Household tumble alarm system based on mobile communication service

Abstract

The invention relates to the technical field of household tumble detection, and discloses a household tumble alarm system based on mobile communication service, which comprises a multi-dimensional acquisition module and an intelligent analysis module. According to the system, test data of daily behaviors of users, sensing data of activity behaviors and monitoring data of activity space are acquired through a multi-dimensional acquisition module, an intelligent analysis module analyzes an activity state under each posture, a daily data set and a feature data set are generated, individual differentiation of the users is emphasized, a unique health base line is established for each user, and the user experience is improved. The system analyzes the activity state of a user in real time, timely recognizes a tumble symptom, analyzes the potential safety hazard of each activity space, generates a danger index, is high in multi-dimensional evaluation precision, prevents the user from staying in a high-risk environment, evaluates the activity state of the user and the tumble risk of the activity space through an intelligent analysis module, judges whether to start a mobile communication service, and improves the safety of the user. And corresponding inquiry and emergency measures are executed, a hierarchical response mechanism is established, and the quick response sensitivity is high.

Description

Technical Field

[0001] The present invention relates to the technical field of home fall detection, and specifically provides a home fall warning system based on mobile communication services. Background Art

[0002] A home fall refers to an accidental fall event that occurs in a home environment, which is particularly common among the elderly and those with limited mobility. The reasons are diverse and mainly divided into two major factors: environmental and physical. Environmental factors include slippery floors, insufficient lighting, unreasonable furniture placement, lack of handrails on stairs, etc. Physical factors involve reduced vision, decreased balance ability, weakened muscle strength, and the impact of chronic diseases.

[0003] To effectively prevent home falls, fall warning systems have emerged. By continuously monitoring user behavior and environmental data, a fall warning system can issue an alarm immediately when a fall occurs and quickly notify the user or their emergency contact. This not only provides comprehensive safety protection for the elderly and those with limited mobility but also significantly reduces the risk of injury caused by falls, making it an essential tool in modern home care.

[0004] Currently, traditional home fall warning systems lack in-depth analysis of users' daily behavior patterns, often ignoring the individual differences of users, unable to establish a health baseline based on each user's behavior habits, with low sensitivity in abnormal monitoring, slow real-time alarm and emergency response speeds. Summary of the Invention

[0005] (I) Technical Problems to be Solved In view of the deficiencies of the prior art, the present invention provides a home fall warning system based on mobile communication services, which has the advantages of high accuracy in multi-dimensional evaluation and high sensitivity in rapid response, and solves the problems of low sensitivity in abnormal monitoring and slow real-time alarm and emergency response speeds of traditional home fall warning systems.

[0006] (II) Technical Solutions To achieve the above object, the present invention provides the following technical solutions: A home fall warning system based on mobile communication services, comprising a multi-dimensional acquisition module and an intelligent analysis module; The multi-dimensional acquisition module consists of a test data unit, a sensing data unit, and an environmental data unit. The test data unit collects a test data set through a network connection to a database. The test data set includes test data of users' daily behaviors. The sensing data unit collects a sensing data set through a network connection to a sensing device. The sensing data set includes sensing data of users' activity behaviors. The environmental data unit collects an environmental data set through a network connection to a camera. The environmental data set includes monitoring data of the activity space where the user is located; The intelligent analysis module consists of a daily evaluation unit, a real-time evaluation unit, an environment evaluation unit, and a fall warning unit. The daily evaluation unit analyzes the activity status of the user in each posture according to the test data set. Among them, a corresponding daily data set is generated for each posture. and a feature data set . The real-time evaluation unit analyzes the activity status of the user in real time according to the sensing data set and generates a corresponding monitoring data set. . The environment evaluation unit analyzes the potential safety hazards of each activity space according to the environment data set and generates a corresponding danger index. . The fall warning unit is set with a danger range of a fixed range. . Then, combined with the sensing data set, the daily data set , the feature data set , the monitoring data set , and the danger index , it evaluates the fall risk of the user's activity status and the activity space, determines whether to start the mobile communication service, and executes corresponding inquiries and emergency measures.

[0007] Preferably, the test data set includes several groups of standing posture test data, several groups of sitting posture test data, several groups of squatting posture test data, several groups of kneeling posture test data, and several groups of lying posture test data. And each group of test data includes heart rate, blood pressure, moving speed, tilt angle, and height from the ground.

[0008] Preferably, the expression of the sensing data set is , to represent the sensing data of the user's first to nth activity behaviors. The sensing data includes heart rate, blood pressure, moving speed, tilt angle, and height from the ground. s represents the time point for obtaining the sensing data of the user's activity behavior.

[0009] Preferably, the expression of the environment data set is , to represent the monitoring data of the first to mth activity spaces. The monitoring data includes environmental brightness, ground flatness, humidity, and storage density. j represents the time point for obtaining the monitoring data of the activity space where the user is located.

[0010] Preferably, the calculation process of the daily data set is as follows: According to the test data set, mark the heart rate in the standing posture test data as , to represent the heart rate during the first to a times of standing posture tests. Mark the heart rate in the sitting posture test data as , to represent the heart rates during the first to the bth sitting posture tests. Mark the heart rates in the squatting posture test data as , to represent the heart rates during the first to the cth squatting posture tests. Mark the heart rates in the kneeling posture test data as , to represent the heart rates during the first to the dth kneeling posture tests. Mark the heart rates in the lying posture test data as , to represent the heart rates during the first to the eth lying posture tests; ; In the formula, represents the average heart rate of the user during daily long - standing behavior, represents the average heart rate of the user during daily long - sitting behavior, represents the average heart rate of the user during daily long - squatting behavior, represents the average heart rate of the user during daily long - kneeling behavior, represents the average heart rate of the user during daily long - lying behavior, represents the active heart rates of the user in standing, sitting, squatting, kneeling, and lying postures, which are the daily data groups corresponding to the user's daily heart rate tests.

[0011] Preferably, the characteristic data group has the following calculation process: ; In the formula, represents the maximum heart rate of the user during daily long - standing behavior, represents the maximum heart rate of the user during daily long - sitting behavior, represents the maximum heart rate difference when the user changes from the standing posture to the sitting posture, represents the maximum heart rate of the user during daily long - squatting behavior, represents the maximum heart rate difference when the user changes from the standing posture to the squatting posture, represents the maximum heart rate of the user during daily long - kneeling behavior, represents the maximum heart rate difference when the user changes from the standing posture to the kneeling posture, represents the maximum heart rate of the user during daily long - lying behavior, represents the maximum heart rate difference when the user changes from the standing posture to the lying posture, represents the minimum heart rate of the user during daily long - standing behavior, represents the minimum heart rate of the user during daily long - sitting behavior, Represents the minimum heart rate difference when the user changes from a standing position to a sitting position, Represents the minimum heart rate during the user's daily long squat behavior, Represents the minimum heart rate difference when the user changes from a standing position to a squatting position, Represents the minimum heart rate during the user's daily long kneeling behavior, Represents the minimum heart rate difference when the user changes from a standing position to a kneeling position, Represents the minimum heart rate during the user's daily long lying behavior, Represents the minimum heart rate difference when the user changes from a standing position to a lying position, Represents the change in the extreme heart rate when the user changes from a standing position to a sitting position, a squatting position, a kneeling position, and a lying position, which is the characteristic data set corresponding to the user's daily heart rate test.

[0012] Preferably, the monitoring data set The calculation process is as follows: According to the sensing data set, extract the sensing data of the user's i-th activity behavior, and mark the heart rate of the user's i-th activity behavior as , mark the blood pressure of the user's i-th activity behavior as , mark the moving speed of the user's i-th activity behavior as , mark the tilt angle of the user's i-th activity behavior as , mark the ground clearance of the user's i-th activity behavior as ; In chronological order, extract the sensing data of the user's (i + 1)-th activity behavior from the sensing data set, where the user's i-th activity behavior is earlier than the (i + 1)-th activity behavior, and then mark the heart rate of the user's (i + 1)-th activity behavior as , mark the blood pressure of the user's (i + 1)-th activity behavior as , mark the moving speed of the user's (i + 1)-th activity behavior as , mark the tilt angle of the user's (i + 1)-th activity behavior as , mark the ground clearance of the user's (i + 1)-th activity behavior as ; ; In the formula, Represents the change in heart rate when the user changes from the i-th activity behavior to the (i + 1)-th activity behavior, Represents the change in blood pressure when the user changes from the i-th activity behavior to the (i + 1)-th activity behavior, Represents the change in moving speed when the user changes from the i-th activity behavior to the (i + 1)-th activity behavior, Represents the change in tilt angle when the user changes from the i-th activity behavior to the (i + 1)-th activity behavior, represents the change in the height above the ground when the user transitions from the i-th activity behavior to the (i + 1)-th activity behavior. represents the monitoring data set when the user transitions from the i-th activity behavior to the (i + 1)-th activity behavior.

[0013] Preferably, the risk index is calculated as follows: According to the environmental data set, extract the monitoring data of the k-th activity space, and mark the environmental brightness of the k-th activity space as , mark the ground flatness of the k-th activity space as , mark the humidity of the k-th activity space as , mark the storage density of the k-th activity space as ; ; In the formula, represents the standard value for measuring the environmental brightness, represents the weight for the ratio of the standard value to the environmental brightness, represents the standard value for measuring the ground flatness, represents the weight for the ratio of the ground flatness to the standard value, represents the standard value for measuring the humidity, represents the weight for the ratio of the humidity to the standard value, represents the weight for the storage density, , , and are all constants, and , represents calculated according to , , and weights, the risk index of the k-th activity space .

[0014] Preferably, in the sensing data set, when the heart rate of the user during any activity behavior is lower than all the values in the daily heart rate test data set , it indicates that the user's heart rate is abnormal and there is a high risk of falling. The user is asked through the mobile communication service whether they are feeling unwell. If no response from the user is received within 5 seconds, the emergency contact of the user is contacted through the mobile communication service. The heart rate change amount of the monitoring data set exceeds the characteristic data set of the heart rate test When all the numerical values are abnormal, it indicates that the user's heart rate has changed abnormally and there is a relatively high risk of falling. The user is asked whether a fall has occurred through the mobile communication service. If no response from the user is received within 5 seconds, the user's emergency contact is contacted through the mobile communication service.

[0015] Preferably, the risk index exceeds the danger range When it does, it indicates that there are many potential safety hazards in the activity space and there is a relatively high risk of falling. The user is reminded to pay attention to safety through the mobile communication service. If no response from the user is received within 5 seconds, the user's emergency contact is reminded through the mobile communication service.

[0016] Compared with the prior art, the present invention provides a home fall warning system based on mobile communication services, which has the following beneficial effects: 1. The present invention connects the database, the sensing device and the camera through the multi-dimensional acquisition module network to obtain the test data of the user's daily behavior, the sensing data of the user's activity behavior and the monitoring data of the user's activity space. The intelligent analysis module analyzes the activity state of the user in each posture according to the test data set. Among them, a corresponding daily data group and characteristic data group are generated for each posture, and a unique health baseline is established for each user, significantly improving the sensitivity of abnormal detection, paying attention to the individual differences of users. The intelligent analysis module analyzes the activity state of the user in real time according to the sensing data set and generates the corresponding monitoring data group , dynamically tracks the real-time data, timely identifies the signs of falling, and the intelligent analysis module analyzes the potential safety hazards of each activity space according to the environmental data set and generates the corresponding risk index , dynamically quantifies the potential safety hazards of the activity space, with high accuracy in multi-dimensional evaluation, and avoids the user staying in a high-risk environment for a long time.

[0017] 2. The present invention sets a fixed range of danger intervals through the intelligent analysis module , and then combines the sensing data set, the daily data group , the characteristic data group , the monitoring data group and the risk index to evaluate the activity state of the user and the risk of falling in the activity space, determine whether to activate the mobile communication service, and execute the corresponding inquiry and emergency measures. If no response from the user is received within 5 seconds, the user's emergency contact is contacted through the mobile communication service, establishing a hierarchical response mechanism, ensuring the efficiency of emergency handling. The hierarchical strategy not only avoids disturbing the user excessively, but also provides effective guarantee at critical moments, with high sensitivity in rapid response. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 is the system flow chart of the present invention. Detailed implementation manners

[0019] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0020] Since the traditional home fall warning system lacks in-depth analysis of users' daily behavior patterns, often ignores the individual differences of users, cannot establish a health baseline according to each user's behavior habits, has low sensitivity for abnormal monitoring, and slow real-time warning and emergency response speed, a home fall warning system based on mobile communication services is provided. Please refer to Figure 1 , the home fall warning system based on mobile communication services includes a multi-dimensional acquisition module and an intelligent analysis module; The multi-dimensional acquisition module is composed of a test data unit, a sensing data unit, and an environmental data unit. Specifically, multi-modal data is collected through wearable devices, which can more comprehensively reflect the user's physical state and the safety level of the surrounding environment, and reduce the probability of misjudgment; The test data unit collects a test data set by connecting to a database through a network. The test data set includes test data of users' daily behaviors. The test data set includes several groups of standing posture test data, several groups of sitting posture test data, several groups of squatting posture test data, several groups of kneeling posture test data, and several groups of lying posture test data. And each group of test data includes heart rate, blood pressure, moving speed, tilt angle, and height from the ground; The sensing data unit collects a sensing data set by connecting to a sensing device through a network. The sensing data set includes sensing data of users' activity behaviors. The expression of the sensing data set is , to represent the sensing data of the user's first to nth activity behaviors. The sensing data includes heart rate, blood pressure, moving speed, tilt angle, and height from the ground. s represents the time point for obtaining the sensing data of the user's activity behavior; The environmental data unit collects an environmental data set by connecting to a camera through a network. The environmental data set includes monitoring data of the activity space where the user is located. The expression of the environmental data set is , to represent the monitoring data of the first to mth activity spaces. The monitoring data includes environmental brightness, ground flatness, humidity, and storage density. j represents the time point for obtaining the monitoring data of the activity space where the user is located; The intelligent analysis module consists of a daily assessment unit, a real-time assessment unit, an environmental assessment unit, and a fall warning unit. The daily assessment unit analyzes the activity status of the user in each posture according to the test data set. Among them, corresponding daily data groups are generated for each posture. And characteristic data groups . Specifically, the standing posture test corresponds to a heart rate test daily data group , a heart rate test characteristic data group , a blood pressure test daily data group, a blood pressure test characteristic data group, a moving speed test daily data group, a moving speed test characteristic data group, an inclination angle test daily data group, an inclination angle test characteristic data group, a ground clearance test characteristic data group, a ground clearance test daily data group, and so on. Similarly, corresponding daily data groups And characteristic data groups Exist for sitting, squatting, kneeling, and lying postures, and the calculation logics of each daily data group And characteristic data group Are the same; The calculation process of the daily data group Is as follows: According to the test data set, mark the heart rate in the standing posture test data as , To Indicating the heart rate during the first to the a-th standing posture tests, mark the heart rate in the sitting posture test data as , To Indicating the heart rate during the first to the b-th sitting posture tests, mark the heart rate in the squatting posture test data as , To Indicating the heart rate during the first to the c-th squatting posture tests, mark the heart rate in the kneeling posture test data as , To Indicating the heart rate during the first to the d-th kneeling posture tests, mark the heart rate in the lying posture test data as , To Indicating the heart rate during the first to the e-th lying posture tests; ; In the formula, Represents the average heart rate of the user during daily long-term standing behavior, Represents the average heart rate of the user during daily long-term sitting behavior, Represents the average heart rate of the user during daily long-term squatting behavior, Represents the average heart rate of the user during daily long-term kneeling behavior, Represents the average heart rate of the user during daily long-term lying behavior, It represents the active heart rates of the user in standing, sitting, squatting, kneeling, and lying postures, which are the daily data sets corresponding to the user's daily heart rate tests. A unique health baseline is established for each user, significantly improving the sensitivity of anomaly detection; Characteristic data set The calculation process is as follows: ; In the formula, represents the maximum heart rate of the user during daily long - standing behavior, represents the maximum heart rate of the user during daily long - sitting behavior, represents the maximum heart rate difference when the user changes from a standing posture to a sitting posture, represents the maximum heart rate of the user during daily long - squatting behavior, represents the maximum heart rate difference when the user changes from a standing posture to a squatting posture, represents the maximum heart rate of the user during daily long - kneeling behavior, represents the maximum heart rate difference when the user changes from a standing posture to a kneeling posture, represents the maximum heart rate of the user during daily long - lying behavior, represents the maximum heart rate difference when the user changes from a standing posture to a lying posture, represents the minimum heart rate of the user during daily long - standing behavior, represents the minimum heart rate of the user during daily long - sitting behavior, represents the minimum heart rate difference when the user changes from a standing posture to a sitting posture, represents the minimum heart rate of the user during daily long - squatting behavior, represents the minimum heart rate difference when the user changes from a standing posture to a squatting posture, represents the minimum heart rate of the user during daily long - kneeling behavior, represents the minimum heart rate difference when the user changes from a standing posture to a kneeling posture, represents the minimum heart rate of the user during daily long - lying behavior, represents the minimum heart rate difference when the user changes from a standing posture to a lying posture, represents the change amount of heart rate extreme values when the user changes from a standing posture to sitting, squatting, kneeling, and lying postures, which is the characteristic data set corresponding to the user's daily heart rate test. It pays attention to the individual differences of users. According to the behavior habits of each user, the change amounts of heart rate extreme values, blood pressure extreme values, moving speed extreme values, tilt angle extreme values, and ground clearance extreme values between different activity behaviors of the user are analyzed according to a unified calculation logic, providing data support for subsequent multi - dimensional comparison; The real - time evaluation unit analyzes the user's activity status in real - time according to the sensing data set and generates the corresponding monitoring data set , and its calculation process is as follows: Extract the sensor data of the user's i-th activity behavior according to the sensor data set, and mark the heart rate of the user's i-th activity behavior as , mark the blood pressure of the user's i-th activity behavior as , mark the moving speed of the user's i-th activity behavior as , mark the tilt angle of the user's i-th activity behavior as , mark the height above the ground of the user's i-th activity behavior as ; Extract the sensor data of the user's (i + 1)-th activity behavior from the sensor data set in chronological order, where the user's i-th activity behavior is earlier than the (i + 1)-th activity behavior, and then mark the heart rate of the user's (i + 1)-th activity behavior as , mark the blood pressure of the user's (i + 1)-th activity behavior as , mark the moving speed of the user's (i + 1)-th activity behavior as , mark the tilt angle of the user's (i + 1)-th activity behavior as , mark the height above the ground of the user's (i + 1)-th activity behavior as ; ; In the formula, represents the change in heart rate when the user switches from the i-th activity behavior to the (i + 1)-th activity behavior, represents the change in blood pressure when the user switches from the i-th activity behavior to the (i + 1)-th activity behavior, represents the change in moving speed when the user switches from the i-th activity behavior to the (i + 1)-th activity behavior, represents the change in tilt angle when the user switches from the i-th activity behavior to the (i + 1)-th activity behavior, represents the change in height above the ground when the user switches from the i-th activity behavior to the (i + 1)-th activity behavior, represents the monitoring data set when the user switches from the i-th activity behavior to the (i + 1)-th activity behavior, dynamically tracks real-time data, and timely identifies fall signs; The environmental assessment unit analyzes the potential safety hazards of each activity space according to the environmental data set and generates a corresponding danger index , and its calculation process is as follows: Extract the monitoring data of the k-th activity space according to the environmental data set, and mark the environmental brightness of the k-th activity space as , mark the ground flatness of the k-th activity space as , mark the humidity of the k-th activity space as , mark the storage density of the k-th activity space as ; ; In the formula, represents the standard value for measuring the environmental brightness, represents the weight for the ratio of the standard value to the environmental brightness, represents the standard value for measuring the ground flatness, represents the weight for the ratio of the ground flatness to the standard value, represents the standard value for measuring the humidity, represents the weight for the ratio of the humidity to the standard value, represents the weight for the storage density, , , and are all constants, and , represents calculating the risk index , , and of the k-th activity space according to the weights, dynamically quantifying the safety hazards of the activity space, with high-precision multi-dimensional evaluation, and avoiding users staying in a high-risk environment for a long time; The fall warning unit sets a dangerous interval with a fixed range , and then combines the sensing data set, daily data set , feature data set , monitoring data set and the risk index to evaluate the fall risk of the user's activity status and activity space, determine whether to activate the mobile communication service, and execute the corresponding inquiries and emergency measures, establishing a hierarchical response mechanism to ensure the efficiency of emergency handling; In the sensing data set, when the heart rate of the user during any activity behavior is lower than all the values in the daily heart rate test data set , it indicates that the user's heart rate is abnormal and there is a relatively high risk of falling. The user is asked through the mobile communication service whether they are feeling unwell. If no response from the user is received within 5 seconds, the user's emergency contact is contacted through the mobile communication service, ensuring that rescue can still be obtained in case of an emergency with no response; When the heart rate change amount of the monitoring data set exceeds all the values in the feature data set of the heart rate test , it indicates that the user's heart rate change is abnormal and there is a relatively high risk of falling. The user is asked through the mobile communication service whether they have fallen. If no response from the user is received within 5 seconds, the user's emergency contact is contacted through the mobile communication service; When the risk index When it indicates that there are many potential safety hazards in the activity space and a relatively high risk of falling, the user will be reminded of safety through mobile communication services. If no response from the user is received within 5 seconds, the user's emergency contacts will be reminded through mobile communication services. The grading strategy can avoid over-disturbing the user while providing effective protection at critical moments and has a high sensitivity for rapid response.

[0021] Example 1: In this experiment, elderly people aged 70 were selected as the experimental subjects. After testing, when the elderly people changed from a standing position to a sitting position, their heart rate increased from 70 beats per minute to 75 beats per minute, blood pressure changed from 120 / 80 mmHg to 125 / 85 mmHg, moving speed increased from 0.5 m / s to 1.0 m / s, tilt angle changed from 0 degrees to 15 degrees, and the height from the ground decreased from 1.7 meters to 0.5 meters. The monitoring data set when the elderly people changed from a standing position to a sitting position The calculation process is as follows: ; In the formula, represents the change in heart rate when the elderly people change from a standing position to a sitting position, represents the change in blood pressure when the elderly people change from a standing position to a sitting position, represents the change in moving speed when the elderly people change from a standing position to a sitting position, represents the change in tilt angle when the elderly people change from a standing position to a sitting position, represents the change in height from the ground when the elderly people change from a standing position to a sitting position. Given the heart rate test characteristic data set of this elderly person is After judgment, when the change in heart rate of the monitoring data set has exceeded all the values in the heart rate test characteristic data set it indicates that the user's heart rate has changed abnormally and there is a relatively high risk of falling. The user will be asked through mobile communication services whether they have fallen. If no response from the user is received within 5 seconds, the user's emergency contacts will be contacted through mobile communication services.

[0022] Example 2: In this experiment, the living room in the home activity space was selected as the experimental object. After testing, the environmental brightness of the living room was 800 lumens, the ground flatness was 0.6 mm / ㎡, the humidity was 40%, and the storage density was 10%. The risk index of this living room The calculation process is as follows: ; In the formula, represents the standard value for measuring environmental brightness, represents the weight for the ratio of the standard value to the environmental brightness, represents the standard value for measuring ground flatness, represents the weight for the ratio of the ground flatness to the standard value, represents the standard value for measuring humidity, represents the weight for the ratio of humidity to the standard value, represents the weight for the storage density, and and and are all constants, and According to and and and weights, the danger index of this living room is calculated to be approximately 0.65, and the danger range is set to 0 - 0.5. After judgment, the danger index of this living room has exceeded the danger range indicating that there are more potential safety hazards in the living room and a relatively high risk of falling. The user is reminded of safety through mobile communication services. If no response from the user is received within 5 seconds, the user's emergency contact is reminded through mobile communication services.

[0023] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A home fall warning system based on mobile communication services, characterized in that: It includes a multi-dimensional acquisition module and an intelligent analysis module; The multi-dimensional acquisition module is composed of a test data unit, a sensing data unit, and an environmental data unit. The test data unit collects a test data set by connecting to a database through a network. The test data set includes test data of users' daily behaviors. The sensing data unit collects a sensing data set by connecting to a sensing device through a network. The sensing data set includes sensing data of users' activity behaviors. The environmental data unit collects an environmental data set by connecting to a camera through a network. The environmental data set includes monitoring data of the activity space where the user is located; The intelligent analysis module consists of a daily assessment unit, a real-time assessment unit, an environmental assessment unit, and a fall warning unit. The daily assessment unit analyzes the activity status of the user in each posture according to the test data set. Among them, a corresponding daily data set is generated for each posture. and a feature data set . The real-time assessment unit analyzes the activity status of the user in real time according to the sensing data set and generates a corresponding monitoring data set. . The environmental assessment unit analyzes the potential safety hazards of each activity space according to the environmental data set and generates a corresponding danger index. . The fall warning unit is set with a danger range of a fixed range. . Then, combined with the sensing data set, the daily data set , the feature data set , the monitoring data set and the danger index , it evaluates the fall risk of the user's activity status and activity space, determines whether to activate the mobile communication service, and executes corresponding inquiries and emergency measures.

2. The home fall warning system based on mobile communication services according to claim 1, characterized in that: The test data set includes several groups of standing posture test data, several groups of sitting posture test data, several groups of squatting posture test data, several groups of kneeling posture test data, and several groups of lying posture test data. And each group of test data includes heart rate, blood pressure, moving speed, tilt angle, and height from the ground.

3. The home fall warning system based on mobile communication services according to claim 2, characterized in that: The expression of the said sensing data set is , to represent the sensing data of the user's first to nth activity behaviors. The sensing data includes heart rate, blood pressure, moving speed, tilt angle, and height from the ground. s represents the time point for obtaining the sensing data of the user's activity behaviors.

4. The home fall warning system based on mobile communication services according to claim 3, wherein: The expression of the environmental data set is , to represent the monitoring data of the first to the mth activity spaces. The monitoring data includes environmental brightness, ground flatness, humidity, and storage density. j represents the time point when the monitoring data of the activity space where the user is located is obtained.

5. The home fall warning system based on mobile communication services according to claim 4, characterized in that: The daily data group The calculation process is as follows: According to the test data set, mark the heart rate in the standing posture test data as , to representing the heart rate during the first to the a-th standing posture tests, mark the heart rate in the sitting posture test data as , to representing the heart rate during the first to the b-th sitting posture tests, mark the heart rate in the squatting posture test data as , to representing the heart rate during the first to the c-th squatting posture tests, mark the heart rate in the kneeling posture test data as , to representing the heart rate during the first to the d-th kneeling posture tests, mark the heart rate in the lying posture test data as , to representing the heart rate during the first to the e-th lying posture tests; ； In the formula, represents the average heart rate of the user during daily long - standing behavior, represents the average heart rate of the user during daily sedentary behavior, represents the average heart rate of the user during daily long - squatting behavior, represents the average heart rate of the user during daily long - kneeling behavior, represents the average heart rate of the user during daily long - lying behavior, represents the activity heart rate of the user in standing, sitting, squatting, kneeling and lying postures, that is, the daily data set corresponding to the user's daily heart rate test.

6. The home fall warning system based on mobile communication services according to claim 5, characterized in that: The said characteristic data group The calculation process is as follows: ； In the formula, represents the maximum heart rate of the user during daily long - standing behavior, represents the maximum heart rate of the user during daily sedentary behavior, represents the difference in maximum heart rate when the user changes from a standing position to a sitting position, represents the maximum heart rate of the user during daily long - squatting behavior, represents the difference in maximum heart rate when the user changes from a standing position to a squatting position, represents the maximum heart rate of the user during daily long - kneeling behavior, represents the difference in maximum heart rate when the user changes from a standing position to a kneeling position, represents the maximum heart rate of the user during daily long - lying behavior, represents the difference in maximum heart rate when the user changes from a standing position to a lying position, represents the minimum heart rate of the user during daily long - standing behavior, represents the minimum heart rate of the user during daily sedentary behavior, represents the difference in minimum heart rate when the user changes from a standing position to a sitting position, represents the minimum heart rate of the user during daily long - squatting behavior, represents the difference in minimum heart rate when the user changes from a standing position to a squatting position, represents the minimum heart rate of the user during daily long - kneeling behavior, represents the difference in minimum heart rate when the user changes from a standing position to a kneeling position, represents the minimum heart rate of the user during daily long - lying behavior, represents the difference in minimum heart rate when the user changes from a standing position to a lying position, represents the change in extreme heart rate values when the user changes from a standing position to a sitting position, a squatting position, a kneeling position, and a lying position, which is the characteristic data set corresponding to the user's daily heart rate test.

7. The home fall warning system based on mobile communication services according to claim 6, wherein: The monitored data group The calculation process is as follows: Extract the sensor data of the user's i-th activity behavior according to the sensor data set, and mark the heart rate of the user's i-th activity behavior as , mark the blood pressure of the user's i-th activity behavior as , mark the moving speed of the user's i-th activity behavior as , mark the tilt angle of the user's i-th activity behavior as , mark the height above the ground of the user's i-th activity behavior as ; Extract the sensor data of the user's (i + 1)-th activity behavior from the sensor dataset in chronological order, where the user's i-th activity behavior is earlier than the (i + 1)-th activity behavior, and then mark the heart rate of the user's (i + 1)-th activity behavior as , mark the blood pressure of the user's (i + 1)-th activity behavior as , mark the moving speed of the user's (i + 1)-th activity behavior as , mark the tilt angle of the user's (i + 1)-th activity behavior as , mark the height above the ground of the user's (i + 1)-th activity behavior as ; ； In the formula, represents the change in heart rate when the user transitions from the i-th activity behavior to the (i + 1)-th activity behavior, represents the change in blood pressure when the user transitions from the i-th activity behavior to the (i + 1)-th activity behavior, represents the change in movement speed when the user transitions from the i-th activity behavior to the (i + 1)-th activity behavior, represents the change in tilt angle when the user transitions from the i-th activity behavior to the (i + 1)-th activity behavior, represents the change in ground clearance when the user transitions from the i-th activity behavior to the (i + 1)-th activity behavior, represents the monitoring data set when the user transitions from the i-th activity behavior to the (i + 1)-th activity behavior.

8. The home fall warning system based on mobile communication services according to claim 7, wherein: The said danger index The calculation process is as follows: Extract the monitoring data of the k-th activity space according to the environmental data set, and mark the environmental brightness of the k-th activity space as , mark the ground flatness of the k-th activity space as , mark the humidity of the k-th activity space as , mark the storage density of the k-th activity space as ; ； In the formula, represents the standard value for measuring the ambient brightness, represents the weight for the ratio of the standard value to the ambient brightness, represents the standard value for measuring the ground flatness, represents the weight for the ratio of the ground flatness to the standard value, represents the standard value for measuring the humidity, represents the weight for the ratio of the humidity to the standard value, represents the weight for the storage density, , , and are all constants, and , represents calculated according to , , and weights, the hazard index of the k-th activity space .

9. The home fall warning system based on mobile communication services according to claim 8, wherein: When the heart rate during any user activity behavior in the described sensing data set is lower than all the values in the daily heart rate test data set, it indicates that the user's heart rate is abnormal and there is a high risk of falling. The user is asked through the mobile communication service whether they are feeling unwell. If no response from the user is received within 5 seconds, the emergency contact of the user is contacted through the mobile communication service. The monitoring data set When the heart rate change amount exceeds all the values in the heart rate test characteristic data set , it indicates that the user's heart rate change is abnormal and there is a high risk of falling. The user is asked through the mobile communication service whether they have fallen. If no response from the user is received within 5 seconds, the emergency contact of the user is contacted through the mobile communication service. ​ 10. The home fall warning system based on mobile communication services according to claim 9, characterized in that: The described danger index exceeds the danger range indicating that there are more potential safety hazards in the activity space and a relatively high risk of falling. The user will be reminded to pay attention to safety through mobile communication services. If no response from the user is received within 5 seconds, the user's emergency contacts will be reminded through mobile communication services.

Images