An indoor and outdoor health monitoring positioning method

Abstract

The application belongs to the technical field of health monitoring, and discloses an indoor and outdoor health monitoring positioning method, which comprises the following steps: acquiring wrist health monitoring data and foot inertial navigation data of a user; acquiring health assessment data of the user based on the wrist health monitoring data, and acquiring posture data and positioning data of the user based on the foot inertial navigation data; judging whether the user needs rescue based on the health assessment data and the posture data of the user; and implementing rescue in combination with the positioning data of the user when the user needs rescue. The technical scheme improves the accuracy and reliability of health monitoring and position tracking, and can provide high-precision health monitoring and position tracking services for the user in a complex environment and reliable emergency support services in an emergency state.

Description

Technical Field

[0001] This invention belongs to the field of health monitoring technology, and in particular relates to an indoor and outdoor health monitoring positioning method. Background Technology

[0002] Health monitoring technology has made significant progress in recent years, with a wide variety of devices available, ranging from professional medical-grade equipment to convenient wearable devices, widely used in clinical and personal health management. Professional medical devices, such as electrocardiogram (ECG) monitors, are widely used for cardiac health monitoring, recording cardiac electrical activity in real time and providing accurate diagnosis of diseases such as arrhythmias and myocardial infarction. These devices offer high accuracy and provide reliable data for clinical analysis, but their large size and high price limit their use in home and non-medical environments. Furthermore, ECG monitors rely on electrodes and wires for connection, which may cause discomfort with prolonged wear. Pulse oximeters non-invasively monitor blood oxygen saturation (SpO2) and heart rate using the principle of light absorption, making them particularly suitable for detecting respiratory problems and widely used in clinical and home environments. While simple to use and providing convenient real-time feedback, their functionality is relatively limited and lacks real-time monitoring capabilities. Smart bracelets are also a popular choice for daily health monitoring, with built-in accelerometers, heart rate sensors, etc., capable of monitoring multiple physiological indicators such as steps, heart rate, and sleep quality. However, smart bracelets often lack precise location coordinates and accurate recognition of the wearer's movement patterns. Smart chest straps are primarily used for monitoring heart rate and respiratory rate, typically applied during athlete training. Compared to smart bracelets, smart chest straps offer higher accuracy in heart rate monitoring, making them suitable for applications requiring high precision. However, chest straps are less comfortable to wear, and prolonged use can cause discomfort. Smart head-mounted devices, such as smart earmuffs and smart glasses, are also used in health monitoring. These devices can monitor multi-modal health signals, such as EEG and EOG. However, despite providing the ability to monitor complex physiological signals, they have high power consumption requirements and still face challenges in terms of comfort and stability.

[0003] Location tracking technology has significant application value in health monitoring devices, enabling timely access to user location information for more precise health management. Currently, major location technologies include Global Navigation Satellite Systems (GNSS, such as GPS), WiFi positioning, Bluetooth positioning, and Ultra-Wideband (UWB) positioning.

[0004] GPS, as a mature global positioning technology, can provide outdoor positioning accuracy of several meters to ten meters, but performs poorly indoors or in obstructed environments. When the signal is blocked or affected by multipath reflection, its positioning accuracy drops significantly, and signal loss may even occur. WiFi positioning has the advantage of widespread availability and low cost. With WiFi networks almost universally covering modern buildings and the widespread use of smartphones, WiFi positioning requires no additional hardware support and has a relatively large coverage area, meeting the needs of most indoor environments. However, WiFi signals are sensitive, highly fluctuating, and greatly affected by the environment; even slight changes in the environment can cause significant changes in RSS, leading to reduced positioning accuracy. Bluetooth positioning uses the signal strength index (RSSI) of Bluetooth Low Energy (BLE) devices for positioning estimation, suitable for short-range positioning scenarios. However, RSSI signals are highly susceptible to interference from obstacles and multipath effects, with positioning accuracy typically around 2-3 meters, and its applicability is limited to small spaces. In recent years, UWB positioning technology has received widespread attention due to its extremely high time resolution and anti-interference capabilities. Ultra-wideband (UWB) technology achieves centimeter-level high-precision positioning by transmitting ultra-wideband wireless signals, making it particularly suitable for scenarios requiring extremely high positioning accuracy, such as real-time location tracking in medical and health monitoring. Furthermore, UWB is well-adapted to multipath effects in complex environments, providing higher positioning stability than WiFi and Bluetooth. Nevertheless, UWB positioning still has certain limitations. It typically requires the deployment of multiple base stations within the target area, and this high dependence on infrastructure increases deployment costs and complexity. In addition, UWB signal transmission consumes relatively high power, placing higher demands on the battery life of portable devices (such as fitness trackers). Summary of the Invention

[0005] The purpose of this invention is to provide an indoor and outdoor health monitoring and positioning method to solve the problems existing in the prior art.

[0006] To achieve the above objectives, the present invention provides an indoor and outdoor health monitoring and positioning method, comprising:

[0007] The system acquires the user's wrist health monitoring data and foot inertial navigation data. The wrist health monitoring data includes PPG data, ECG data, and accelerometer data, and the foot inertial navigation data includes accelerometer data, gyroscope data, and magnetometer data.

[0008] Based on the wrist health monitoring data, obtain the user's health assessment data; based on the foot inertial navigation data, obtain the user's posture data and positioning data.

[0009] Based on the health assessment data and the user posture data, it is determined whether the user needs rescue. When the user needs rescue, rescue is carried out in conjunction with the user's location data.

[0010] Optionally, obtaining the user's health assessment data based on the wrist health monitoring data specifically includes:

[0011] The wrist health monitoring data is filtered, and the user's heart rate and blood oxygen data are calculated based on the filtered PPG data and filtered accelerometer data. The user's heart rate variability data is calculated based on the filtered ECG data, and the user's blood pressure data is calculated by combining the filtered PPG data and filtered ECG data. Based on the heart rate data, blood oxygen data, heart rate variability data, blood pressure data and preset thresholds, the user's health status is assessed to obtain health assessment data.

[0012] Optionally, the step of calculating the user's heart rate and blood oxygen data based on the filtered PPG data and filtered accelerometer data includes the following specific calculation process:

[0013] Filtering:

[0014]

[0015] S(f) = w(f)·P(f)

[0016] In the formula, P(t,k) is the value of the PPG signal of length a collected at time t after Fourier transform at the k-th frequency point, N(t,k) is the acceleration signal, w represents the filter function, c represents the selected window size, and S represents the filtered frequency domain signal.

[0017] Peak detection is performed on the filtered frequency domain signal, and the frequency of the peak is set to k. i Calculate heart rate (HR):

[0018] HR = k i *60

[0019] The filtered frequency domain signal is then restored to the time domain using an inverse Fourier transform to obtain s(t,a):

[0020] AC = max(s(t,a)) - min(s(t,a))

[0021]

[0022] In the formula, AC is the AC component of the PPG signal, DC is the DC component of the signal, and R is the normalized infrared absorption ratio.

[0023] Based on the preset standard value, linear regression was performed on R to obtain blood oxygen data.

[0024] Optionally, the calculation of the user's heart rate variability data based on the filtered ECG data is specifically calculated using the following formula:

[0025] Filtering and R-peak detection of ECG data:

[0026]

[0027] In the formula, x(n) is the ECG time-domain signal sampled at time n, L is the signal length, and x norm For the normalized signal, E shannon It is the Shannon energy of the signal, E smooth To smooth the Shannon energy, locs(i) is the index of the R peak in the ECG signal. Representing the Hilbert transform, Z is the set of indices of the crossing zeros; ∈ is a very small constant used to avoid invalid inputs to the logarithmic function, RR i F is the time interval between adjacent R waves. s It is the sampling rate. It is a signal that has undergone Hilbert transformation.

[0028] Calculate health-related parameters:

[0029]

[0030] In the formula, N is the number of RR intervals in the calculation window. SDSD is the average RR value, SDNN is the standard deviation of the difference between adjacent normal heartbeat intervals, NN50 is the number of intervals between adjacent normal heartbeat intervals exceeding 50ms, and pNN50 is the percentage of NN50 in the total number of heartbeats.

[0031] The user's health status is analyzed based on health-related parameters.

[0032] Optionally, the user's blood pressure data is calculated by combining the filtered PPG data and the filtered ECG data, and the specific calculation formula is as follows:

[0033] Select the calculation window W f oot and W d The time interval between the R wave and the maximum first derivative point corresponding to the PPG data was calculated, and the diastolic time was extracted from the PPG signal. Combined with the user's own physiological parameters, a personalized blood pressure regression calculation model was then established for each user to obtain the blood pressure:

[0034]

[0035]

[0036] Dia t2 =t foot -t min

[0037] In the formula, P(t) is the PPG signal, and t max 'The point where the maximum derivative of P(t) is calculated within the window, t foot W represents the time point at which the waveform begins during the PPG signal period. foot W is the time window that includes the beginning portion of the waveform. dia t is the time window that includes the diastolic phase of the signal. min The lowest point of the signal's diastolic phase, PTT dp t represents the pulse wave conduction time. R波 Dia represents the time when the R peak appears within the corresponding time window. t2 This refers to the diastolic time interval.

[0038] Optionally, the calculation process of the attitude data specifically includes:

[0039]

[0040] In the formula, p n v n and These are the position, velocity, and attitude rotation matrices of the downloaded volume in the n-frame (navigation coordinate system); f n For accelerometer observations of the n-system; Let be the Earth's rotational angular velocity in the n-system; and These are the velocity and angular velocity of the n-frame relative to the e-frame in the n-frame; g n Let gravitational acceleration be the acceleration due to gravity in the n-system. These are the original observations of the gyroscope; Let n be the angular velocity of the n-system relative to the i-system in the b-system, where n is the navigation coordinate system, b is the vehicle coordinate system, and i is the geocentric inertial coordinate system.

[0041] Optionally, the specific calculation process for the positioning data is as follows:

[0042]

[0043] In the formula, N is the length of the calculation window, μ represents the signal mean, σ represents the standard deviation, K represents kurtosis, S represents skewness, R represents the root mean square, and x i The data returned by the accelerometer and gyroscope at time i is used to input the feature vector v into the ECOC classification model. A pre-trained classifier generates a 31-bit codeword, which is then matched against the closest class.

[0044]

[0045] In the formula, N is the length of the calculation window, μ represents the signal mean, σ represents the standard deviation, K represents kurtosis, S represents skewness, R represents the root mean square, and xi This represents the data returned by the accelerometer and gyroscope at time i.

[0046] Optionally, when a user needs rescue, a distress signal containing the user's current location data, heart rate data, blood oxygen data, heart rate variability data, and blood pressure data is sent to the cloud, and rescue is carried out based on the distress signal.

[0047] The technical effects of this invention are as follows:

[0048] This invention integrates foot inertial sensors and wristband devices, combining location tracking and vital sign monitoring functions to achieve precise indoor and outdoor seamless positioning and continuous health monitoring. This system effectively solves the technical challenges of positioning accuracy and real-time vital sign monitoring in complex environments, providing reliable support for national safety, health management, smart healthcare, and emergency response.

[0049] This lightweight wearable system is characterized by its miniaturization, intelligence, and low power consumption, making it widely applicable in fields such as smart healthcare, health management, and emergency rescue. It provides continuous location, behavior recognition, and vital sign monitoring. Based on a foot-based inertial navigation system (INS) algorithm and a precise gait recognition algorithm, the system achieves seamless indoor and outdoor positioning with an accuracy better than 1 meter, providing stable 3D position tracking even in complex environments. Its gait recognition algorithm accurately identifies walking, running, climbing stairs, elevator use, and falls, with an accuracy rate exceeding 98%. For emergencies such as falls, the system responds in less than 3 seconds, quickly locating the cause and triggering relevant alarms to ensure timely response in emergencies. Regarding vital sign monitoring, the system can monitor physiological parameters such as heart rate, blood oxygen, and blood pressure in real time, providing continuous and accurate health status feedback and reliable health management support for users.

[0050] This invention not only realizes a miniaturized and intelligent health monitoring and positioning system, but also solves the stability and reliability problems of positioning and monitoring systems in complex environments by integrating multiple technologies such as foot inertial algorithms, gait recognition algorithms, emergency response modules, and multimodal physiological indicator monitoring modules. The system integrates multiple health indicators to provide a comprehensive assessment of the user's health status, including healthy, sub-healthy, good, and risky states, ensuring comprehensive and reliable health management. Simultaneously, through the deep integration of gait recognition and inertial navigation, it effectively overcomes the problem of long-term cumulative errors in inertial navigation systems, significantly improving the stability and reliability of positioning and monitoring, and providing users with continuous and accurate service support. Attached Figure Description

[0051] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0052] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0053] Figure 1 A structural diagram of a device for continuous health monitoring, location tracking, and emergency response provided in an embodiment of the present invention;

[0054] Figure 2 A schematic diagram showing the results of health monitoring and location tracking of a device for continuous health monitoring and location tracking, and emergency response. Detailed Implementation

[0055] Various exemplary embodiments of the present invention will now be described in detail. This detailed description should not be considered as a limitation of the present invention, but rather as a more detailed description of certain aspects, features, and embodiments of the present invention.

[0056] It should be understood that the terminology used in this invention is merely for describing particular embodiments and is not intended to limit the invention. Furthermore, with respect to numerical ranges in this invention, it should be understood that each intermediate value between the upper and lower limits of the range is also specifically disclosed. Every smaller range between any stated value or intermediate value within a stated range, and any other stated value or intermediate value within said range, is also included in this invention. The upper and lower limits of these smaller ranges may be independently included or excluded from the range.

[0057] Unless otherwise stated, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art. While only preferred methods have been described herein, any methods similar or equivalent to those described herein may be used in the implementation or testing of this invention. All references to this specification are incorporated by way of citation to disclose and describe the methods associated with those references. In the event of any conflict with any incorporated reference, the content of this specification shall prevail.

[0058] Various modifications and variations can be made to the specific embodiments described in this specification without departing from the scope or spirit of the invention, as will be apparent to those skilled in the art. Other embodiments derived from this specification will also be obvious to those skilled in the art. This application specification and embodiments are merely exemplary.

[0059] The terms “include,” “including,” “have,” “contain,” etc., used in this article are all open-ended terms, meaning that they include but are not limited to.

[0060] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0061] like Figure 1 - Figure 2 As shown in the figure, this embodiment provides an indoor and outdoor health monitoring and positioning method, including: acquiring a user's wrist health monitoring data and foot inertial navigation data, wherein the wrist health monitoring data includes PPG data, ECG data, and accelerometer data, and the foot inertial navigation data includes accelerometer data, gyroscope data, and magnetometer data; acquiring the user's health assessment data based on the wrist health monitoring data, and acquiring the user's posture data and positioning data based on the foot inertial navigation data; determining whether the user needs rescue based on the health assessment data and the user's posture data, and implementing rescue in conjunction with the user's positioning data when the user needs rescue.

[0062] This embodiment discloses a device, equipment, and medium for continuous health monitoring, location tracking, and emergency response. The steps include: acquiring PPG (Photoplethysmography), ECG (Electrocardiogram), and accelerometer data from a wrist health monitoring unit, and obtaining health information through a signal processing and analysis unit; extracting data from a triaxial accelerometer, triaxial gyroscope, and triaxial magnetometer from a foot inertial unit, and obtaining inertial positioning coordinates and user posture through a zero-speed update unit, a mechanical orchestration unit, and an ECOC classification unit; fusing the health information and user posture through a state assessment unit to determine whether the user needs rescue; and finally, providing the user's health information and location information through an output unit, and deciding whether to send a distress signal based on the user's status. This embodiment, by constructing a fusion system and algorithm for health sensors and foot inertial sensors, greatly improves the accuracy and reliability of health monitoring and location tracking, providing users with high-precision health monitoring and location tracking services in complex environments and reliable emergency support services in emergency situations.

[0063] like Figure 1As shown in the diagram, this embodiment provides a structural diagram of a device for continuous health monitoring, location tracking, and emergency response. The device includes a wrist health detection unit, a foot inertial sensor unit, a signal processing and analysis unit, a zero-speed update unit, a mechanical programming unit, an ECOC classification unit, and an output unit. Based on miniaturized and intelligent wrist health monitoring and foot positioning devices, high-precision health monitoring and location tracking services can be achieved in complex environments.

[0064] Specifically, this embodiment describes a positioning method and system integrating a head-mounted BeiDou inertial camera fusion, the positioning method comprising the following steps:

[0065] S1: Obtain PPG, ECG, and accelerometer data from the wrist health monitoring unit, and obtain health information through the signal processing and analysis unit;

[0066] S2: Extract data from the three-axis accelerometer, three-axis gyroscope, and three-axis magnetometer from the foot inertial unit, and obtain inertial positioning coordinates and user attitude through the zero-speed update unit, mechanical orchestration unit, and ECOC classification unit;

[0067] S3: The health information and user posture are assessed by the status evaluation unit to determine whether the user needs rescue, and then output the results through the output unit.

[0068] Implementable, step S1 includes:

[0069] S1-1: Obtain PPG and accelerometer data from the wrist health monitoring unit, and obtain health information through the Wiener filter unit and the heart rate and blood oxygen calculation unit.

[0070] Assuming that the PPG signal of length a acquired at time t has a value of P(t,k) at the k-th frequency point after Fourier transform, and the acceleration signal is N(t,k), then the filtering equation in the Wiener filter unit is:

[0071]

[0072] S(f) = W(f)·P(f)

[0073] Here, w represents the filter function, c represents the selected window size, and S represents the filtered frequency domain signal. Finally, peak detection is performed on S, and the frequency of the peak is set to k. i The heart rate (HR) can be obtained using the following formula:

[0074] HR = k i *60

[0075] If we use the inverse Fourier transform to restore S to the time domain, we get s(t,a), then we have

[0076] AC = max(s(t,a)) - min(s(t,a))

[0077]

[0078] In the formula, AC is the AC component of the PPG signal, DC is the DC component of the signal, and R is the normalized infrared absorption ratio.

[0079] Finally, by referring to the standard value, linear regression of R can be performed to obtain the blood oxygen saturation.

[0080] Also feasible are:

[0081] S1-2: Obtain ECG data from the wrist health monitoring unit, and obtain HRV analysis information through the combined filtering unit and HRV analysis unit.

[0082] Suppose that at time i, an ECG time-domain signal of length x is sampled. Normalizing this signal yields xi. norm Then the smoothed Shannon energy E smooth The calculation equation is:

[0083]

[0084] In the formula, ∈ is a very small constant used to avoid invalid input to the logarithmic function. After performing a Hilbert transform, the zero-crossing point is located, and then the index locs(i) corresponding to the local maximum near the zero-crossing point of the original signal is found, which is the position of the R peak in the ECG signal. Finally, RR is calculated. i .

[0085]

[0086] In the formula, This represents the Hilbert transform, where Z is the set of indices of the cross zeros, locs(i) is the index corresponding to the position of the R peak, and RR... i F is the time interval between adjacent R waves. s It's the sampling rate. It is a signal that has undergone Hilbert transformation.

[0087] RR i Then, health-related parameters can be obtained using the following formula:

[0088]

[0089] In the formula, N is the number of RR intervals in the calculation window. The parameters are: SDSD (mean RR), SDNN (standard deviation of the interval between consecutive normal heartbeats), NN50 (number of intervals between consecutive normal heartbeats exceeding 50ms), and pNN50 (percentage of NN50 intervals out of the total heartbeats). These parameters are then compared to standard values ​​to analyze the user's health status.

[0090] Also feasible are:

[0091] S1-3: The ECG signal output by the combined filtering unit, combined with the PPG signal output by the Wiener filtering unit, is used to calculate the blood pressure by the blood pressure calculation unit.

[0092] Select the calculation window W f oot and W d ia, calculate the time interval between the R-wave and the point of maximum first derivative of PPG (PTT) dp The analysis of PPG signal extraction diastolic time (Dia_t2) is combined with the user's own physiological parameters (height, age, weight, blood pressure value within one hour after waking up).

[0093]

[0094] Dia t2 =t foot -t min

[0095] In the formula, P(t) is the PPG signal, and t max ' is the point where the maximum derivative of P(t) is calculated within the window, t foot It is the time point at which the waveform begins during the PPG signal period, W foot It is a time window that includes the beginning portion of the waveform, W dia It is a time window that includes the diastolic phase of the signal, t min It is the lowest point of the signal's diastolic phase, PTT. dp It is the pulse wave conduction time, t R波 This corresponds to the time when the R peak appears within the corresponding time window. t2 It is the diastolic interval.

[0096] Then, a personalized blood pressure regression calculation model is built for each user to obtain blood pressure;

[0097] Implementable, step S2 includes:

[0098] S2-1: Extract data from the three-axis accelerometer, three-axis gyroscope, and three-axis magnetometer from the foot inertial unit, and obtain inertial positioning coordinates through the zero-speed update unit and the mechanical arrangement unit.

[0099] MEMS sensors use inertial navigation mechanical programming to integrate and recursively calculate a person's position, velocity, and attitude (PVA). The differential equations are expressed as follows:

[0100]

[0101] In the formula, p n v n and These are the position, velocity, and attitude rotation matrices of the downloaded volume in the n-frame (navigation coordinate system); f n For accelerometer observations of the n-system; Let be the Earth's rotational angular velocity in the n-system; and These are the velocity and angular velocity of the n-frame relative to the e-frame in the n-frame; g n Let gravitational acceleration be the acceleration due to gravity in the n-system. These are the original observations of the gyroscope; Let n be the angular velocity of the n-system relative to the i-system (geocentric inertial coordinate system) in the b-system (carrier coordinate system).

[0102] Also feasible are:

[0103] S2-2: Extract data from the three-axis accelerometer and three-axis gyroscope from the foot inertial unit, and obtain the user's posture through the ECOC classification unit.

[0104] Assume the accelerometer and gyroscope data collected at time i are a i ,g i Its length is N. Features are extracted from it according to the following formula to form a feature vector v:

[0105]

[0106] In the formula, N is the length of the calculation window, μ represents the signal mean, σ represents the standard deviation, K represents kurtosis, S represents skewness, and R represents the root mean square. i This represents the data returned by the accelerometer and gyroscope at time i. The feature vector v is input into the ECOC classification model, and a 31-bit codeword is obtained through a pre-trained classifier. Finally, it is matched with the class closest to the codeword. The matching equation is as follows:

[0107]

[0108] Where c represents the index of the action, d(·) is the Hamming distance function, and M c This represents the c-th row in the encoding matrix. This is the final output prediction result.

[0109] Implementable, step S3 includes:

[0110] S3-1: The health information and user posture are fused and judged through the status assessment unit to determine whether the user needs rescue;

[0111] When a user's health parameters show significant abnormalities, this system can promptly identify the abnormal state and send out a distress signal. Specifically, in the event of a fall, entrapment, or severe impact, the user's heart rate rises rapidly, blood pressure increases abnormally, and the ECG signal exhibits irregular waveforms. Simultaneously, the system uses motion recognition results from foot inertial sensors to accurately determine if the user has fallen or is in need of emergency rescue. After determining the emergency, the system automatically sends a distress signal containing parameters such as the user's current location, heart rate, and blood oxygen saturation to the cloud. The process from detecting the abnormality to sending the distress signal takes less than 3 seconds, allowing the user to receive rapid and accurate medical assistance.

[0112] Also feasible are:

[0113] S3-2: Provides user health information and location information through the output unit, and decides whether to send a distress signal to the outside world based on the user's status;

[0114] The device in this embodiment includes a wristband-type health monitoring device, a foot inertial sensor unit, a signal processing and analysis unit, and a posture recognition and positioning unit. These units achieve integrated positioning and health monitoring through the method proposed in this embodiment.

[0115] Figure 2 This diagram illustrates the real-time location and health monitoring of a user in a complex urban environment, using the continuous health monitoring, location tracking, and emergency response device installed in this embodiment. As can be seen from the diagram, the continuous health monitoring, real-time location, and motion recognition method proposed in this embodiment can effectively provide users with real-time health information, location information, and posture information, better supporting emergency rescue efforts in case of emergencies.

[0116] This embodiment integrates foot inertial sensors and wristband devices to achieve precise indoor and outdoor seamless positioning and continuous health monitoring by deeply fusing foot inertial sensors with a wristband-style bracelet. This system effectively solves the technical challenges of positioning accuracy and real-time vital sign monitoring in complex environments, providing reliable support for national safety, health management, smart healthcare, and emergency response.

[0117] This lightweight wearable system is characterized by its miniaturization, intelligence, and low power consumption, making it widely applicable in fields such as smart healthcare, health management, and emergency rescue. It provides continuous location, behavior recognition, and vital sign monitoring. Based on a foot-based inertial navigation system (INS) algorithm and a precise gait recognition algorithm, the system achieves seamless indoor and outdoor positioning with an accuracy better than 1 meter, providing stable 3D position tracking even in complex environments. Its gait recognition algorithm accurately identifies walking, running, climbing stairs, elevator use, and falls, with an accuracy rate exceeding 98%. For emergencies such as falls, the system responds in less than 3 seconds, quickly locating the cause and triggering relevant alarms to ensure timely response in emergencies. Regarding vital sign monitoring, the system can monitor physiological parameters such as heart rate, blood oxygen, and blood pressure in real time, providing continuous and accurate health status feedback and reliable health management support for users.

[0118] This embodiment not only realizes a miniaturized and intelligent health monitoring and positioning system, but also solves the stability and reliability issues of positioning and monitoring systems in complex environments by integrating multiple technologies such as foot inertial algorithms, gait recognition algorithms, emergency response modules, and multimodal physiological indicator monitoring modules. The system integrates multiple health indicators to provide a comprehensive assessment of the user's health status, including various states such as healthy, sub-healthy, good, and at risk, ensuring comprehensive and reliable health management. Simultaneously, through the deep integration of gait recognition and inertial navigation, it effectively overcomes the problem of long-term cumulative errors in inertial navigation systems, significantly improving the stability and reliability of positioning and monitoring, and providing users with continuous and accurate service support.

[0119] The above description is merely a preferred embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for indoor and outdoor health monitoring and positioning, characterized in that, include: The system acquires the user's wrist health monitoring data and foot inertial navigation data. The wrist health monitoring data includes PPG data, ECG data, and accelerometer data, and the foot inertial navigation data includes accelerometer data, gyroscope data, and magnetometer data. Based on the wrist health monitoring data, obtain the user's health assessment data; based on the foot inertial navigation data, obtain the user's posture data and positioning data. Based on the health assessment data and the user posture data, it is determined whether the user needs rescue. When the user needs rescue, rescue is carried out in conjunction with the user's location data. The process of obtaining user health assessment data based on the wrist health monitoring data specifically includes: The wrist health monitoring data is filtered, and the user's heart rate and blood oxygen data are calculated based on the filtered PPG data and filtered accelerometer data. The user's heart rate variability data is calculated based on the filtered ECG data, and the user's blood pressure data is calculated by combining the filtered PPG data and filtered ECG data. Based on the heart rate data, blood oxygen data, heart rate variability data, blood pressure data and preset thresholds, the user's health status is assessed to obtain health assessment data. The calculation of the user's heart rate and blood oxygen data based on the filtered PPG data and filtered accelerometer data includes the following specific calculation process: Filtering: S(f) = W(f)·P(f) In the formula, P(t,k) is the value of the PPG signal of length a collected at time t after Fourier transform at the k-th frequency point, N(t,k) is the acceleration signal, w represents the filter function, c represents the selected window size, and S represents the filtered frequency domain signal. Peak detection is performed on the filtered frequency domain signal, and the frequency of the peak is set to k. i Calculate heart rate (HR): HR=k i *60 The filtered frequency domain signal is then restored to the time domain using an inverse Fourier transform to obtain s(t,a): AC = max(s(t,a)) - min(s(t,a)) In the formula, AC is the AC component of the PPG signal, DC is the DC component of the signal, and R is the normalized infrared absorption ratio. Based on the preset standard value, linear regression was performed on R to obtain blood oxygen data; The specific formula for calculating the user's heart rate variability data based on the filtered ECG data is as follows: Filtering and R-peak detection of ECG data: In the formula, x(n) is the ECG time-domain signal sampled at time n, L is the signal length, and x norm For the normalized signal, E shannon It is the Shannon energy of the signal, E smooth To smooth the Shannon energy, locs(i) is the index of the R peak in the ECG signal. Representing the Hilbert transform, Z is the set of indices of the crossing zeros; ∈ is a very small constant used to avoid invalid inputs to the logarithmic function, RR i F is the time interval between adjacent R waves. s It is the sampling rate. It is a signal that has undergone Hilbert transform; Calculate health-related parameters: In the formula, N is the number of RR intervals in the calculation window. SDSD is the average RR value, SDSD is the standard deviation of the difference between adjacent normal heartbeat intervals, SDNN is the standard deviation of the normal heartbeat intervals, NN50 is the number of intervals between adjacent normal heartbeat intervals that exceed 50ms, and pNN50 is the percentage of NN50 in the total number of heartbeats. Analyze users' health status based on health-related parameters; The user's blood pressure data is calculated by combining the filtered PPG data and the filtered ECG data. The specific calculation formula is as follows: Select calculation window W foot With W dia The time interval between the R wave and the maximum first derivative point corresponding to the PPG data was calculated, and the diastolic time was extracted from the PPG signal. Combined with the user's own physiological parameters, a personalized blood pressure regression calculation model was then established for each user to obtain the blood pressure: In the formula, P(t) is the PPG signal, and t max 'The point where the maximum derivative of P(t) is calculated within the window, t foot W represents the time point at which the waveform begins during the PPG signal period. foot W is the time window that includes the beginning portion of the waveform. dia t is the time window that includes the diastolic phase of the signal. min The lowest point of the signal's diastolic phase, PTT dp t represents the pulse wave conduction time. R波 Dia represents the time when the R peak appears within the corresponding time window. t2 This refers to the diastolic time interval; The calculation process for the attitude data specifically includes: In the formula, p n v n and These are the position, velocity, and attitude rotation matrices of the n-system downloader, respectively; f n For accelerometer observations of the n-system; Let be the Earth's rotational angular velocity in the n-system; and These are the velocity and angular velocity of the n-frame relative to the e-frame in the n-frame; g n Let gravitational acceleration be the acceleration due to gravity in the n-system. These are the original observations of the gyroscope; Let n be the angular velocity of the n-system relative to the i-system in the b-system, where n is the navigation coordinate system, b is the vehicle coordinate system, and i is the geocentric inertial coordinate system. The specific calculation process for the location data is as follows: In the formula, N is the length of the calculation window, μ represents the signal mean, σ represents the standard deviation, K represents kurtosis, S represents skewness, R represents the root mean square, and x i The data returned by the accelerometer and gyroscope at time i is used to input the feature vector v into the ECOC classification model. A pre-trained classifier generates a 31-bit codeword, which is then matched against the closest class. In the formula, c represents the index of the action, d(·) is the Hamming distance function, and M c This represents the c-th row in the encoding matrix. This is the final output of the predicted localization result.

2. The indoor / outdoor health monitoring and positioning method according to claim 1, characterized in that, When a user needs rescue, a distress signal containing the user's current location data, heart rate data, blood oxygen data, heart rate variability data, and blood pressure data is sent to the cloud, and rescue is carried out based on the distress signal.

Images