning sequence adopts a differentiated warning strategy and content design according to different levels of gathering density. When the gathering density is within the safe range, the maintenance warning of "maintain the current evacuation rhythm and pay attention to following the person in front" is sent. When the gathering density approaches the warning line, the adjustment warning of "there are many people in front, please slow down and pass orderly" is sent. When the gathering density exceeds the safety threshold, the shunting warning of "the front is too crowded, please detour other evacuation routes" is sent. The spatial pertinence of the warning sequence is realized by identifying high-density areas and low-density areas. The personnel in the high-density areas receive deceleration and shunting instructions, and the personnel in the low-density areas receive acceleration and gathering instructions. The time scheduling of the sequence considers the emergency degree of evacuation and the response time of the warning, and shortens the warning interval and increases the scheduling frequency in emergency situations. The personalized design of the warning content considers the position, movement state and individual characteristics of the receiver, and provides personalized evacuation guidance for different users.

[0070] In step S150, active escape response data is generated based on the displacement coordination warning sequence, and the inertia swing amplitude of the active escape response data is analyzed to extract a motion coordination efficiency coefficient, and a disaster emergency coefficient is generated based on the motion coordination efficiency coefficient.

[0071] Specifically, active escape response data is generated based on the displacement coordination warning sequence. According to the content characteristics and emergency level of the displacement coordination warning sequence, the corresponding protection response strategy is divided into four basic types of active escape, passive waiting, in-place protection and assisting others. The maintenance warning sequence such as "maintain the current evacuation rhythm" corresponds to the passive waiting type of protection strategy, and the user maintains the current state and waits for further instructions. The adjustment warning sequence such as "please slow down and pass orderly" corresponds to the in-place protection type of protection strategy, and the user takes corresponding adjustment measures at the current position. The shunting warning sequence such as "please detour other evacuation routes" corresponds to the active escape type of protection strategy, and the user immediately starts to change the moving path for evacuation. The assisting warning sequence such as "please help the disabled person" corresponds to the assisting others type of protection strategy, and the user turns to help other personnel for evacuation. After the protection classification is completed, the system focuses on monitoring and analyzing the active escape type of user response data, and detects whether the user starts to move quickly, changes direction or adjusts path according to the warning instruction through the bracelet. The active escape response data includes behavior parameters such as response time, movement speed change, movement direction adjustment, duration and response strength. The collection of response data is realized through the motion detection devices such as three-axis acceleration sensor, gyroscope and magnetometer in the bracelet.

[0072] The inertial swing amplitude analysis is performed on the active escape response data to extract the motion coordination efficiency coefficient. The obtained active escape response data is analyzed kinematically, and the inertial swing characteristics and motion coordination ability of the user during the escape process are analyzed. The inertial swing amplitude analysis extracts the periodic swing component and amplitude characteristics from the bracelet acceleration data through signal processing. The human body will produce inertial motion such as upper limb swing, pace rhythm and body center of gravity shift during rapid movement. The coordination degree of these motions reflects the motion ability and escape efficiency of the user. The swing amplitude is obtained by calculating the peak value and periodic characteristics of the acceleration signal. The coordinated motion is characterized by moderate swing amplitude and stable rhythm. The motion coordination efficiency coefficient is calculated by comprehensively calculating parameters such as swing frequency, amplitude stability and left-right symmetry. The swing frequency reflects the rhythm characteristics of the motion, and the swing frequency of normal fast walking or slow running is 1.5-2.5Hz. The amplitude stability describes the degree of change of the swing amplitude, and stable swing indicates good motion coordination. The left-right symmetry is quantified by comparing the swing difference between the left and right arms, and good symmetry indicates good motion balance. The calculation formula of the coordination efficiency coefficient is: motion coordination efficiency coefficient = frequency stability x amplitude consistency x symmetry index, the numerical range is 0-1, and the higher the value, the better the motion coordination.

[0073] In some embodiments, the disaster emergency coefficient is generated based on the motion coordination efficiency coefficient, including: based on the motion coordination efficiency coefficient, a coordination degree decay parameter is obtained by timing change analysis; the coordination degree decay parameter is used to adjust the vibration intensity of the bracelet to form an adaptive vibration mode; a vibration alarm coefficient is generated by matching the adaptive vibration mode with the bracelet alarm function; and the coordination degree decay parameter and the vibration alarm coefficient are combined to generate a disaster emergency coefficient.

[0074] The coordination degree attenuation parameter is obtained by time series analysis based on the motion coordination efficiency coefficient. The motion coordination efficiency coefficient obtained is used as the analysis reference for time series analysis. The sliding window method is used to observe the downward trend and change law of the motion coordination efficiency coefficient over time. The coordination degree attenuation parameter reflects the change characteristics of the user's coordination ability from being mentally full to being physically exhausted during continuous movement. The attenuation analysis considers the influence of factors such as physical consumption caused by long-time fast evacuation and fatigue accumulation caused by continuous stress state on motion coordination. The coordination degree attenuation parameter includes key features such as attenuation rate, attenuation amplitude, attenuation start time, and attenuation stable value. The attenuation rate describes the speed of the decline in coordination ability. Some people may experience a decline in coordination soon after the evacuation begins, while others can maintain a good state for a long time. The attenuation amplitude reflects the overall degree of decline in coordination ability, which is the difference between the good coordination state at the beginning of the evacuation and the fatigue state later. The attenuation start time identifies the time node at which the coordination ability begins to decline significantly, similar to the time at which physical strength begins to decline significantly. The attenuation stable value represents the stable level of coordination ability after it has declined, reflecting the basic movement ability that the user can maintain in a state of fatigue.

[0075] The vibration intensity of the bracelet is adjusted using the coordination degree attenuation parameter to form an adaptive vibration mode. The vibration reminder is adjusted intelligently based on the coordination degree attenuation parameter: when the user's attenuation rate is detected to be fast, it means that the user's physical strength is declining quickly, and the bracelet increases the vibration reminder frequency to maintain their attention; when the attenuation amplitude is large, it means that the user's decline from a good state to a fatigue state is large, and the bracelet uses more obvious vibration intensity to compensate for the decline in attention. The attenuation start time parameter is used to determine when to start strengthening the vibration reminder, such as when the user first shows signs of fatigue. The attenuation stable value parameter is used to set a long-term vibration strategy, and when the user's coordination ability stabilizes at a low level, the bracelet continues to use a strong vibration mode to ensure that the user can perceive important information. Adjusting the vibration intensity is like adjusting the alarm volume according to a person's fatigue level: when you are mentally good, a gentle reminder is enough, but when you are tired, you need a stronger stimulus to attract attention. The adaptive vibration mode selects the appropriate vibration strategy according to different combinations of attenuation parameters: for users with good physical strength and slow attenuation, a gentle reminder is used, and for users with poor physical strength and fast attenuation, a strong reminder is used. The vibration mode switches gradually to avoid sudden strong vibrations that may disturb tired users.

[0076] The adaptive vibration mode is matched with the bracelet alarm function to generate a vibration alarm coefficient. The adaptive vibration mode formed is matched with various alarm functions of the smart bracelet to generate a quantitative vibration alarm effect evaluation index. The bracelet alarm function includes vibration alarm, sound alarm, light alarm, and haptic feedback and other reminding methods. The vibration alarm coefficient reflects the coordination effect of the vibration mode and other alarm functions, and the higher the coefficient, the better the alarm effect. The matching analysis considers the perception characteristics and attention attracting ability of different alarm methods. The vibration alarm is suitable for quiet environments, the sound alarm is suitable for noisy environments, and the light alarm is suitable for dark environments. The combination strategy of the alarm function is optimized according to the environmental conditions and the user state. In a noisy evacuation environment, the vibration alarm and the sound alarm are simultaneously enabled to achieve better results. The calculation of the vibration alarm coefficient comprehensively considers vibration intensity, alarm frequency, duration, and user response rate and other factors. The vibration intensity affects the perception of the user, and the perception effect is best when the intensity is moderate. The alarm frequency affects the persistence of attention, and a high frequency is easy to produce adaptive neglect. The duration affects the effectiveness of the alarm, and a short time may be missed, and a long time may interfere with action. The user response rate is statistically obtained by monitoring the behavior reaction of the user after receiving the alarm. The value range of the vibration alarm coefficient is set to 0-10 points, and the higher the score, the better the alarm effect.

[0077] The coordination degree attenuation parameter and the vibration alarm coefficient are combined to generate a disaster emergency coefficient. The obtained coordination degree attenuation parameter and the vibration alarm coefficient are comprehensively analyzed and numerically fused to generate a final disaster emergency ability comprehensive evaluation index. The coordination degree attenuation parameter reflects the user's internal emergency ability, including physical factors such as physical level, motor coordination, and endurance. The vibration alarm coefficient reflects the effectiveness of external auxiliary means, including technical factors such as device performance, alarm effect, and user acceptance. The calculation formula of the emergency coefficient is: disaster emergency coefficient = 0.7 x coordination degree comprehensive score + 0.3 x vibration alarm coefficient, and the weight distribution reflects the principle that internal ability is more important than external assistance. The coordination degree comprehensive score is obtained by reverse conversion of the attenuation parameter, and the slower the attenuation, the higher the score, and the faster the attenuation, the lower the score. The value range of the emergency coefficient is standardized to 0-100 points, which is convenient for comparison and analysis between different users. The grading standard of the coefficient divides the emergency ability into five levels: A level (more than 90 points), B level (75-90 points), C level (60-75 points), D level (45-60 points), and E level (less than 45 points). Different levels of users adopt differentiated evacuation strategies and auxiliary measures in disaster response. A-level users can evacuate independently, and E-level users need special help.

[0078] In step S160, a multi-device linkage triggering area is identified based on the disaster emergency coefficient, a sound wave positioning network is constructed for the multi-device linkage triggering area, sound wave network cooperative alarm data is generated based on the sound wave positioning network, and the smart bracelet earthquake early warning service is completed.

[0079] Specifically, the multi-device linkage triggering area is identified based on the disaster emergency coefficient. The disaster emergency coefficient generated in step S150 is used as the basis for area division to identify the key area range that needs to start multi-device collaborative response. When the average emergency coefficient of users in the area is lower than the set threshold, the area is marked as a linkage triggering area. The threshold setting of linkage triggering considers the emergency ability distribution and risk tolerance level of users. The area with an emergency coefficient lower than 60 points needs to start device linkage support. The triggering area includes three types of bracelet device intensive area, weak emergency ability area and high risk impact area. The bracelet device intensive area corresponds to the area with a large number of bracelet users, and device linkage can play a greater collaborative effect. The weak emergency ability area corresponds to the area with a low average emergency coefficient of users, which needs to provide additional guidance and support through device linkage. The high risk impact area corresponds to the area with serious earthquake impact and difficult evacuation, and device linkage helps to improve evacuation efficiency and safety. The spatial resolution of area identification is set to a grid unit of 50m x 50m, and the emergency coefficient and linkage triggering condition are calculated independently for each grid unit. The dynamic adjustment function of the triggering area updates the area boundary in real time according to the user movement and emergency coefficient change. The linkage device includes smart bracelet, emergency broadcast, lighting device and evacuation indication device and other types.

[0080] A sound wave positioning network is constructed for the multi-device linkage triggering area. In the identified linkage triggering area, a sound wave network for precise positioning and collaborative communication is constructed using the sound wave emission and reception functions of smart bracelets and other devices. The sound wave positioning network achieves precise positioning by measuring the propagation time and time difference of arrival of sound waves between different devices. The network construction uses self-organizing network technology, and each device automatically establishes connection relationship and communication link according to signal strength and propagation quality. The working frequency of sound wave positioning is selected in the ultrasonic frequency band outside the human ear hearing range to avoid auditory interference to users. The network topology structure adopts mesh structure design, each device establishes connection with multiple surrounding devices to improve the redundancy and reliability of the network. The positioning accuracy is guaranteed by triangulation and multi-point correction techniques, and the positioning error is controlled within 1-3 meters. The coverage range of the sound wave network is consistent with the linkage triggering area, ensuring that all users in the area can receive positioning and communication services.

[0081] In some embodiments, the sound wave network collaborative alarm data is generated based on the sound wave positioning network, including: performing propagation distance analysis on the sound wave positioning network to identify near-field sound waves and far-field sound waves; performing height difference influence evaluation based on the near-field sound waves and the far-field sound waves to form a height compensation parameter; using the height compensation parameter to perform multi-bracelet sound wave time calibration to generate a synchronous sound wave signal; and generating sound wave network collaborative alarm data according to the synchronous sound wave signal.

[0082] The propagation distance of the sound wave positioning network is analyzed to identify near-field sound waves and far-field sound waves. In the constructed sound wave positioning network, the propagation distance of the sound wave signal is classified and analyzed, and the sound wave signal is divided into near-field and far-field according to the nearness of the propagation distance. The near-field sound wave is defined as the sound wave signal with a propagation distance of less than 50 meters, which has the characteristics of short propagation time, high signal strength and small environmental interference. The far-field sound wave is defined as the sound wave signal with a propagation distance of more than 50 meters, which has the characteristics of long propagation time, obvious signal attenuation and easy to be affected by the environment. The propagation distance analysis is calculated by measuring the time delay and signal strength attenuation of the sound wave from transmission to reception. The distance calculation formula is: distance = sound speed x propagation time, wherein the sound speed is corrected according to the environmental temperature and humidity. The signal processing of the near-field sound wave adopts a direct receiving mode, and the signal quality is good and does not need complex compensation processing. The signal processing of the far-field sound wave needs to consider the influence factors such as propagation loss, multipath effect and environmental noise. The boundary threshold of signal classification is set according to the network coverage demand and positioning accuracy requirement, and the 50-meter demarcation distance can balance the signal quality and coverage range. The analysis result is used to optimize the network topology and signal routing, and the near-field connection is preferentially used for high-precision positioning, and the far-field connection is used to expand the network coverage.

[0083] The height difference influence evaluation is based on the near-field sound wave and the far-field sound wave to form the height compensation parameter. The identified near-field sound wave and far-field sound wave signal are analyzed for height difference influence, and the influence degree of different height positions on sound wave propagation is evaluated. For example, when the bracelet of the user on the first floor communicates with the bracelet of the user on the fifth floor, the sound wave needs to pass through multiple building structures, and the propagation path is about 15-20 meters longer than the horizontal propagation between users on the same floor, and the propagation time is increased by 0.04-0.06 seconds. The height difference is calculated by the air pressure sensor and altimeter built-in the bracelet, and the air pressure change can reflect the height difference. The influence evaluation analysis includes propagation path lengthening, signal attenuation increase and reflection scattering effect. The propagation path lengthening leads to the increase of sound wave arrival time, which affects the accuracy of positioning calculation. The signal attenuation increase reduces the strength of the received signal, which may affect the identification and analysis of the signal. The reflection scattering effect makes the sound wave propagation complex, resulting in multipath propagation and interference phenomenon. The height compensation parameter includes path compensation coefficient, attenuation compensation coefficient and time compensation coefficient. The path compensation coefficient is used to correct the propagation distance error caused by the height difference. The attenuation compensation coefficient is used to compensate for the signal strength loss caused by height propagation. The time compensation coefficient is used to correct the time delay error caused by the change of the propagation path.

[0084] Synchronous acoustic signals are generated by multi-wristband acoustic time calibration using height compensation parameters. The compensation parameters are applied to the acoustic time synchronization calibration process between multiple wristbands: the path compensation coefficient is used to correct the propagation distance calculation, and the actual propagation path length difference caused by the height difference is compensated into the distance measurement result. The attenuation compensation coefficient is used to adjust the intensity threshold of the received signal, and the signal attenuation caused by the height propagation is compensated to ensure that the wristbands at different heights can accurately identify and analyze the acoustic signals. The time compensation coefficient is used to correct the acoustic arrival time measurement, and the additional time delay caused by the change of the propagation path is deducted from the original measurement time. Multi-wristband acoustic time calibration adopts a master-slave synchronization mechanism, selects the wristband with the best signal quality in the network as the master clock, and other wristbands as the slave clock for synchronization calibration. The process of time calibration includes three steps of time deviation measurement, three compensation parameter application and synchronization error correction. The time deviation measurement is realized by the round-trip time measurement method, and the total time of the acoustic signal from transmission to reception and return is calculated. The three compensation parameter application adds the path, attenuation, and time compensation coefficients to the original measurement result according to the propagation physical law, eliminating the systematic error caused by the height difference.

[0085] The sound wave network cooperative alarm data is generated according to the synchronous sound wave signal. Based on the obtained synchronous sound wave signal, the cooperative alarm and information synchronization of multiple bracelet devices in the network are realized, and the cooperative alarm data in a unified format is generated. The sound wave network cooperative alarm data contains key information such as user identity, real-time position coordinates, movement state information, emergency capability level and alarm content. The data generation adopts a distributed summary mode, and each bracelet device contributes its own state information to summarize the global cooperative alarm data through a network protocol. The content design of the cooperative alarm considers the coordination needs of group evacuation, and contains two levels of individual guidance information and group coordination information. The individual guidance information provides personalized evacuation suggestions and safety reminders for specific users. The group coordination information provides overall evacuation strategies and collective action guidance. The bracelet device directly displays the warning information on the screen, including epicenter location, estimated magnitude, warning time, local estimated intensity, epicenter distance, warning level and other core content. The warning level is divided into four levels of red, orange, yellow and blue according to the threat degree, and the color coding and text prompt are combined to facilitate users to quickly identify the danger level and response measures in an emergency. For users with different physiological characteristics, the cooperative alarm data adopts a multi-modal output adaptation mechanism. For visually impaired users, the sound alarm module outputs voice warning information, and the voice broadcast includes key content such as magnitude, warning time and evacuation direction; for hearing-impaired users, the display screen outputs visual warning information, and the display interface uses high-contrast color matching and flashing prompts to enhance the visual warning effect; for users with visual and auditory impairments, the vibration alarm module outputs a tactile warning signal, and the vibration mode is designed with different vibration rhythms and intensities according to the warning level, so that users can identify the danger level through tactile perception; the multi-modal alarm mechanism supports simultaneous activation or selective activation of the three output modes of sound, display and vibration, ensuring that different user groups can receive earthquake warning information in a timely and accurate manner. The transmission of the cooperative alarm data adopts a combination of broadcasting and multicasting, important information is broadcast to all devices, and personalized information is multicast to specific devices, finally completing the intelligent bracelet earthquake warning service.

[0086] In order to perform the intelligent bracelet earthquake warning service method corresponding to the above-mentioned method embodiment, the corresponding functions and technical effects are realized. Referring to Figure 2 , Figure 2 The structure block diagram of the intelligent bracelet earthquake warning service system 200 provided by the embodiment of the application is shown. For ease of illustration, only the parts related to the embodiment are shown, and the intelligent bracelet earthquake warning service system 200 provided by the embodiment of the application includes:

[0087] The signal acquisition module 201 is configured to acquire the earthquake warning signal sent by the earthquake monitoring network and the user physiological parameter, extract the magnitude parameter from the earthquake warning signal, and establish an individual disaster bearing benchmark based on the magnitude parameter and the user physiological parameter;

[0088] The phase detection module 202 is configured to identify a high-risk period of earthquake based on the individual disaster bearing reference, detect a phase difference between a wristband vibration and a ground vibration according to the high-risk period of earthquake, generate disaster damage evaluation data by using the phase difference, and determine an individual disaster risk table according to the disaster damage evaluation data;

[0089] The gravity sensing module 203 is configured to acquire a real-time geographic position of a user and a building seismic grade, generate a building micro-inclination angle by performing gravity sensing detection according to the building seismic grade, and perform coupling analysis on the real-time geographic position of the user and the building micro-inclination angle to form a gravity field risk feature;

[0090] The cooperative positioning module 204 is configured to identify a disaster early warning urgency based on the individual disaster risk table and the gravity field risk feature, determine a relative position change between multiple wristbands based on the disaster early warning urgency, and generate a displacement cooperative early warning sequence by using the relative position change between the multiple wristbands;

[0091] The motion analysis module 205 is configured to perform protection classification based on the displacement cooperative early warning sequence to generate active escape response data, perform inertia swing amplitude analysis on the active escape response data to extract a motion coordination efficiency coefficient, and generate a disaster emergency coefficient based on the motion coordination efficiency coefficient;

[0092] The early warning output module 206 is configured to identify a multi-device linkage triggering area based on the disaster emergency coefficient, construct a sound wave positioning network for the multi-device linkage triggering area, generate sound wave network cooperative alarm data based on the sound wave positioning network, and complete an intelligent wristband earthquake early warning service.

[0093] The intelligent wristband earthquake early warning service system 200 described above can implement an intelligent wristband earthquake early warning service method described above. The optional items in the method embodiment described above are also applicable to this embodiment, and will not be described in detail here. The remaining content of the present embodiment can refer to the content of the method embodiment described above, and will not be described in detail in this embodiment.

[0094] The purpose of the above embodiments is to exemplarily reproduce and deduce the technical solutions of the present application, and to completely describe the technical solutions, purposes and effects of the present application, so as to make the public more thoroughly and comprehensively understand the disclosed content of the present application, and not to limit the protection scope of the present application.

[0095] The above embodiments are not based on an exhaustive enumeration of the present application, and there can be many other unlisted embodiments. Any replacement and improvement made without violating the concept of the present application is within the protection scope of the present application.

Claims

1. A smart bracelet earthquake early warning service method, characterized in that, The method comprises the following steps: obtaining an earthquake early warning signal sent by a seismic monitoring network and a user physiological parameter, extracting a magnitude intensity parameter from the earthquake early warning signal, establishing an individual disaster bearing benchmark based on the magnitude intensity parameter and the user physiological parameter, including: generating a magnitude threat time sequence graph through the magnitude intensity parameter; forming a physiological-magnitude coupling curve by combining the user physiological parameter and the magnitude threat time sequence graph; extracting a safe bearing interval from the physiological-magnitude coupling curve; setting the threshold value of the safe bearing interval as the individual disaster bearing benchmark; wherein the combination of the user physiological parameter and the magnitude threat time sequence graph to form a physiological-magnitude coupling curve comprises: identifying a magnitude sensitive interval according to the magnitude threat time sequence graph to determine an analysis window, the magnitude sensitive interval including a magnitude rising gradient, a duration and a decay rate; forming a physiological response atlas by tracking physiological changes in the analysis window based on the user physiological parameter; extracting the time coordinates of each response point in the physiological response atlas; arranging the response intensity of the time coordinates to form a physiological-magnitude coupling curve; identifying a high-risk period of earthquake based on the individual disaster bearing benchmark, detecting the phase difference between the bracelet vibration and the ground vibration according to the high-risk period of earthquake, generating earthquake damage assessment data using the phase difference, and determining an individual disaster risk table according to the earthquake damage assessment data; obtaining the real-time geographical position of the user and the building seismic resistance level, generating the building micro-inclination angle through gravity sensing detection according to the building seismic resistance level, and coupling the real-time geographical position of the user with the building micro-inclination angle to form a gravity field risk feature; identifying the disaster warning urgency based on the individual disaster risk table and the gravity field risk feature, determining the relative position change between multiple bracelets based on the disaster warning urgency, and generating a displacement collaborative warning sequence using the relative position change between multiple bracelets; based on the displacement collaborative warning sequence, the protective classification produces active escape response data, the inertia swing amplitude analysis extracts the motion coordination efficiency coefficient based on the active escape response data, and the disaster emergency coefficient is generated based on the motion coordination efficiency coefficient; based on the disaster emergency coefficient, a multi-device linkage triggering area is identified, a sound wave positioning network is constructed for the multi-device linkage triggering area, sound wave network collaborative alarm data is generated based on the sound wave positioning network, and the intelligent bracelet earthquake early warning service is completed.

2. The method of claim 1, wherein, The use of the phase difference to generate earthquake damage assessment data comprises: decomposing the phase difference into fast wave phase and slow wave phase; projecting a phase detection sequence from the fast wave phase to the slow wave phase; record the position of the amplitude enhancement point on the phase detection sequence to form an enhancement point set; the largest point in the enhancement point set is marked to generate earthquake damage assessment data.

3. The method of claim 1, wherein, The coupling analysis of the real-time geographical position of the user and the building micro-inclination angle to form a gravity field risk feature comprises: performing gravity field intensity analysis on the building micro-inclination angle to obtain inclination parameters; performing risk matching on the real-time geographical position of the user through the inclination parameters to form a position risk coefficient; spatially interpolating the position risk coefficients to generate a risk level distribution map; forming a gravity field risk feature according to the risk level distribution map.

4. The method of claim 1, wherein, The generating a displacement coordination early warning sequence according to the direction coordination coefficient comprises: performing mobile pattern analysis on the relative position changes among the multiple wristbands to obtain group mobile features; performing rate anomaly detection based on the group mobile features to form a mobile rate deviation value; analyzing the consistency of the moving direction through the mobile rate deviation value to generate a direction coordination coefficient; generating a displacement coordination early warning sequence according to the direction coordination coefficient.

5. The method of claim 1, wherein, The generating a disaster emergency coefficient based on the motion coordination efficiency coefficient comprises: performing timing change analysis based on the motion coordination efficiency coefficient to obtain a coordination degree attenuation parameter; adjusting the wristband vibration intensity using the coordination degree attenuation parameter to form an adaptive vibration mode; matching the adaptive vibration mode with the wristband alarm function to generate a vibration alarm coefficient; combining the coordination degree attenuation parameter and the vibration alarm coefficient to generate a disaster emergency coefficient.

6. The method of claim 1, wherein, The generating sound wave network coordination alarm data based on the sound wave positioning network comprises: performing propagation distance analysis on the sound wave positioning network to identify near-field sound waves and far-field sound waves; performing height difference influence evaluation based on the near-field sound waves and the far-field sound waves to form a height compensation parameter; performing multi-wristband sound wave time calibration using the height compensation parameter to generate a synchronous sound wave signal; generating sound wave network coordination alarm data according to the synchronous sound wave signal.

7. The method of claim 4, wherein, The generating a displacement coordination early warning sequence according to the direction coordination coefficient comprises: performing mobile pattern analysis on the relative position changes among the multiple wristbands to obtain group mobile features; performing rate anomaly detection based on the group mobile features to form a mobile rate deviation value; analyzing the consistency of the moving direction through the mobile rate deviation value to generate a direction coordination coefficient; generating a displacement coordination early warning sequence according to the direction coordination coefficient.

8. A smart bracelet earthquake early warning service system, characterized in that, The signal acquisition module is configured to acquire an earthquake early warning signal sent by an earthquake monitoring network and a user physiological parameter, extract a magnitude intensity parameter from the earthquake early warning signal, establish an individual disaster bearing benchmark based on the magnitude intensity parameter and the user physiological parameter, and include: generating a magnitude threat timing diagram based on the magnitude intensity parameter; forming a physiological-magnitude coupling curve by combining the user physiological parameter and the magnitude threat timing diagram; extracting a safe bearing interval from the physiological-magnitude coupling curve; setting a threshold value of the safe bearing interval as the individual disaster bearing benchmark; wherein the forming a physiological-magnitude coupling curve by combining the user physiological parameter and the magnitude threat timing diagram includes: identifying a magnitude sensitive interval according to the magnitude threat timing diagram to determine an analysis window, the magnitude sensitive interval including a magnitude rising gradient, a duration, and an attenuation rate; tracing a physiological change process in the analysis window based on the user physiological parameter to form a physiological response map; extracting time coordinates of each response point in the physiological response map; and arranging the response intensity of the time coordinates to form a physiological-magnitude coupling curve. ​ A phase detection module is configured to identify a high-risk period of earthquake based on the individual disaster endurance benchmark, detect a phase difference between a vibration of the smart band and a ground vibration according to the high-risk period of earthquake, generate disaster damage evaluation data by using the phase difference, and determine an individual disaster risk table according to the disaster damage evaluation data; A gravity sensing module is configured to acquire a real-time geographic position of a user and a building seismic resistance level, generate a building micro-inclination angle by performing gravity sensing detection according to the building seismic resistance level, and perform coupling analysis on the real-time geographic position of the user and the building micro-inclination angle to form a gravity field risk feature; A cooperative positioning module is configured to identify a disaster early warning urgency based on the individual disaster risk table and the gravity field risk feature, determine a relative position change among multiple smart bands based on the disaster early warning urgency, and generate a displacement cooperative early warning sequence by using the relative position change among the multiple smart bands; A motion analysis module is configured to perform protection classification based on the displacement cooperative early warning sequence to generate active escape response data, perform inertia swing amplitude analysis on the active escape response data to extract a motion coordination efficiency coefficient, and generate a disaster emergency coefficient based on the motion coordination efficiency coefficient; An early warning output module is configured to identify a multi-device linkage triggering area based on the disaster emergency coefficient, construct an acoustic positioning network for the multi-device linkage triggering area, generate acoustic network cooperative alarm data based on the acoustic positioning network, and complete a smart band earthquake early warning service.

Images