Motion monitoring and analysis method and device based on smart wearable device control chip

Abstract

The present invention relates to the field of motion data analysis, and in particular to a motion monitoring and analysis method and apparatus based on a smart wearable device control chip. The method comprises the following steps: collecting the user's initial state monitoring parameters, identifying abnormal data, and performing abnormality elimination processing to obtain optimized motion monitoring parameters; performing multimodal state fusion perception based on the optimized motion monitoring parameters to obtain a user's real-time state perception map; obtaining the positioning information of the smart wearable device in real time, performing behavioral pattern feature analysis and dynamic behavior trajectory evolution, and generating a user's real-time behavior trajectory map; and predicting fatigue sign limit points and calculating optimal rest time points for the user's real-time state perception map based on the user's real-time behavior trajectory map to generate a rest warning signal. The present invention achieves accurate analysis of the user's real-time motion state and minimizes health risks.

Description

Technical Field

[0001] The present invention relates to the field of motion data analysis, and in particular to a motion monitoring and analysis method and device based on a smart wearable device control chip. Background Art

[0002] With the rapid development of the Internet of Things (IoT) and mobile computing technologies, smart wearable devices, serving as a vital bridge connecting people, objects, and the cloud, have been widely used in a variety of scenarios, including health monitoring, fitness tracking, and rehabilitation assistance. Motion monitoring technology based on smart wearable devices, with its real-time availability, convenience, and continuous data collection, has become a crucial support for modern health management and personalized fitness guidance. Especially with growing public health awareness and the rapid development of industries such as smart healthcare, remote rehabilitation, and smart fitness, the application boundaries of smart wearable devices are constantly expanding, and the requirements for the accuracy and intelligence of motion monitoring and analysis are also becoming increasingly demanding.

[0003] Current mainstream smart wearable devices typically integrate multiple sensor modules, such as accelerometers, gyroscopes, heart rate sensors, and blood oxygen sensors. These sensors, coordinated by the device's control chip, continuously collect multimodal motion data from the user, including key metrics such as cadence, gait, body rhythm, and heart rate variability. Leveraging this fundamental data, motion monitoring systems can make preliminary assessments of the user's exercise intensity, state changes, and energy expenditure, providing users with exercise recommendations or health warnings. However, in practical applications, traditional motion monitoring methods commonly face the following technical challenges: First, it's difficult to balance acquisition frequency with device power consumption. While high-frequency sampling improves accuracy, it can also lead to rapid battery drain. Second, the accuracy of motion state recognition is limited by data synchronization capabilities and the adaptability of analysis algorithms, making misjudgments particularly prone to occur in dynamic environments. Third, the ability to process abnormal data is insufficient. Some devices are unable to effectively correct and identify sudden changes in motion state or abnormal sensor signals.

[0004] Furthermore, existing methods generally employ fixed acquisition frequencies and regularized analysis mechanisms, lacking the ability to adapt to and learn from individual user characteristics. This makes it impossible to dynamically adjust sampling strategies based on the user's real-time motion state, thus impacting the system's real-time performance and personalized responsiveness. Especially during prolonged wear or high-intensity exercise, issues such as data time synchronization failure, sensor drift, and external interference further diminish the availability and reliability of motion data. To address these issues, a more intelligent motion monitoring and analysis method is urgently needed. Summary of the Invention

[0005] In order to solve the above technical problems, the present invention proposes a motion monitoring and analysis method and device based on a smart wearable device control chip to solve at least one of the above technical problems.

[0006] To achieve the above objectives, the present invention provides a motion monitoring and analysis method based on a smart wearable device control chip, comprising the following steps:

[0007] Step S1: Collect the user's initial state monitoring parameters, identify abnormal data, and perform abnormal elimination processing to obtain optimized motion monitoring parameters;

[0008] Step S2: performing multimodal state fusion perception based on the optimized motion monitoring parameters to obtain a real-time state perception map of the user;

[0009] Step S3: Acquire the positioning information of the smart wearable device in real time, analyze the behavioral pattern characteristics and dynamic behavior trajectory evolution, and generate a real-time user behavior trajectory map;

[0010] Step S4: predicting the fatigue sign limit point and calculating the optimal rest time point of the user's real-time state perception map based on the user's real-time behavior trajectory map, and generating a rest warning signal;

[0011] Step S5: Obtain historical motion monitoring parameters, and perform transient state mutation identification and state offset calculation for each state point on the user's real-time state perception graph to obtain multiple state point offset values;

[0012] Step S6: Quantitatively assess the user status risk of multiple status point offset values, make intelligent health decisions, and build an intelligent risk handling strategy.

[0013] By collecting initial user status monitoring parameters and combining them with abnormal data identification, the system can effectively eliminate errors, noise, and invalid data, ensuring more accurate final motion monitoring parameters. This process is fundamental and critical, improving the reliability of subsequent analysis. After removing abnormal data, the user's motion data is purer, facilitating subsequent behavioral analysis and health assessment. Data inaccuracies may result from improper device wear or short-term environmental interference; this step cleans this data to ensure the accuracy of user health monitoring. Through multimodal state fusion, the system integrates different types of sensor data (such as acceleration, heart rate, temperature, and location) to obtain a more comprehensive real-time user status. This process overcomes the limitations of a single data source and provides richer information for health status monitoring. By integrating multiple sensor data, the real-time state perception map can provide a more detailed analysis of user behavior, assessing information such as physical strength, fatigue, and exercise intensity. This is of great significance for health management systems, especially for accurate assessments in motion monitoring. By using real-time positioning information and behavioral pattern analysis, the system can accurately track user activity trajectories. The generated behavior trajectory map not only displays the user's real-time location but also reflects the evolution of their dynamic behavior, providing a basis for further health management decisions. By tracking the user's activity trajectory, the system can adjust exercise intensity or provide rest recommendations in real time. This helps determine whether to reduce workload or increase exercise intensity based on the user's real-time exercise status, thereby developing a personalized health plan. By analyzing the user's behavior trajectory map in real time and combining it with their real-time state perception map, the system can predict the user's fatigue threshold, preventing excessive exercise. Accurate fatigue prediction can effectively reduce the risk of sports injuries. Based on the fatigue threshold prediction, the system calculates the optimal rest time and generates a rest warning signal. Users can use these recommendations to adjust their exercise plan in a timely manner to avoid excessive fatigue and optimize exercise results. By identifying transient state mutations, the system can promptly detect abnormal fluctuations in the user's exercise or health status and identify potential health risks. Abnormal changes such as sudden heart rate acceleration and fluctuations in exercise intensity may indicate the emergence of health issues. By calculating the offset of each state point, the system can provide more personalized health analysis reports, helping users understand their health trends and make appropriate adjustments. By quantifying and assessing the user's status deviation values, the system can comprehensively analyze the user's health risks and provide a more scientific and accurate health status assessment. This helps users understand their health status in real time and prevent potential health problems. Based on the assessment results, the system makes intelligent health decisions and can automatically recommend exercise plans, rest time, or dietary advice to users. This intelligent decision-making mechanism not only provides personalized health plans but also reduces the risks caused by human misjudgment. By combining risk quantification with treatment strategies, the system can proactively intervene in the user's health status and provide customized solutions.When a user shows signs of excessive fatigue, the system may automatically suggest a break or remind the user to reduce exercise intensity to minimize health risks.

[0014] In this specification, a motion monitoring and analysis device based on a smart wearable device control chip is provided, which is used to perform the motion monitoring and analysis method based on a smart wearable device control chip as described above, including:

[0015] The data processing module is used to collect the user's initial state monitoring parameters, identify abnormal data, and perform abnormal elimination processing to obtain optimized motion monitoring parameters;

[0016] A state perception module, configured to perform multimodal state fusion perception based on the optimized motion monitoring parameters to obtain a real-time state perception map of the user;

[0017] The trajectory evolution module is used to obtain the positioning information of smart wearable devices in real time, analyze the characteristics of behavioral patterns and dynamically evolve behavioral trajectories, and generate a real-time user behavior trajectory map;

[0018] The fatigue sign prediction module is used to predict the fatigue sign limit point and calculate the optimal rest time point based on the user's real-time behavior trajectory map and the user's real-time state perception map, and generate a rest warning signal;

[0019] The state offset module is used to obtain historical motion monitoring parameters, identify transient state mutations in the user's real-time state perception map, and calculate the state offset of each state point to obtain multiple state point offset values;

[0020] The status risk analysis module is used to quantitatively assess user status risks based on multiple status point offset values, make intelligent health decisions, and build intelligent risk handling strategies.

[0021] This invention collects initial user status monitoring parameters and integrates an abnormal data identification mechanism to ensure the quality of monitoring data. By eliminating invalid, erroneous, or noisy data, the optimized motion monitoring parameters are more representative, providing a more reliable data source for subsequent analysis. After removing abnormal data, the system can more accurately capture the user's health dynamics. If data errors are caused by improper wear or sensor failure, the module automatically identifies and eliminates inaccurate monitoring data to prevent it from affecting subsequent status perception and decision-making. By fusing data from multiple sensors (such as acceleration, heart rate, location, and body temperature), the system provides a comprehensive assessment of the user's health status. Fusion perception not only eliminates bias in single sensor data but also provides more accurate real-time health monitoring. Multimodal state fusion enables the system to monitor and perceive user health parameters such as exercise intensity, fatigue, and physical exertion in real time, helping users fully understand their current exercise status and make timely adjustments. By obtaining positioning information from the smart wearable device, the module can create a real-time map of the user's behavior trajectory, reflecting their specific activity patterns and location changes. By analyzing behavioral pattern features, the system can identify the user's activity type, exercise intensity, and the evolution of their behavior patterns. As user behavior evolves, the system can adjust exercise recommendations or health guidance based on real-time patterns. If a user exhibits signs of excessive fatigue or poor posture, the system can recommend adjusting exercise patterns or taking a break. By analyzing user behavior in real time, the module can predict fatigue thresholds, promptly identifying whether a user has reached the threshold of overexertion or fatigue. This helps prevent injuries or health problems caused by excessive exercise. Based on the user's fatigue status, the module calculates optimal rest times and provides personalized rest recommendations, preventing fatigue accumulation and overexertion, and improving exercise effectiveness and safety. By comparing historical exercise data with real-time status graphs, the system can detect transient changes in the user's status (such as a sudden increase in heart rate or a sharp change in exercise intensity), which may indicate health risks or signs of excessive exercise. Promptly identifying these sudden changes helps prevent injury or increased health risks. By calculating offsets at each status point, the system can identify subtle changes in the user's health status and use these offsets to assess health risks, providing a reference for subsequent intelligent decision-making. By quantifying the offsets across multiple status points, the system can calculate the current health risk for each user and assess their health status in real time. The system can determine whether a user is at risk of overexertion, excessive fatigue, or other related issues, and can then intervene accordingly. Based on the risk assessment results, the system can provide users with personalized health decision-making recommendations. If the system detects excessive exercise intensity or a high risk of fatigue, it will automatically generate a health intervention strategy, recommending appropriate rest or adjustment of exercise intensity. This strategy can effectively prevent health problems and improve the effectiveness and safety of exercise. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 This is a schematic flow chart of the steps of a motion monitoring and analysis method based on a smart wearable device control chip of the present invention;

[0023] Figure 2 Detailed implementation flow chart of step S1;

[0024] Figure 3 Detailed implementation flow chart of step S2;

[0025] Figure 4 Schematic diagram of the detailed implementation steps of step S3. DETAILED DESCRIPTION

[0026] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0027] This application provides a motion monitoring and analysis method and apparatus based on a smart wearable device control chip. The execution entities of the motion monitoring and analysis method and apparatus based on a smart wearable device control chip include, but are not limited to, the following: mechanical equipment, data processing platforms, cloud server nodes, network upload devices, etc. equipped with the system, which can be regarded as general computing nodes of this application. The data processing platform includes, but is not limited to, at least one of an audio and image management system, an information management system, and a cloud data management system.

[0028] See also Figures 1 to 4 The present invention provides a motion monitoring and analysis method based on a smart wearable device control chip, the motion monitoring and analysis method based on a smart wearable device control chip comprises the following steps:

[0029] Step S1: Collect the user's initial state monitoring parameters, identify abnormal data, and perform abnormal elimination processing to obtain optimized motion monitoring parameters;

[0030] Step S2: performing multimodal state fusion perception based on the optimized motion monitoring parameters to obtain a real-time state perception map of the user;

[0031] Step S3: Acquire the positioning information of the smart wearable device in real time, analyze the behavioral pattern characteristics and dynamic behavior trajectory evolution, and generate a real-time user behavior trajectory map;

[0032] Step S4: predicting the fatigue sign limit point and calculating the optimal rest time point of the user's real-time state perception map based on the user's real-time behavior trajectory map, and generating a rest warning signal;

[0033] Step S5: Obtain historical motion monitoring parameters, and perform transient state mutation identification and state offset calculation for each state point on the user's real-time state perception graph to obtain multiple state point offset values;

[0034] Step S6: Quantitatively assess the user status risk of multiple status point offset values, make intelligent health decisions, and build an intelligent risk handling strategy.

[0035] By collecting initial user status monitoring parameters and combining them with abnormal data identification, the system can effectively eliminate errors, noise, and invalid data, ensuring more accurate final motion monitoring parameters. This process is fundamental and critical, improving the reliability of subsequent analysis. After removing abnormal data, the user's motion data is purer, facilitating subsequent behavioral analysis and health assessment. Data inaccuracies may result from improper device wear or short-term environmental interference; this step cleans this data to ensure the accuracy of user health monitoring. Through multimodal state fusion, the system integrates different types of sensor data (such as acceleration, heart rate, temperature, and location) to obtain a more comprehensive real-time user status. This process overcomes the limitations of a single data source and provides richer information for health status monitoring. By integrating multiple sensor data, the real-time state perception map can provide a more detailed analysis of user behavior, assessing information such as physical strength, fatigue, and exercise intensity. This is of great significance for health management systems, especially for accurate assessments in motion monitoring. By using real-time positioning information and behavioral pattern analysis, the system can accurately track user activity trajectories. The generated behavior trajectory map not only displays the user's real-time location but also reflects the evolution of their dynamic behavior, providing a basis for further health management decisions. By tracking the user's activity trajectory, the system can adjust exercise intensity or provide rest recommendations in real time. This helps determine whether to reduce workload or increase exercise intensity based on the user's real-time exercise status, thereby developing a personalized health plan. By analyzing the user's behavior trajectory map in real time and combining it with their real-time state perception map, the system can predict the user's fatigue threshold, preventing excessive exercise. Accurate fatigue prediction can effectively reduce the risk of sports injuries. Based on the fatigue threshold prediction, the system calculates the optimal rest time and generates a rest warning signal. Users can use these recommendations to adjust their exercise plan in a timely manner to avoid excessive fatigue and optimize exercise results. By identifying transient state mutations, the system can promptly detect abnormal fluctuations in the user's exercise or health status and identify potential health risks. Abnormal changes such as sudden heart rate acceleration and fluctuations in exercise intensity may indicate the emergence of health issues. By calculating the offset of each state point, the system can provide more personalized health analysis reports, helping users understand their health trends and make appropriate adjustments. By quantifying and assessing the user's status deviation values, the system can comprehensively analyze the user's health risks and provide a more scientific and accurate health status assessment. This helps users understand their health status in real time and prevent potential health problems. Based on the assessment results, the system makes intelligent health decisions and can automatically recommend exercise plans, rest time, or dietary advice to users. This intelligent decision-making mechanism not only provides personalized health plans but also reduces the risks caused by human misjudgment. By combining risk quantification with treatment strategies, the system can proactively intervene in the user's health status and provide customized solutions.When a user shows signs of excessive fatigue, the system may automatically suggest a break or remind the user to reduce exercise intensity to minimize health risks.

[0036] In the embodiment of the present invention, see Figure 1 , is a schematic flow chart of the steps of a motion monitoring and analysis method based on a smart wearable device control chip of the present invention. In this example, the steps of the motion monitoring and analysis method based on a smart wearable device control chip include:

[0037] Step S1: Collect the user's initial state monitoring parameters, identify abnormal data, and perform abnormal elimination processing to obtain optimized motion monitoring parameters;

[0038] In this embodiment, it is necessary to determine the parameters for initial user status monitoring, including heart rate, blood oxygen saturation, body temperature, and activity level (static or dynamic). These parameters can be acquired using sensors within the smart wearable device, ensuring high-precision physiological monitoring capabilities. Appropriate equipment should be selected, such as a heart rate monitor, pulse oximetry monitor, and temperature sensor. The data acquisition frequency should be set to 1 Hz (once per second) to obtain real-time information about the user's physiological status. Data collection should begin with the user in a static state (typically while sitting or standing) and last for 3 to 5 minutes. This period is sufficient to capture the user's baseline physiological parameters. Assume that the heart rate recorded during static monitoring is 70-75 bpm, blood oxygen saturation is 95%-98%, and body temperature is 36.5-37.0°C. The data is stored within the device with a timestamp, forming a preliminary dataset for subsequent analysis. Abnormal data is defined as measurements outside the normal physiological range. The normal range for heart rate is typically 60-100 bpm; blood oxygen saturation should be between 90% and 100%; and the normal range for body temperature is 36.1-37.2°C. A threshold for identifying abnormal data is determined. For example, if the heart rate is below 60 bpm or above 100 bpm, the blood oxygen saturation is below 90%, or the body temperature exceeds 37.5°C, it is marked as abnormal. Statistical analysis is performed on the initial state monitoring parameters, calculating the mean and standard deviation of each parameter. If the mean heart rate is 73 bpm and the standard deviation is 5 bpm, the Z-score method can be used to identify abnormal data: Z = (X - μ) / σ, where X is the observed value, μ is the mean, and σ is the standard deviation. If the absolute value of the Z-score is greater than 3, the data point is marked as abnormal. For example, if a heart rate measurement is 110 bpm and the Z-score is 3.5, the data point is identified as abnormal. Once outliers are identified, they must be removed to ensure that subsequent analysis is based on accurate data. This can be done by directly deleting outliers from the dataset or replacing them with the mean. In the preliminary dataset, the values ​​of each parameter are examined individually to identify all outliers. For example, if two values ​​(110 bpm and 55 bpm) in the heart rate data are marked as outliers, they are removed from the dataset. After these removals, the mean and standard deviation of the remaining data are recalculated to obtain a more accurate baseline. After removing the outliers, the new mean heart rate might be 71 bpm with a standard deviation of 4 bpm. After removing the outliers, the resulting parameters are called optimized exercise monitoring parameters. These parameters should more accurately reflect the user's true physiological state and provide a foundation for subsequent exercise monitoring analysis. The optimized parameters are stored for future use. A report can be generated that includes the user's initial state monitoring parameters, the outliers, and the optimized results after removal. The optimized heart rate is recorded as 71 bpm, blood oxygen saturation is 96%, and body temperature is 36.8°C.These optimized parameters will be used for subsequent analysis of user exercise status, providing a data basis for personalized exercise recommendations and health management.

[0039] Step S2: performing multimodal state fusion perception based on the optimized motion monitoring parameters to obtain a real-time state perception map of the user;

[0040] In this embodiment, the goal of multimodal state fusion perception is first defined: to generate a real-time state perception map of the user by integrating data from various physiological sensors. This perception map is intended to comprehensively reflect the user's physiological state, exercise intensity, and fitness level. The parameters to be fused are determined, including optimized heart rate, blood oxygen saturation, body temperature, acceleration, gait, and other data, to form a comprehensive overview of the user's state. After determining the fusion parameters, the collection frequency and importance of each parameter are listed. Heart rate and blood oxygen saturation are collected once per second, body temperature is collected once per minute, and acceleration and gait data are updated once per second. Heart rate and blood oxygen saturation are given higher weights, while body temperature is given lower weights. To achieve effective multimodal data fusion, data from different sources must first be preprocessed. This includes data cleaning, denoising, and normalization steps to ensure that each parameter is compared on the same scale. Cleaning thresholds are set to, for example, remove unreasonable values ​​(such as heart rate exceeding 200 bpm or body temperature outside the 35-38°C range). Each parameter is cleaned and normalized. If the heart rate data contains an outlier value of 110 bpm (identified as abnormal), it is discarded. During the normalization process, heart rate, blood oxygen saturation, and body temperature data are converted to values ​​between 0 and 1: x^' = (x - min(X)) / (max(X) - min(X)). This normalization ensures that all parameters are in the same dimension, laying the foundation for subsequent fusion operations. Based on the fusion goal, select an appropriate multimodal data fusion algorithm. Methods such as weighted averaging, principal component analysis (PCA), or deep learning models (such as long short-term memory networks (LSTMs)) can be used. In this solution, the weighted averaging method is chosen because it is simple and easy to implement, making it particularly suitable for processing real-time data streams. Weights are assigned to each parameter, adjusting them according to their importance. The weight for heart rate is set to 0.4, for blood oxygen saturation to 0.3, for body temperature to 0.2, and for acceleration to 0.1. Based on the standardized data, calculate the total sensory value: Total sensory value = 0.4 × heart rate + 0.3 × blood oxygen saturation + 0.2 × body temperature + 0.1 × acceleration. Use visualization tools to generate a real-time sensory map of the user's status. Charts, dashboards, and other formats can be used to display the user's current physiological state and health level. Configure the sensory map's display content, including the real-time values ​​of each parameter, the fused total sensory value, and health status indicators. Use visualization software such as Tableau or Power BI to construct the user's sensory map. For example, a heart rate of 75 bpm, blood oxygen saturation of 96%, body temperature of 36.8°C, and acceleration of 1.5 m / s yields a fused total sensory value of 0.78. Different colors can be used in the sensory map to indicate health status, such as green for normal, yellow for warning, and red for high risk. If the user's heart rate is abnormally elevated, the corresponding area in the map will be marked in red to alert the user.To ensure that the user's state perception map reflects physiological changes in real time, a real-time monitoring and feedback mechanism must be established. When a user's state changes significantly, the system should automatically update the state perception map and provide corresponding health recommendations. A feedback mechanism can be implemented. For example, if the heart rate exceeds 120 bpm, the device will immediately send a reminder to the user, "Attention, your heart rate is high." During exercise, the system monitors various physiological parameters in real time and continuously updates the state perception map based on preset conditions. If the user's heart rate rises from 75 bpm to 130 bpm, the system will automatically recalculate the total perception value and update the chart. Through this real-time monitoring and feedback mechanism, users can obtain timely information on their health status, promoting more scientific exercise management.

[0041] Step S3: Acquire the positioning information of the smart wearable device in real time, analyze the behavioral pattern characteristics and dynamic behavior trajectory evolution, and generate a real-time user behavior trajectory map;

[0042] In this embodiment, appropriate positioning technologies, such as GPS, Wi-Fi, or Bluetooth, are selected to ensure that the smart wearable device can obtain the user's location information in real time. GPS is the most commonly used positioning method and is suitable for outdoor sports. The data collection frequency is set to 1Hz to ensure that the user's location information is updated every second, thereby capturing the user's dynamic changes. When the user exercises, the smart wearable device begins recording the user's latitude and longitude information and timestamp in real time. For example, if the user is running in a park, the device records positioning information once per second. The device should also record the user's movement status, such as speed and cadence, for subsequent analysis. If the user's speed is 5 km / h at a certain moment, the system will record this speed information accordingly. After obtaining the user's positioning information, behavioral pattern characteristics are analyzed. Parameters such as acceleration, speed change, and frequency can be used to identify the user's movement behavior patterns, such as walking, running, and stationary. A threshold for identifying behavioral patterns is set. For example, a speed greater than 4 km / h is marked as "running," a speed between 1 and 4 km / h is marked as "walking," and a speed of 0 is marked as "stationary." By analyzing the user's speed data, the system determines the user's current behavior in real time. If the user's speed data is [0, 1.2, 3.0, 5.5, 0] km / h within a certain time period, the system can identify the user's behavior pattern as varying between "stationary," "walking," "running," and "stationary." The identified behavior pattern is recorded in a database, forming user behavior pattern feature data for subsequent analysis and display. Based on real-time positioning information and behavior pattern features, the system evolves the dynamic behavior trajectory. Algorithms (such as Kalman filters) are used to smooth the trajectory data and predict the user's future location. A time window for trajectory evolution is determined, for example, updating the user's trajectory every 5 minutes. Within each time window, the user's trajectory is updated based on the user's current positioning information and behavior pattern. If the user's positioning information at time t=1 is (40.7128, -74.0060) and the speed is 5 km / h, the user's new location at time t=2 is calculated to be (40.7129, -74.0058). By updating the user's trajectory in real time, a continuous motion trajectory is formed. Assuming a user completes a run within 30 minutes, the trajectory will show the user's starting point, end point, and intermediate path. Use visualization tools to generate a real-time behavioral trajectory map of the user, showing the user's path changes, behavior patterns, and exercise intensity during exercise. Set the content of the graphical display, including the user's movement trajectory, current speed, behavior pattern, and use different colors to distinguish different movement states. In the trajectory map, draw the user's movement trajectory line and mark the user's starting point and end point. If the user reaches the end point B from the starting point A, the trajectory map will show the user's complete path, and use different colors on the path to represent the "walking", "running" and "stationary" states. Update the graphics in real time so that the user can see the current movement status and trajectory.The system will automatically adjust the chart with each update to ensure that users always see the latest exercise information.

[0043] Step S4: predicting the fatigue sign limit point and calculating the optimal rest time point of the user's real-time state perception map based on the user's real-time behavior trajectory map, and generating a rest warning signal;

[0044] In this embodiment, a user's real-time behavioral trajectory is used to analyze changes in their exercise patterns and physiological parameters to identify signs of fatigue. By observing factors such as exercise intensity, duration, and frequency, it is determined whether the user is approaching a state of fatigue. Fatigue-sign identification criteria are set, such as a heart rate exceeding 85% of the user's maximum heart rate or an exercise duration exceeding a specific threshold (e.g., 30 minutes). During exercise, the user's heart rate, speed, and activity time are monitored in real time. If the user's heart rate remains at 150 bpm (assuming a maximum heart rate of 180 bpm) and the exercise duration is 35 minutes, a preliminary determination of fatigue can be made. The system will mark these key moments on the trajectory for subsequent analysis and decision-making. This data will constitute the basis for understanding the user's fatigue status. Time series analysis methods are used to predict signs of fatigue. A sliding window method can be used to calculate the user's fatigue index over different time periods, combined with physiological data (such as heart rate and speed) for evaluation. The user's fatigue index is calculated within each time window. For example, if a user's heart rate is 150 bpm at a certain moment and they exercise for 35 minutes, the fatigue index for each time period is recorded and analyzed to identify fatigue thresholds. When the fatigue index exceeds a certain threshold (e.g., 30), it is marked as a fatigue threshold. Based on the fatigue threshold, the optimal rest time is calculated for the user. An empirical formula based on the fatigue index and exercise duration can be set to determine the appropriate rest time. After identifying multiple fatigue thresholds, the system calculates the corresponding optimal rest time based on the fatigue index. If the fatigue index for a certain time period is 35 and the time window is 5 minutes, the optimal rest time is: Optimal Rest Time = 35 × 5 = 175 minutes. The system will output a recommended rest time. If a 175-minute rest is recommended, a corresponding rest notification will be displayed on the user's device to help them manage their exercise load. Based on the calculated optimal rest time, a rest warning signal can be designed. When the user approaches the fatigue threshold, the system will automatically send a rest reminder to help them plan their rest time appropriately. Alert trigger conditions can be set, for example, when the fatigue index exceeds 25, the system will issue a rest reminder. While exercising, the system monitors the user's fatigue index in real time. Once the index exceeds a set threshold, the device will alert the user through vibration, sound, or mobile phone notifications. When the user's fatigue index reaches 30, the system will pop up a prompt: "Recommend rest for 30 minutes and monitor your heart rate." This real-time monitoring and feedback mechanism ensures that users can receive timely health advice, adjust their exercise plans appropriately, and avoid excessive fatigue.

[0045] Step S5: Obtain historical motion monitoring parameters, and perform transient state mutation identification and state offset calculation for each state point on the user's real-time state perception graph to obtain multiple state point offset values;

[0046] In this embodiment, historical exercise monitoring parameters are extracted from the smart wearable device. These parameters include the user's heart rate, number of steps, exercise type, duration, acceleration, etc. This ensures that these data are accurately recorded and stored during the user's exercise.

[0047] Set the data collection frequency to once per second to obtain enough detailed information to ensure that the recorded data reflects the user's actual exercise status. The heart rate data for the past week may be displayed as [70, 72, 75, 80, 85, 78, 74] bpm.

[0048] After the user completes their workout, the device uploads historical data to the cloud or local storage. After each workout, the system automatically saves the data, creating a structured database. A user's weekly exercise history includes information such as heart rate, steps, and acceleration over time.

[0049] These historical parameters are collated to facilitate subsequent analysis and comparison, providing benchmark data for identifying sudden changes in state. Using the user's historical motion monitoring parameters, the real-time state perception graph is used to identify transient sudden changes in state. A transient sudden change in state refers to a significant change in a user's physiological parameters within a short period of time.

[0050] A sudden change threshold is determined. A change in heart rate exceeding 10 bpm or a change in step count exceeding 100 steps is considered a sudden change. During exercise, the system monitors changes in physiological parameters in real time. If the heart rate data shows a sharp increase from 75 bpm to 90 bpm, and this occurs within a short period of time (e.g., 5 seconds), it is marked as a transient state sudden change.

[0051] Collect the timestamps and corresponding parameter values ​​of these sudden state points to form a sudden state list. If the heart rate is 90 bpm at t = 10 seconds and 85 bpm at t = 11 seconds, record t = 10 seconds as the sudden state point. For each identified transient state sudden point, calculate its state offset value. The state offset value refers to the deviation of the physiological parameter at a certain moment from its historical normal state.

[0052] Extract each sudden change state point one by one and calculate its corresponding offset value. If the heart rate at a sudden change state point is 90 bpm, while the baseline heart rate in the historical data is 75 bpm, the state offset value is: State offset value = 90 - 75 = 15 bpm. Record the offset values ​​of all sudden change state points in the dataset to form a state offset value list for subsequent analysis. If the offset values ​​of five sudden change state points are recorded as [15, 10, 20, -5, 12] bpm, these data will be used for subsequent state analysis. Further analysis of the calculated multiple state point offset values ​​can be performed to identify physiological trends. The mean, standard deviation, and rate of change of these offset values ​​can be calculated to assess the user's state fluctuations.

[0053] Set the analysis period to the past 30 minutes to observe changes in status over a short period of time. Statistically calculate the mean and standard deviation of the status offsets. If the offsets are [15, 10, 20, -5, 12] bpm, the average offset is: Average offset = (15 + 10 + 20 - 5 + 12) / 5 = 10.4 bpm. Based on these analysis results, a user status change report is generated, recording physiological changes in each time period to help users understand their exercise status and health risks.

[0054] Step S6: Quantitatively assess the user status risk of multiple status point offset values, make intelligent health decisions, and build an intelligent risk handling strategy.

[0055] In this embodiment, risk assessment indicators are set based on multiple state point offset values. These indicators should reflect changes in the user's physiological state and potential health risks. Common risk assessment indicators include heart rate offset, exercise intensity, and fatigue index. Risk level standards are determined, and state offset values ​​are categorized as "low risk" (offset value ≤ 5 bpm), "medium risk" (5 < offset value ≤ 15 bpm), and "high risk" (offset value > 15 bpm). Offset data for all state points is collected, assuming the state offset values ​​are [15, 10, 20, -5, 12] bpm. The corresponding risk level is calculated for each offset value. Offset values ​​of 15 bpm and 20 bpm are marked as high risk, 10 bpm as medium risk, and -5 bpm as low risk. Statistical analysis is used to calculate the number of state points at different risk levels. In the above data, there are two high-risk points, one medium-risk point, and one low-risk point. A risk assessment report is generated based on these results to clarify the user's health status. Based on the risk assessment results, intelligent health decisions are made. A rule-based decision model can be used to generate personalized health recommendations based on the user's historical health data, exercise history, and risk level. Decision criteria are determined. If the user is at high risk, the recommendation is to stop exercising and rest immediately; if the user is at medium risk, the recommendation is to reduce exercise intensity; and if the user is at low risk, the recommendation is to continue the current activity. Health recommendations are generated for different risk levels. If the user's deviation value is 20 bpm (high risk), the system will automatically prompt: "Your heart rate is elevated. Please stop exercising immediately and rest." If it is 10 bpm (medium risk), the system will prompt: "Reducing exercise intensity and monitoring heart rate is recommended." These recommendations are provided to the user through smart wearable devices via notifications, vibrations, or sounds, ensuring timely access to health information. Based on the user's risk assessment and health decisions, an intelligent risk management strategy is constructed. This strategy aims to help the user take appropriate measures based on different health conditions to reduce potential risks and improve health. A strategy implementation process is established, combining exercise plans with physiological feedback to develop personalized exercise and rest plans. In high-risk situations, strategies may include immediate rest, heart rate monitoring, and recommendations for deep breathing or relaxation exercises. In medium-risk situations, strategies may include adjusting exercise intensity and increasing rest frequency. The system will dynamically adjust strategies based on the user's real-time status. If the user fails to reduce exercise intensity in a medium-risk situation, the system will update the strategy, reminding the user to take more rest, ensuring more adaptive health management. To verify the effectiveness of intelligent risk management strategies, effectiveness evaluation indicators need to be established. These indicators should include changes in the user's physiological state (such as heart rate and fatigue index) as well as user feedback (such as exercise satisfaction). After implementing the intelligent risk management strategy, the user's physiological parameters are continuously monitored. If the user's heart rate drops from 150 bpm to 120 bpm after resting in a high-risk situation, the strategy is effective.Collect user feedback on the strategy, such as through questionnaires or user interface feedback, to assess user satisfaction with the health recommendations. Positive user feedback indicates successful strategy implementation; otherwise, further optimization is needed.

[0056] In this embodiment, refer to Figure 2 , is a flowchart of the detailed implementation steps of step S1. In this embodiment, the detailed implementation steps of step S1 include:

[0057] Step S11: collecting user initial state monitoring parameters based on the smart wearable device control chip; calculating the user's heart rate and heart rate change frequency based on the user's initial state monitoring parameters;

[0058] Step S12: Analyzing the user's exercise state based on the user's heart rate and heart rate change frequency to obtain the user's real-time exercise state;

[0059] Step S13: Adaptively adjusting the acquisition frequency of the control chip based on the user's real-time motion state to obtain an adaptive acquisition frequency;

[0060] Step S14: sampling multimodal motion monitoring parameters based on the adaptive acquisition frequency to extract the user's real-time multimodal motion monitoring parameters;

[0061] Step S15: performing time stamp synchronization processing on the user's real-time multimodal motion monitoring parameters to obtain time-synchronized motion monitoring parameters;

[0062] Step S16: performing abnormal data identification on the time-synchronized motion monitoring parameters and performing abnormality elimination processing to obtain optimized motion monitoring parameters.

[0063] In this embodiment, after obtaining device and user authorization, the device's initial parameter collection frequency is set to 100Hz to ensure sufficient dynamic information is captured. Initial state monitoring parameters include heart rate, heart rate variability, gait, and exercise intensity. The device's photoelectric sensor records the user's heart rate data, calculates the user's average heart rate to be 75 beats per minute (bpm), and monitors heart rate fluctuations. Using this collected heart rate data, the user's real-time heart rate and its frequency of variation are calculated. The frequency of heart rate variation is calculated by detecting the peak interval of the heartbeat signal. A detection threshold is set for the heartbeat signal waveform, and the time interval between heartbeat peaks is recorded. If the average value of consecutive heartbeat intervals is 0.8 seconds, the heart rate is: HeartRate = 60 / AverageInterval = 60 / 0.8 = 75 bpm. Further analysis of the frequency of heart rate variation is performed, and the standard deviation of the heart rate in a static state is calculated to assess heart rate stability. A smaller standard deviation indicates less heart rate fluctuation, while a smaller standard deviation indicates greater variability. If the heart rate recorded within a minute is 70, 75, 80, 72, and 78 bpm, calculate the standard deviation of these values ​​to understand the user's heart rate variability. Based on the acquired heart rate data and its frequency of variation, establish a user exercise state analysis model. This model can be combined with machine learning algorithms to analyze the user's exercise intensity (such as sedentary, light, moderate, and vigorous). Exercise state thresholds are set: a heart rate ≤ 70 bpm is considered sedentary, 71-85 bpm is considered light, 86-100 bpm is considered moderate, and ≥ 101 bpm is considered vigorous. The model evaluates real-time heart rate data to determine the user's current exercise state. If the current heart rate is 82 bpm, the user is considered to be in a light exercise state. This method continuously monitors changes in the user's exercise state for subsequent adaptive frequency adjustment. An adaptive acquisition frequency adjustment strategy is set based on the user's real-time exercise state. When the user is sedentary, the acquisition frequency can be reduced; when the user's exercise state increases, the acquisition frequency can be increased. The acquisition frequency is set to 50Hz in a static state, 75Hz for light exercise, 100Hz for moderate exercise, and 150Hz for high-intensity exercise. The sensor's acquisition frequency is dynamically adjusted during real-time monitoring. By evaluating the user's motion state in real time, the data acquisition settings are promptly responded to and adjusted. If the user transitions from a static state to light exercise, the control chip automatically increases the acquisition frequency to 75Hz to ensure that more motion details are captured. Based on the adaptive acquisition frequency, a sampling scheme for multimodal motion monitoring parameters is designed. These parameters include heart rate, gait, acceleration, angular velocity, etc., to ensure that the user's motion state is fully reflected. At an acquisition frequency of 100Hz, the sampling interval for each parameter is set to 10ms to ensure data integrity and accuracy. At the set acquisition frequency, multimodal motion monitoring parameters are collected in real time and the data is stored in the control chip for subsequent processing and analysis.During dynamic exercise, heart rate and acceleration data are recorded simultaneously to form a multimodal dataset with timestamps. Timestamp synchronization of multimodal data requires ensuring that data collected by different sensors accurately match. Linear interpolation or time alignment algorithms can be used for synchronization. The heart rate data is set as the primary timestamp, and the data from other sensors are adjusted based on time delays to align with the heart rate data. Timestamps are synchronized for all sampled motion monitoring parameters to ensure that all relevant data is acquired at the same time point for subsequent analysis. For example, if the heart rate data is timestamped at 0.010s and the acceleration data is collected at 0.012s, the acceleration data is time-shifted by 0.002s to align with the heart rate data. An algorithm for identifying abnormal data is typically developed, using statistical analysis methods (such as the Z-score or IQR) to identify and eliminate data anomalies. A reasonable threshold is set to determine data validity. The Z-score threshold is set to ±3; data points with a Z-score outside this range are marked as abnormal. After time synchronization, abnormal data is identified for motion monitoring parameters and unreasonable data points are removed to ensure the accuracy and reliability of the final dataset. If a data point in the gait data is found to have a speed of -5 m / s (an unreasonable value), it is marked as an anomaly and removed from the dataset.

[0064] In this embodiment, refer to Figure 3 , is a flowchart of the detailed implementation steps of step S2. In this embodiment, the detailed implementation steps of step S2 include:

[0065] Step S21: Calculating the user's heart rate, respiratory rate, body temperature monitoring data and electrocardiogram signal based on the optimized motion monitoring parameters;

[0066] Step S22: performing a heart rate time difference fluctuation analysis on the heart rate to extract a heart rate time difference fluctuation curve;

[0067] Step S23: performing time series variation statistics on the respiratory frequency to obtain respiratory frequency variation characteristics;

[0068] Step S24: performing multimodal state fusion perception on the body temperature monitoring data, electrocardiogram signal, heart rate time difference fluctuation curve and respiratory rate change characteristics to obtain a real-time state perception map of the user.

[0069] In this embodiment, optimized motion monitoring parameters are extracted from a smart wearable device. These parameters include heart rate, respiratory rate, body temperature monitoring data, and electrocardiogram (ECG) signals. The device's sensors are ensured to accurately capture these physiological data. The heart rate acquisition frequency is set to 100 Hz, the respiratory rate acquisition frequency to 50 Hz, the body temperature monitoring frequency to 10 Hz, and the ECG signal acquisition frequency to 500 Hz. These settings ensure the real-time and accuracy of the data. The above physiological parameters are collected in real time while the user performs routine activities. Heart rate can be monitored using a photoelectric sensor, respiratory rate can be obtained using a chest motion sensor, body temperature can be measured using a thermocouple sensor, and ECG signals can be recorded using electrode patches. During a 30-minute light exercise session, the user's heart rate is recorded to be 72-120 bpm, respiratory rate to be 16-24 beats / minute, and body temperature to be maintained in the range of 36.5-37.5°C. ECG signal data is continuously recorded to form a complete data set. Time difference fluctuation analysis is performed on the heart rate data to extract a heart rate time difference fluctuation curve. A sliding window method can be used to calculate the instantaneous rate of change of heart rate, with a window size of 5 seconds. By calculating the differences between consecutive heart rate data points, a fluctuation curve is generated. This method clearly reflects the changing trends of heart rate over different time periods. Process the heart rate data, calculate the average heart rate within each time window, and record the heart rate fluctuation for each frame. If the heart rate data within 5 seconds is [75, 78, 80, 76, 77], the average heart rate for this window is 77 bpm, with a fluctuation of ±3 bpm. By extracting these time differences, a heart rate time difference fluctuation curve is generated, which can intuitively demonstrate the patterns and characteristics of heart rate fluctuations. Time-series statistics of respiratory rate changes are analyzed to observe their characteristics and trends. Statistical analysis is performed by recording respiratory rate at different time points, with a time period of 1 minute. Sample the respiratory rate data, calculate the average respiratory rate per minute, and record its fluctuations. Calculate the change in respiratory rate within each minute. If the respiratory rate data for a certain time period is [16, 17, 18, 16, 19], the average respiratory rate for that minute is 17 bpm. The maximum and minimum values ​​are recorded for subsequent analysis. This statistical method generates a time series of respiratory rate variation characteristics, providing a data foundation for subsequent multimodal fusion analysis. A multimodal state fusion perception model is designed to comprehensively analyze body temperature monitoring data, electrocardiogram (ECG) signals, heart rate time difference fluctuation curves, and respiratory rate variation characteristics. Data fusion can be achieved using weighted averaging or principal component analysis (PCA). Weights are assigned to each parameter based on their importance in monitoring the user's health status. The weights for heart rate and respiratory rate are set to 0.4, and for body temperature and ECG signals to 0.3. These parameters are integrated according to the set weights to generate a real-time state perception map of the user. The fusion results are displayed visually to facilitate health status assessment by users and medical staff.If the fusion result is a health status score of 85 / 100, it can indicate that the user's current physiological state is good. Different colors and graphics can be used in the real-time status perception map to represent different physiological parameter states, helping users quickly identify potential health risks.

[0070] In this embodiment, refer to Figure 4 , is a flowchart of the detailed implementation steps of step S3. In this embodiment, the detailed implementation steps of step S3 include:

[0071] Step S31: Acquire the positioning information of the smart wearable device in real time; calculate the real-time spatial position coordinates based on the positioning information;

[0072] Step S32: performing real-time position tracking on the real-time spatial position coordinates to generate a real-time user position coordinate sequence;

[0073] Step S33: Calculating the user's movement speed, acceleration, and direction change frequency based on the real-time user position coordinate sequence;

[0074] Step S34: Analyzing the behavior pattern characteristics based on the user's movement speed, acceleration, and direction change frequency to generate user behavior pattern characteristics;

[0075] Step S35: Dynamically evolve the behavior trajectory according to the user's behavior pattern characteristics to generate a real-time user behavior trajectory graph.

[0076] In this embodiment, appropriate positioning technologies (such as GPS, Wi-Fi, and Bluetooth) are selected, ensuring that the smart wearable device is equipped with the appropriate sensors. These sensors can acquire the user's location information in real time, ensuring accurate and timely positioning. The GPS positioning update frequency is set to 1Hz (update once per second) to facilitate acquisition of the user's location information in dynamic environments. Real-time positioning information is acquired during user activity. This can be accomplished by enabling an automatic positioning function on the user's wearable device to continuously record location data during exercise. Recorded data should include longitude, latitude, and a timestamp. For example, if the user is in a park, the device records positioning information once per second. The acquired latitude and longitude information is converted to planar coordinates (such as metric coordinates) using a geographic information system (GIS) or a corresponding algorithm. This can be achieved by using the Haversine formula or other geographic coordinate conversion algorithms. The conversion accuracy is set to 1 meter to ensure that the generated spatial coordinates reflect the user's actual location in three-dimensional space. The real-time positioning information is converted to spatial coordinates. Assume that at a certain point in time, the user's latitude and longitude are (40.7128, -74.0060). Using a conversion algorithm, the corresponding spatial coordinates are calculated as (x, y, z), where z can be set to 0 (assuming movement on a plane). If the converted spatial coordinates are (500, 600, 0), these form the user's real-time spatial location coordinates. Using these real-time spatial coordinates, a user location tracking system is established. This system continuously records the user's location information and generates a sequence of location coordinates. The location tracking interval is set to 1 second to ensure real-time and accurate data. As the user moves, the spatial coordinates are recorded at each point in time. This can be implemented using a loop structure or data stream processing, ensuring that the user's new location is recorded every second. The generated user location coordinate sequence is [(500, 600, 0), (501, 601, 0), (502, 602, 0), …], representing the user's location changes at different points in time. Based on the real-time location coordinate sequence, the user's movement velocity and acceleration are calculated. The movement speed can be calculated by the ratio of the position change to the time change. The formula is: Speed ​​= Current position - Previous position / Δt, where Δt is the time interval (for example, 1 second). Traverse the user position coordinate sequence and gradually calculate the speed and acceleration between each two points. If the coordinates at time points t1 and t2 are (500, 600) and (501, 601) respectively, then the speed is: Speed ​​= Acceleration is calculated similarly, using velocity changes. The frequency of directional changes can be identified by the angular change between consecutive locations. The calculated velocity, acceleration, and directional change frequency are used to analyze user behavior patterns. Clustering algorithms (such as K-means or DBSCAN) can be used to analyze user behavior patterns and identify common motion states (such as walking, running, and stationary). A time window for feature aggregation, for example, analysis every 5 minutes, is set to extract stable behavioral features. User movement data is fed into the clustering algorithm to identify different behavioral patterns. If the user's speed is greater than 3 meters per second over a certain period, it can be labeled as "running"; if the speed is less than 1 meter per second, it can be labeled as "stationary." A user behavior pattern feature dataset is generated, recording the duration and frequency of each behavior pattern. Based on the user behavior pattern features, a dynamic behavior trajectory evolution model is designed. This model predicts future movement trajectories based on the user's current behavior state. Trajectory prediction and updates can be achieved using a Kalman filter or state-space model. A real-time user trajectory map is generated based on the real-time user location sequence and behavior pattern features. Use visualization tools (such as graphics libraries or data visualization platforms) to display behavioral trajectories and ensure that the trajectories are clearly visible. Dynamically draw the user's movement trajectory on the map, showing each change in the user's movement status, and use different colors to distinguish between walking, running, and stationary states.

[0077] In this embodiment, step S4 includes the following steps:

[0078] Perform motion load calculation on the user's real-time state perception map to obtain the user's motion load state characteristics;

[0079] Conduct time-series estimation of physical energy consumption based on the user's exercise load status characteristics and construct a time-series physical energy consumption curve;

[0080] Based on the user's real-time behavior trajectory, the time series physical energy consumption curve is deeply mined to obtain the user's sports fatigue status information;

[0081] Predicting the fatigue sign limit point based on the user's exercise fatigue status information to obtain the fatigue sign limit point;

[0082] Calculate the optimal rest time point based on the fatigue sign limit point and mark the optimal rest time point;

[0083] A rest warning signal is generated based on the optimal rest time point.

[0084] In this embodiment, the user's exercise load state characteristics are calculated based on their real-time state perception map. Exercise load typically takes into account multiple factors such as heart rate, acceleration, and exercise type. Exercise load can be assessed by combining heart rate with MET (metabolic equivalent) values. A MET value range is set, and calculations are performed based on the MET values ​​corresponding to different exercise types (such as walking, running, and cycling). The MET value for walking is 3, and the MET value for running is 8. As the user exercises, heart rate, acceleration, and exercise type data are extracted in real time. Exercise load is calculated using the formula: Exercise load = MET × weight (kg) × activity time (h). For example, if a user weighs 70 kg and runs for one hour with a heart rate of 150 bpm and a MET value of 8, the exercise load is: Exercise load = 8 × 70 × 1 = 560 kcal. Based on the user's exercise load state characteristics, a time series estimation of physical energy expenditure is performed. A linear regression model or time series analysis method can be used to predict future physical energy expenditure based on historical exercise load data. A time window is set, for example, to estimate physical energy expenditure every 5 minutes, to ensure data timeliness. Within each time window, the user's exercise load is recorded and estimated to generate an energy expenditure curve. If the exercise load within each 5-minute window is [100, 120, 140, 160] kcal, the corresponding energy expenditure can be estimated. Using this data, a time-series energy expenditure curve is constructed, which visually displays the user's energy expenditure changes over different time periods. Combined with the user's real-time behavioral trajectory, the time-series energy expenditure curve can be used to deeply analyze exercise fatigue. Fatigue indices (such as RPE (Rate of Perceived Exertion)) can be combined with physiological data for comprehensive analysis. A fatigue threshold is set; for example, an RPE value exceeding 7 is considered high fatigue. The energy expenditure curve is combined with the user's behavioral trajectory to analyze the user's fatigue symptoms under different exercise states. If the user experiences a decrease in speed or slow heart rate recovery during exercise, this indicates increasing fatigue. This information is collected to form an information profile of the user's exercise fatigue status, facilitating subsequent fatigue threshold prediction. Based on the user's exercise fatigue status information, a predictive model (such as time series prediction) is used to identify the fatigue threshold. A fatigue index can be set and combined with historical data to predict the threshold. The fatigue index is calculated using the following formula: Fatigue Index = Exercise Load / Heart Rate Recovery Time. The fatigue index is calculated for each time period and its changes are monitored. If the fatigue index exceeds the set limit, it is marked as a fatigue threshold. If the fatigue index reaches 5 during a particular exercise session, exceeding the set threshold of 4, this point is marked as a fatigue threshold for subsequent rest time calculations. The optimal rest time is calculated based on the fatigue threshold. Empirical formulas or physiological models can be used to determine the appropriate rest time, combining the user's exercise load and fatigue index.The recommended formula for setting rest times is: Recommended Rest Time = Fatigue Index × Exercise Duration (minutes). The optimal rest time after each exercise session is calculated based on the user's exercise history and fatigue thresholds. For example, if the user's Fatigue Index is 5 and their exercise duration is 30 minutes, the recommended rest time is 150 minutes. Based on this calculated optimal rest time, a rest warning signal is set. When the user approaches their fatigue threshold or their exercise workload is too high, the system automatically issues a rest reminder. A threshold is set such that when the user's Fatigue Index exceeds 3, the system issues a rest recommendation. During exercise, the Fatigue Index and exercise workload are monitored in real time. Once they approach the preset threshold, a rest warning signal is generated. This can be alerted via vibration, sound, or mobile notifications. If the user's Fatigue Index is detected to be 4, the device will vibrate and a "15-minute rest recommended" notification will pop up to help the user adjust their exercise plan appropriately.

[0085] In this embodiment, the specific steps of performing motion load calculation on the user's real-time state perception map to obtain the user's motion load state characteristics are:

[0086] Perform running scene analysis based on the user's real-time state perception map to obtain the current user's motion scene;

[0087] Identify the start time of exercise based on the user's real-time state perception map;

[0088] Calculate the duration of the exercise based on the start time of the exercise;

[0089] Get the weight information entered by the user;

[0090] The exercise load is calculated based on the weight information, the current user's exercise scene and the duration of the exercise to obtain the user's exercise load state characteristics.

[0091] In this embodiment, the user's real-time state perception map is used to analyze the user's current exercise scene. This process involves integrating multimodal data, including heart rate, acceleration, GPS location, and other information. Using this data, the system can identify the user's exercise environment (e.g., running, walking, cycling, etc.). A scene recognition algorithm, such as a decision tree or support vector machine (SVM), is developed to classify the user based on different motion characteristics. While the user is exercising, various physiological and motion data are collected in real time. For example, the heart rate is maintained at 140 bpm, the speed is 10 km / h, and the current location is a park. By analyzing this data, the system can determine the user's exercise scene. Assuming that the user's heart rate and speed combination meets the characteristics of the "running" exercise scene, the system will label the current user's exercise scene as "running." To identify the start time of the user's exercise, the system monitors the user's state changes. The start time of the exercise is when the user transitions from a stationary state (e.g., sitting or standing) to an active state (e.g., starting to run). A monitoring threshold is set such that when the heart rate rapidly rises from 60 bpm to above 80 bpm and the speed exceeds 1 km / h, the exercise is marked as started. The moment a user begins exercise, heart rate and speed changes are monitored in real time. When a heart rate reaches 80 bpm and a speed exceeds 1 km / h, this time is recorded as the exercise start time. If the user's heart rate is 60 bpm at t=0 and rises to 82 bpm at t=1, while the speed is 1.5 km / h, t=1 is marked as the exercise start time. Once the exercise start time is identified, the user's exercise duration is calculated. This can be done by taking the difference between the current time and the exercise start time. Setting the time unit to seconds or minutes provides a more intuitive display of the user's exercise duration. As the exercise continues, the system continuously updates the current time and compares it with the exercise start time. If the exercise start time is t=1 and the current time is t=10, the exercise duration is 10-1=9 seconds. If the user starts running at t=1 and the current time is t=300 seconds, the exercise duration is 299 seconds (approximately 5 minutes). When using smart wearable devices, users need to enter their basic personal information, including weight. Usually, during the initial setup of the device, users will be asked to enter this information to ensure the accuracy of subsequent calculations. The user interface (UI) can be designed to provide a simple and easy-to-use weight input function to ensure that users can easily update their weight information. The user's weight information is stored in the system, assuming that the user weighs 70 kg. The system can call this information before and after each exercise for subsequent exercise load calculation. The device automatically prompts the user to confirm the weight information before the user starts exercising to ensure the accuracy of the data. The exercise load is calculated based on the user's weight, the current exercise scene and the duration of the exercise. The exercise load can be calculated by combining the MET value (metabolic equivalent) with weight and activity time.Set MET values ​​for different exercise scenarios, for example, a MET value of 8 for running and a MET value of 3 for walking. Assuming the user's exercise scenario is "Running" and the exercise duration is 5 minutes, use the following formula to calculate the exercise load: Exercise load = MET × weight (kg) × activity time (hours). For example, if the user weighs 70 kg and the exercise duration is 5 minutes (i.e., 1 / 12 hour), the exercise load is: Exercise load = 8 × 70 × 5 / 60 = 46.67 kcal.

[0092] In this embodiment, step S5 includes the following steps:

[0093] Obtain historical motion monitoring parameters based on the smart wearable device control chip;

[0094] Performing user-personalized regular state statistical analysis on the historical motion monitoring parameters to construct a normalized baseline state curve;

[0095] Divide the user's real-time state perception graph into multiple time periods to generate multiple state perception windows;

[0096] Identify transient state mutations in multiple state perception windows and extract irregular transient mutation state points;

[0097] Based on the normalized reference state curve, the state offset of the irregular transient mutation state points is calculated one by one to obtain multiple state point offset values.

[0098] In this embodiment, the control chip of a smart wearable device acquires the user's historical exercise monitoring parameters. These parameters may include heart rate, step count, exercise type, duration, acceleration, GPS location, etc. The device should have data storage capabilities, allowing it to upload data to the cloud or local storage after the user's exercise. The data collection cycle is set to once per second to ensure sufficient detailed information is captured for subsequent analysis. As the user exercises, the device records relevant data in real time. Assuming the user has engaged in various types of exercise over the past week, the recorded parameters may include: heart rate (average heart rate of 120 bpm), step count (average daily step count of 8,000 steps), and exercise type (running, cycling, fitness, etc.). A personalized routine status statistical analysis is performed on the acquired historical exercise monitoring parameters. Descriptive statistical methods (such as mean, standard deviation, maximum, and minimum values) can be used to construct a normalized baseline status curve for the user. The analysis cycle is set to one week, and the average and fluctuation of each exercise parameter for the user during this period are calculated. For the user's heart rate data, the average heart rate over the past week is calculated to be 120 bpm with a standard deviation of 5 bpm. The same statistics were performed on the step count data, resulting in an average of 8,000 steps and a standard deviation of 1,200 steps. Based on these statistical values, a normalized baseline state curve was plotted, showing parameter changes under different motion states, providing a baseline reference for subsequent state monitoring. Based on the user's real-time state perception map, the timeline was divided into multiple state perception windows. Each window could represent a different motion state or static state. Each state perception window was set to 5 minutes in duration to capture changes in motion state during dynamic monitoring. Data was monitored and recorded in real time while the user was exercising. The monitored time series data was divided into multiple 5-minute windows. The first window was from t = 0 to t = 5 minutes, the second window was from t = 5 to t = 10 minutes, and so on. The system extracted and saved the corresponding state data within each window, including heart rate, step count, acceleration, and other data, forming a data set for multiple state perception windows. Within each state perception window, transient state mutation identification was performed. This was achieved by setting a threshold to determine significant state changes, typically using the standard deviation or absolute change. A threshold is set such that a sudden change in heart rate is marked when it exceeds 10 bpm or when the step count exceeds 100 steps within a time window. Heart rate and step count data are analyzed within each time window. The heart rate data for the first window is [115, 118, 120, 125, 130] bpm. If the heart rate suddenly increases to 140 bpm at a certain time, it is identified as a transient sudden change. All identified irregular transient sudden change state points are recorded to form a sudden change state point list for subsequent analysis. Based on the normalized baseline state curve, the state offset of each extracted irregular transient sudden change state point is calculated. The offset calculation is performed by comparing the difference between each state point and the normalized baseline value.The offset calculation formula is: Offset = Sudden change point value - Normalized baseline value. For each extracted sudden change point, calculate its offset from the normalized baseline curve. If the heart rate at a sudden change point is 140 bpm and the normalized baseline value is 120 bpm, the offset value is 20 bpm. Record the offset values ​​of all state points to form a list of state point offsets, providing basic data for subsequent analysis.

[0099] In this embodiment, the specific steps of step S6 are:

[0100] Extracting the abnormal timestamp of the irregular transient mutation state point;

[0101] Perform multi-period offset trend evolution on multiple state point offset values ​​according to the abnormal timestamp to generate a multi-period state offset trend;

[0102] Predicting the long-term deviation trend of the multi-period state deviation trend to generate a future state deviation prediction situation;

[0103] Conduct quantitative assessment of user status risk based on the future status deviation prediction trend and construct a user status risk assessment report;

[0104] Make intelligent health decisions based on user status risk assessment reports and build intelligent risk handling strategies.

[0105] In this embodiment, abnormal timestamps are identified and extracted from previously extracted irregular transient mutation state points. Abnormal timestamps are those timestamps indicating significant state changes during the monitoring period. A detection threshold is set. When the magnitude of the state change (e.g., a sudden change in heart rate or step count) exceeds a preset standard (e.g., 10 bpm or 100 steps), the time point is considered an abnormal timestamp. For each identified sudden state point, the corresponding timestamp is recorded. For example, if during the monitoring process, the heart rate suddenly increases from 120 bpm to 150 bpm at a certain moment, the timestamp at that moment will be recorded as an abnormal timestamp. If the detected sudden state points are: [(t1, 140 bpm), (t2, 130 bpm), (t3, 155 bpm)], where t1, t2, and t3 are different time points, and the magnitude of the change in t3 exceeds the threshold, the system will extract the timestamp of t3 as an abnormal timestamp. Based on the extracted abnormal timestamps, the offset values ​​of multiple state points are analyzed to generate state offset trends for multiple time periods. Data can be divided into time periods, such as 5-minute or 10-minute periods, to facilitate observation of shift trends. Set the analysis period to the past 24 hours to ensure coverage of multiple changes in the user's motion state. Within each time period, calculate the shift values ​​for all state points within that period and find their average. If the shift values ​​recorded in a certain time period are [15, 20, 25], the average shift value for that period is 20 bpm. Use statistical methods to visualize the average shift values ​​for all time periods and generate a multi-period state shift trend chart to facilitate observation of patterns in state fluctuations. Use time series analysis to predict long-term shift trends for multiple state points. Use the ARIMA (Autoregressive Integrated Moving Average) model or exponential smoothing to predict future state shift trends. Set the forecast timeframe, such as the next hour or day, to assess changes in the user's motion state. Input the multi-period state shift trend data into the selected forecasting model to extrapolate future state shifts. If the historical shift values ​​are [5, 10, 15, 20], the model will use these data to predict future shift trends. Through model analysis, a future state deviation prediction trend map is generated, showing the difference between the predicted deviation value and the actual value, helping to identify potential risk changes. Based on the predicted future state deviation trend, the user's health status is quantitatively assessed. Risk assessment indicators can be set, such as the absolute value of the deviation value and the rate of change, and a risk assessment model can be constructed based on the user's exercise load and physiological data. Risk level classification criteria, such as low risk, medium risk, and high risk, are set to facilitate subsequent decision-making. For a predicted deviation value, if it exceeds a set threshold (such as 5 bpm), it is marked as a high-risk state. If the future deviation prediction value is 8 bpm, the system will generate a status risk assessment report for the user, indicating potential health risks.Based on the assessment results, a user status risk assessment report is generated, detailing the current status, future forecast, and risk level. Based on the user status risk assessment report, corresponding intelligent health decisions and risk management strategies are formulated. Machine learning algorithms can be used to analyze users' historical data and identify effective intervention measures. Decision criteria are set; for example, when the risk level is high, the user is advised to take a moderate rest or adjust exercise intensity. If the risk assessment report detects a high risk level for the user, the system will automatically generate health recommendations, such as "It is recommended to rest for 30 minutes and monitor heart rate." Through intelligent algorithms, risk management strategies are optimized to ensure that users receive timely health advice and support at the appropriate time. At the same time, actionable recommendations are provided, such as adjusting training plans and increasing rest time.

[0106] In this embodiment, a motion monitoring and analysis device based on a smart wearable device control chip is provided, which is used to execute the motion monitoring and analysis method based on a smart wearable device control chip as described above, including:

[0107] The data processing module is used to collect the user's initial state monitoring parameters, identify abnormal data, and perform abnormal elimination processing to obtain optimized motion monitoring parameters;

[0108] A state perception module, configured to perform multimodal state fusion perception based on the optimized motion monitoring parameters to obtain a real-time state perception map of the user;

[0109] The trajectory evolution module is used to obtain the positioning information of smart wearable devices in real time, analyze the characteristics of behavioral patterns and dynamically evolve behavioral trajectories, and generate a real-time user behavior trajectory map;

[0110] The fatigue sign prediction module is used to predict the fatigue sign limit point and calculate the optimal rest time point based on the user's real-time behavior trajectory map and the user's real-time state perception map, and generate a rest warning signal;

[0111] The state offset module is used to obtain historical motion monitoring parameters, identify transient state mutations in the user's real-time state perception map, and calculate the state offset of each state point to obtain multiple state point offset values;

[0112] The status risk analysis module is used to quantitatively assess user status risks based on multiple status point offset values, make intelligent health decisions, and build intelligent risk handling strategies.

[0113] This invention collects initial user status monitoring parameters and integrates an abnormal data identification mechanism to ensure the quality of monitoring data. By eliminating invalid, erroneous, or noisy data, the optimized motion monitoring parameters are more representative, providing a more reliable data source for subsequent analysis. After removing abnormal data, the system can more accurately capture the user's health dynamics. If data errors are caused by improper wear or sensor failure, the module automatically identifies and eliminates inaccurate monitoring data to prevent it from affecting subsequent status perception and decision-making. By fusing data from multiple sensors (such as acceleration, heart rate, location, and body temperature), the system provides a comprehensive assessment of the user's health status. Fusion perception not only eliminates bias in single sensor data but also provides more accurate real-time health monitoring. Multimodal state fusion enables the system to monitor and perceive user health parameters such as exercise intensity, fatigue, and physical exertion in real time, helping users fully understand their current exercise status and make timely adjustments. By obtaining positioning information from the smart wearable device, the module can create a real-time map of the user's behavior trajectory, reflecting their specific activity patterns and location changes. By analyzing behavioral pattern features, the system can identify the user's activity type, exercise intensity, and the evolution of their behavior patterns. As user behavior evolves, the system can adjust exercise recommendations or health guidance based on real-time patterns. If a user exhibits signs of excessive fatigue or poor posture, the system can recommend adjusting exercise patterns or taking a break. By analyzing user behavior in real time, the module can predict fatigue thresholds, promptly identifying whether a user has reached the threshold of overexertion or fatigue. This helps prevent injuries or health problems caused by excessive exercise. Based on the user's fatigue status, the module calculates optimal rest times and provides personalized rest recommendations, preventing fatigue accumulation and overexertion, and improving exercise effectiveness and safety. By comparing historical exercise data with real-time status graphs, the system can detect transient changes in the user's status (such as a sudden increase in heart rate or a sharp change in exercise intensity), which may indicate health risks or signs of excessive exercise. Promptly identifying these sudden changes helps prevent injury or increased health risks. By calculating offsets at each status point, the system can identify subtle changes in the user's health status and use these offsets to assess health risks, providing a reference for subsequent intelligent decision-making. By quantifying the offsets across multiple status points, the system can calculate the current health risk for each user and assess their health status in real time. The system can determine whether a user is at risk of overexertion, excessive fatigue, or other related issues, and can then intervene accordingly. Based on the risk assessment results, the system can provide users with personalized health decision-making recommendations. If the system detects excessive exercise intensity or a high risk of fatigue, it will automatically generate a health intervention strategy, recommending appropriate rest or adjustment of exercise intensity. This strategy can effectively prevent health problems and improve the effectiveness and safety of exercise.

[0114] The present invention is therefore intended to be illustrative and non-restrictive in all respects, with the scope of the invention being defined by the appended claims rather than the foregoing description, and all changes that come within the meaning and range of equivalents of the application documents are intended to be embraced therein.

[0115] The foregoing description is intended only to provide specific embodiments of the present invention, which will enable those skilled in the art to understand and implement the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not intended to be limited to the embodiments shown herein, but is to be construed in the widest possible manner consistent with the principles and novel features disclosed herein.

Claims

1. A motion monitoring and analysis method based on a smart wearable device control chip, characterized in that: The following steps are involved: Step S1: Collect the user's initial state monitoring parameters, identify abnormal data, and perform abnormal elimination processing to obtain optimized motion monitoring parameters; Step S2: performing multimodal state fusion perception based on the optimized motion monitoring parameters to obtain a real-time state perception map of the user; Step S3: Acquire the positioning information of the smart wearable device in real time, analyze the behavioral pattern characteristics and dynamic behavior trajectory evolution, and generate a real-time user behavior trajectory map; Step S4: predicting the fatigue sign limit point and calculating the optimal rest time point of the user's real-time state perception map based on the user's real-time behavior trajectory map, and generating a rest warning signal; Step S5: Obtain historical motion monitoring parameters, and perform transient state mutation identification and state offset calculation for each state point on the user's real-time state perception graph to obtain multiple state point offset values; Step S6: Quantitatively assess the user status risk of multiple status point offset values, make intelligent health decisions, and build an intelligent risk handling strategy; The specific steps of step S4 are: Perform motion load calculation on the user's real-time state perception map to obtain the user's motion load state characteristics; Conduct time-series estimation of physical energy consumption based on the user's exercise load status characteristics and construct a time-series physical energy consumption curve; Based on the user's real-time behavior trajectory, the time series physical energy consumption curve is deeply mined to obtain the user's sports fatigue status information; Predicting the fatigue sign limit point based on the user's exercise fatigue status information to obtain the fatigue sign limit point; Calculate the optimal rest time point based on the fatigue sign limit point and mark the optimal rest time point; generating a rest warning signal based on the optimal rest time point; The specific steps of performing motion load calculation on the user's real-time state perception map to obtain the user's motion load state characteristics are as follows: Perform running scene analysis based on the user's real-time state perception map to obtain the current user's motion scene; Identify the start time of exercise based on the user's real-time state perception map; Calculate the duration of the exercise based on the start time of the exercise; Get the weight information entered by the user; Calculate the user's exercise load based on the weight information, the current user's exercise scene, and the duration of the exercise to obtain the user's exercise load state characteristics; The specific steps of step S5 are: Obtain historical motion monitoring parameters based on the smart wearable device control chip; Performing user-personalized regular state statistical analysis on the historical motion monitoring parameters to construct a normalized baseline state curve; Divide the user's real-time state perception graph into multiple time periods to generate multiple state perception windows; Identify transient state mutations in multiple state perception windows and extract irregular transient mutation state points; Based on the normalized reference state curve, the state offset of the irregular transient mutation state point is calculated one by one to obtain multiple state point offset values; The specific steps of step S6 are: Extracting the abnormal timestamp of the irregular transient mutation state point; Perform multi-period offset trend evolution on multiple state point offset values ​​according to the abnormal timestamp to generate a multi-period state offset trend; Predicting the long-term deviation trend of the multi-period state deviation trend to generate a future state deviation prediction situation; Conduct quantitative assessment of user status risk based on the future status deviation prediction trend and construct a user status risk assessment report; Make intelligent health decisions based on user status risk assessment reports and build intelligent risk handling strategies.

2. The motion monitoring and analysis method based on the smart wearable device control chip according to claim 1 is characterized in that: The specific steps of step S1 are: Collecting user initial state monitoring parameters based on the smart wearable device control chip; calculating the user's heart rate and heart rate change frequency based on the user's initial state monitoring parameters; Analyze the user's exercise status based on the user's heart rate and heart rate change frequency to obtain the user's real-time exercise status; Adaptively adjusting the acquisition frequency of the control chip based on the user's real-time motion state to obtain an adaptive acquisition frequency; Multimodal motion monitoring parameters are sampled based on the adaptive acquisition frequency to extract the user's real-time multimodal motion monitoring parameters; Performing time stamp synchronization processing on the user's real-time multimodal motion monitoring parameters to obtain time-synchronized motion monitoring parameters; Abnormal data identification is performed on the time-synchronized motion monitoring parameters, and abnormality elimination is performed to obtain optimized motion monitoring parameters.

3. The motion monitoring and analysis method based on the smart wearable device control chip according to claim 1 is characterized in that: The specific steps of step S2 are: Calculate the user's heart rate, respiratory rate, body temperature monitoring data and electrocardiogram signal based on the optimized motion monitoring parameters; performing a heart rate time difference fluctuation analysis on the heart rate to extract a heart rate time difference fluctuation curve; Perform frequency time series variation statistics on respiratory frequency to obtain respiratory frequency variation characteristics; Multimodal state fusion perception is performed on the body temperature monitoring data, electrocardiogram signal, heart rate time difference fluctuation curve and respiratory rate change characteristics to obtain a user's real-time state perception map.

4. The motion monitoring and analysis method based on the smart wearable device control chip according to claim 1 is characterized in that: The specific steps of step S3 are: Obtain positioning information of smart wearable devices in real time; calculate real-time spatial position coordinates based on positioning information; Performing real-time position tracking on the real-time spatial position coordinates to generate a real-time user position coordinate sequence; Calculating user movement speed, acceleration, and direction change frequency based on the real-time user position coordinate sequence; Analyze the behavior pattern characteristics according to the user's movement speed, acceleration and direction change frequency to generate user behavior pattern characteristics; Dynamic behavior trajectory evolution is performed based on user behavior pattern characteristics to generate a real-time user behavior trajectory graph.

5. A motion monitoring and analysis device based on a smart wearable device control chip, characterized in that: The method for performing the motion monitoring and analysis method based on the smart wearable device control chip according to claim 1 comprises: The data processing module is used to collect the user's initial state monitoring parameters, identify abnormal data, and perform abnormal elimination processing to obtain optimized motion monitoring parameters; A state perception module, configured to perform multimodal state fusion perception based on the optimized motion monitoring parameters to obtain a real-time state perception map of the user; The trajectory evolution module is used to obtain the positioning information of smart wearable devices in real time, analyze the characteristics of behavioral patterns and dynamically evolve behavioral trajectories, and generate a real-time user behavior trajectory map; The fatigue sign prediction module is used to predict the fatigue sign limit point and calculate the optimal rest time point based on the user's real-time behavior trajectory map and the user's real-time state perception map, and generate a rest warning signal; The state offset module is used to obtain historical motion monitoring parameters, identify transient state mutations in the user's real-time state perception map, and calculate the state offset of each state point to obtain multiple state point offset values; The status risk analysis module is used to quantitatively assess user status risks based on multiple status point offset values, make intelligent health decisions, and build intelligent risk handling strategies.

Images