External medical equipment positioning method based on mobile low-power-consumption Bluetooth base station

Abstract

The invention belongs to the technical field of medical equipment positioning in the range of an intelligent medical system, and provides an external medical equipment positioning method based on a mobile low-power-consumption Bluetooth base station, which comprises the following steps: judging whether an edge error phenomenon exists or not by acquiring historical positioning data and screening and marking edge error positioning; performing stability analysis on the marginal area, optimizing base station deployment in the fixed area, and constructing a correlation model through linear fitting or clustering of a positioning error and a signal value in the non-fixed area; when the current device is positioned in real time, whether the device is located in the marginal area is judged, base station weights are distributed according to the association model, and the position of the device is calculated through a weighted least square method, so that the problem of low positioning accuracy of multi-base-station coverage edges is solved, the method can adapt to stability and signal relations of different marginal areas, the positioning accuracy is improved, and the energy consumption is reduced. And the requirements of medical scenes on equipment positioning are met.

Description

Technical Field

[0001] This invention belongs to the field of medical device positioning technology within the scope of intelligent medical systems, specifically a method for locating external medical devices based on a mobile low-power Bluetooth base station. Background Technology

[0002] In the field of smart healthcare, real-time positioning of mobile intelligent terminals for external medical devices is crucial for improving diagnostic and treatment efficiency, asset turnover management, and emergency response speed. Mobile Bluetooth Low Energy (BLE) base stations have become the mainstream technology solution for positioning external devices in hospitals due to their advantages such as flexible deployment, low power consumption, and controllable cost. By coordinating multiple base stations to receive Bluetooth signals broadcast by device beacons, and combining RSSI (Received Signal Indicator) or TOF (Time-of-Flight) ranging technology to calculate device coordinates, a positioning accuracy of 1-1.5 meters can be achieved in everyday scenarios, basically meeting basic medical needs.

[0003] However, in signal transition zones such as the edges of multi-base station coverage areas (e.g., at the ends of corridors, corners of wards, and operating room entrances), existing positioning technologies face significant accuracy bottlenecks. These problems manifest in two main ways: First, single-base station positioning errors increase dramatically. Signal attenuation is severe in edge areas, resulting in generally low RSSI values ​​received by devices from single base stations. This leads to a sharp increase in distance conversion errors based on RSSI, compounded by base station coordinate drift, ultimately expanding the single-base station positioning error to 3-5 meters, which is completely unacceptable for the accuracy requirements of medical device positioning. Second, multi-base station joint positioning struggles to overcome signal reliability bottlenecks. Edge area signals are affected by multipath interference and electromagnetic interference, with RSSI fluctuation variance often exceeding 5dBm². The low proportion of direct TOF signals further reduces signal reliability. Even with multi-base station fusion algorithms, the large deviation in the input ranging data and the lack of optimization for edge area signal characteristics in existing fusion strategies prevent effective filtering of low-reliability data. Ultimately, the joint positioning accuracy of multi-base stations remains less than 2 meters, failing to support high-precision requirements such as rapid location of emergency equipment and real-time tracking of surgical instruments.

[0004] Therefore, the present invention provides a method for locating external medical devices based on a mobile low-power Bluetooth base station. Summary of the Invention

[0005] In order to overcome the shortcomings of the prior art, at least one technical problem raised in the background art is solved.

[0006] The technical solution adopted by this invention to solve its technical problem is: a method for locating external medical devices based on a mobile low-power Bluetooth base station, characterized in that it includes: By performing edge area feature recognition and positioning error analysis on multiple historical positioning data, edge error positioning is identified, and statistical analysis of edge error positioning is performed to determine whether edge error phenomena exist. If it exists, perform stability analysis on the edge region of the edge error location to determine whether the edge region corresponding to the edge error location is fixed. If it is fixed, optimize the base station deployment for the fixed edge region. If not fixed, perform correlation analysis between the signal value and the positioning error for edge error localization, construct a correlation model between the signal value and the positioning error based on the correlation analysis results, and determine the edge region judgment conditions; By comparing and analyzing the signal values ​​of each base station received by the device at the current positioning time with the edge area judgment conditions, it is determined whether the device is located in the edge area. If so, weights are assigned to each base station according to the correlation model between signal value and positioning error, and the final device positioning coordinates are calculated according to the weighted least squares method.

[0007] Furthermore, the method for identifying the edge error localization is as follows: Obtain multiple historical location data from the positioning system. For any given historical location: Calculate the difference between the two top-ranked signal values. If the difference between the two signal values ​​is close to the threshold and the top-ranked signal value meets the weak signal threshold, then it meets the characteristics of the edge region. Calculate the distance between the positioning coordinates of a single base station and the fused positioning coordinates of multiple base stations to obtain the positioning error of a single base station, and calculate the average error of a single base station. Calculate the distance between the fused coordinates of multiple base stations and the actual coordinates of the device to obtain the fused error of multiple base stations. If both the average error of a single station and the fusion error of multiple base stations meet the error determination criteria and conform to the characteristics of an edge area, then the historical positioning will be marked as edge error positioning.

[0008] Furthermore, the method for determining whether edge error exists is as follows: The proportion of edge error positioning in historical positioning is statistically analyzed. If the requirements are met, then edge error exists when positioning external medical devices based on mobile low-power Bluetooth base stations.

[0009] Furthermore, the method for determining whether the edge region corresponding to the edge error positioning is fixed is as follows: The hospital's location area is divided into spatial units. The number of times each spatial unit is marked as an edge area is counted, and the ratio of this to the total number of edge error locations is calculated to obtain the percentage of edge area occurrences. From all spatial units, select those with an edge region ratio greater than or equal to a preset ratio, and denot them as the high-ratio unit set; By statistically analyzing adjacent units in a high-proportion unit set, it can be determined whether the spatial distribution of high-proportion spatial units is spatially concentrated. If there is one or more high-proportion spatial units in the high-proportion unit set, and the spatial distribution of high-proportion spatial units is concentrated, then the edge region corresponding to the edge error positioning is fixed.

[0010] Furthermore, the method for determining whether the spatial distribution of the high-proportion spatial units is spatially concentrated is as follows: If the difference between the horizontal coordinates and the difference between the vertical coordinates of two spatial units are both less than or equal to 1, they are determined to be adjacent units. For each high-proportion unit in the high-proportion unit set, count the number of adjacent units in the high-proportion unit set; Add up the number of adjacent units of all high-percentage units and divide by 2 to get the total number of adjacent unit pairs. Calculate the proportion of adjacent units by comparing the total number of adjacent unit pairs with the theoretical maximum number of adjacent unit pairs. If the proportion of adjacent units is greater than or equal to the concentration threshold, the spatial distribution is concentrated; otherwise, the spatial distribution is dispersed.

[0011] Furthermore, the process of optimizing base station deployment in the fixed edge area includes: The spatial extent formed by high-proportion spatial units is marked as a fixed edge zone; The frequency of device occurrence in each space within a fixed edge zone is counted. Spaces with a device occurrence frequency that meets the requirements are marked as core activity areas for the devices, while those with a frequency that does not meet the requirements are marked as non-core activity areas for the devices. If the distance between the fixed edge area and the nearest base station meets the requirements and there are no other base stations in between, the defect type is insufficient base station density. In this case, additional edge micro base stations should be deployed between the fixed edge area and the nearest base station. If the base station is fixed in a non-core activity area for a long time, the defect type is that the base station location is deviated from the core area. The existing mobile base station should be adjusted to be near the geometric center of the core activity area of ​​the equipment.

[0012] Furthermore, the process of constructing the correlation model between the signal value and the positioning error includes: The strongest signal value received by the device in each edge error positioning and the multi-base station fusion positioning error are screened and standardized to obtain the signal value and positioning error; Using the signal value as the independent variable and the multi-base station fusion positioning error as the dependent variable, a linear fit is performed on the signal value and the positioning error, and the fitting equation is solved using the least squares method. Calculate the goodness of fit. If the goodness of fit satisfies the linear condition, then the signal value and the positioning error have a linear relationship; otherwise, the relationship is non-linear. If the signal value and the positioning error have a linear relationship, then the fitted equation is the correlation model between the signal value and the positioning error. If the relationship between the signal value and the positioning error is non-linear, clustering can be performed based on the positioning error and the signal value according to the edge error positioning to obtain multiple cluster groups.

[0013] Furthermore, the method for determining the edge region judgment condition is as follows: If the positioning error is linearly related to the signal value, then the edge region threshold obtained by substituting the acceptable error threshold into the fitting equation is the edge region judgment condition. If the positioning error and the signal value have a non-linear relationship, then the signal value range and positioning error range of each cluster group are statistically analyzed. The total range of signal values ​​for each cluster group is used as the threshold range for the edge region, i.e., the edge region judgment condition.

[0014] Furthermore, the calculation process for the final device positioning coordinates includes: Real-time acquisition of signal values ​​received by the device from each base station when the device is currently positioned; If the positioning error is linearly related to the signal value, and if the maximum signal value received by the device is less than or equal to the edge zone threshold, then the device is located in the edge zone. Substitute the signal values ​​of each base station received by the device into the fitting equation to obtain the expected error of each base station participating in the positioning. For any participating positioning base station, calculate the proportion of the reciprocal of the expected error to the sum of the reciprocals of the expected errors of all participating positioning base stations to obtain the weight of each participating positioning base station; Substituting the real-time coordinates, signal values, and weights of each participating positioning base station into the weighted least squares formula yields the final device positioning coordinates.

[0015] Furthermore, the calculation process for the final device positioning coordinates also includes: If the positioning error has a non-linear relationship with the signal value, and if the maximum signal value received by the current device is not within the threshold range of the edge area, then the device is located in the edge area. The signal values ​​of each base station received by the current device are matched with the signal value range of each cluster group to determine the cluster group to which each base station signal value belongs. Calculate the average positioning error of each cluster group, and the weight of each base station is the reciprocal of the average positioning error of each cluster group; Substituting the coordinates and weights of each base station into the weighted least squares formula, the final device positioning coordinates are obtained.

[0016] The beneficial effects of this invention are as follows: By filtering edge error positioning through historical data and constructing an association model by combining linear fitting or clustering, weights are accurately allocated during real-time positioning, improving the positioning accuracy of edge areas and meeting the needs of medical scenarios. Base station deployment is optimized for fixed edge areas, and linear / nonlinear relationships are adapted for non-fixed edge areas, adapting to different edge area characteristics. This solves the problem of poor adaptability to edge area characteristics in existing methods. Micro base stations are added or base station positions are adjusted as needed for fixed edge areas, avoiding blind deployment and reducing hardware costs. Weights are dynamically allocated through the association model for non-fixed edge areas, reducing invalid calculations, reducing energy consumption, and optimizing resource allocation. During real-time positioning, only signal features need to be collected, and the device location is quickly obtained through threshold comparison and weight calculation, resulting in a short response time. This adapts to the real-time tracking needs of medical devices and is applicable to different types of external medical devices and different hospital scenarios (corridors, wards, operating rooms, etc.). There is no need to develop separate algorithms for each scenario, reducing application costs. Attached Figure Description

[0017] The invention will now be further described with reference to the accompanying drawings.

[0018] Figure 1 This is a flowchart of the steps of the external medical device positioning method based on a mobile low-power Bluetooth base station according to the present invention. Figure 2 This is a logic diagram of the external medical device positioning method based on a mobile low-power Bluetooth base station as described in this invention. Detailed Implementation

[0019] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments.

[0020] Please see Figure 1 As shown, the present invention describes a method for locating external medical devices based on mobile low-power Bluetooth base stations. This invention primarily involves edge region feature identification and positioning error analysis of historical positioning data, filtering and marking edge error locations, determining the existence of edge error phenomena, and performing stability analysis on edge regions. For fixed areas, base station deployment is optimized; for non-fixed areas, a correlation model is constructed through linear fitting or clustering of positioning errors and signal values. When performing real-time positioning of the current device, it determines whether the device is located in an edge region, assigns base station weights based on the correlation model, and calculates the device location using weighted least squares. This solves the problem of low positioning accuracy at the edge of multi-base station coverage, improving positioning accuracy. Specifically, the method includes the following steps: Step 1: By performing edge area feature recognition and positioning error analysis on multiple historical positioning operations, edge error positioning is identified, and statistical analysis is performed on the edge error positioning to determine whether edge error phenomena exist. Please see Figure 2As shown, in step one, the edge error positioning refers to the historical positioning where there is a positioning error due to the device being located at the edge of multiple device coverage, where the edge of multiple device coverage is marked as the edge area; In step one, the identification method for edge error localization includes: It should be noted that the edge area is characterized by the absence of a dominant base station and weak and unstable signal. The logic for edge error localization and identification is: to simultaneously satisfy the edge area characteristics and the historical localization where both single base station localization error and multi-base station fusion error exist. The first point to clarify is that the verification process for edge region features includes: The system retrieves multiple historical positioning data, including: positioning timestamp, device ID, list of participating base stations (ID + real-time coordinates), single base station signal characteristics, single base station positioning coordinates, multi-base station fused positioning coordinates, and device motion status. For any given historical location: From the list of base stations participating in the positioning, filter out the base station with the highest signal value and extract its signal value; The strongest signal value is compared with the weak signal threshold. If the strongest signal value is less than or equal to the weak signal threshold, then the strongest signal received by the device meets the weak signal threshold. Among them, the weak signal threshold refers to the critical value of the strongest base station signal strength received by the device. When the strongest signal value is less than or equal to this threshold, the signal is determined to meet the weak signal characteristics. It is one of the core indicators of the edge area and is set according to the signal quality requirements of medical positioning. It should be noted that another key characteristic of the edge area is that the signal values ​​of multiple base stations are close, with no clearly dominant base station. Therefore, it is necessary to verify whether the signal difference between the two base stations with the strongest signals is less than the proximity threshold. Specifically: From the list of base stations participating in the positioning, select the two base stations with the highest signal values ​​and calculate the signal difference between them; If the signal difference is less than the proximity threshold, it satisfies the rule of no dominant base station, indicating that the signals of multiple base stations are close and conform to the characteristics of the edge area; The proximity threshold refers to the critical value of the signal difference between the two base stations with the strongest signals. When the signal difference between the two base stations is less than this threshold, it is determined that the signals of multiple base stations are close and there is no obvious dominant base station, which meets the characteristics of the edge area. It is set according to the signal distribution characteristics of the edge area. Secondly, it should be noted that the verification process for the existence of both single-base station positioning errors and multi-base station fusion errors includes: For any given historical location: Calculate the distance between the positioning coordinates of a single base station and the fused positioning coordinates of multiple base stations to obtain the positioning error of a single base station, and calculate the mean of all positioning errors of a single base station to obtain the average error of a single base station. The distance between the fused coordinates of multiple base stations and the actual coordinates of the devices is calculated to obtain the fusion error of the multiple base stations; If both the average error of a single base station and the fusion error of multiple base stations are greater than the acceptable error, then the problem is satisfied that both the positioning error of a single base station and the fusion error of multiple base stations exist. Among them, acceptable error refers to the maximum allowable value of single base station positioning error and multi-base station fusion positioning error. When both exceed this value, the error is judged to be out of standard. It is the core result feature of edge error positioning and is set according to the positioning accuracy requirements of medical scenarios. If the device is located at the edge of multi-base station coverage, and both single-base station positioning error and multi-base station fusion error exist, then the historical positioning will be marked as edge error positioning. In step one, the method for determining whether edge error exists is as follows: The proportion of edge error positioning in historical positioning is statistically analyzed and compared with the preset proportion. If it is greater than the preset proportion, then there is an edge error phenomenon when positioning external medical devices based on mobile low-power Bluetooth base stations. The preset percentage refers to the maximum allowable proportion of the number of edge error positioning to the total number of valid historical positioning. When the actual percentage is greater than this value, it is determined that the edge error phenomenon is common and subsequent optimization needs to be initiated. If the actual percentage is less than or equal to this value, the edge error is an occasional case and no large-scale adjustment is required. It is set according to the positioning accuracy requirements of the medical scenario. It should be noted that the purpose of edge error judgment is to accurately identify positioning errors caused by edge areas in historical data, determine whether the problem is widespread, provide data basis for subsequent targeted optimization, and avoid blindly investing resources. Step 2: If it exists, perform stability analysis on the edge region of the edge error location to determine whether the edge region corresponding to the edge error location is fixed. If it is fixed, optimize the base station deployment for the fixed edge region. In step two, the process of determining whether the edge region corresponding to the edge error positioning is fixed includes: The hospital's location area is divided into spatial units using a grid, and each spatial unit is identified by a unique ID. For example, the origin of the coordinate system is a fixed landmark in the hospital (such as the bottom of the central pillar in the outpatient hall). The x-axis is the direction along the length of the main corridor of the hospital (such as the direction from the outpatient hall to the inpatient department, with the east direction being the positive direction), and the y-axis is the direction along the width of the corridor (such as the direction from the north wall of the corridor to the south wall, with the north direction being the positive direction). Select a 1m×1m grid and plot the coordinates (x, y, z) of any physical location within the hospital. 实 y 实 ), which are mapped to the corresponding mesh cells according to the following rules: Grid x-coordinate (horizontal number): The physical x-coordinate is rounded down (ignoring the decimal part) and denoted as x. 格(e.g. x) 实 =5.3m→x 格 =5; x 实 =5.9m→x 格 =5); Grid y-coordinate (vertical number): The physical y-coordinate is rounded down and denoted as y. 格 (e.g., y) 实 =0.8m→y 格 =0; y 实 =1.2m→y 格 =1); Grid coverage area: numbered (x 格 y 格 The grid of ) corresponds to a physical space range of x∈[x 格 x 格 +1) meters, y∈[y 格 y 格 +1) meters (left closed, right open, to avoid duplicate coordinate assignment); For all edge error localizations, the spatial units to which the edge regions belong in each edge error localization are statistically analyzed. Among them, the edge area is the boundary area where the positioning device is located in the coverage of multiple base stations in edge error positioning, which is a physical area with weak signal, no dominant base station, and positioning error exceeding the standard. For any spatial unit: The frequency of edge regions is obtained by counting the number of times a spatial unit is marked as an edge region. The proportion of edge region occurrences in the total number of edge error localization attempts is obtained by statistically analyzing the frequency of edge region occurrences in spatial units. From all spatial units, select those with a proportion of edge regions greater than or equal to the proportion threshold, and denot them as the high proportion unit set; Among them, the proportion threshold refers to the minimum qualified value of the proportion of the edge area of ​​a spatial unit. Only when the proportion of the edge area of ​​a spatial unit is greater than or equal to this value will it be included in the high proportion unit set and regarded as a unit with a long-term high frequency of edge area occurrence. It is set according to the stability requirements of the medical scenario. Based on the physical location of spatial units, if the difference between the horizontal coordinates and the difference between the vertical coordinates of two spatial units are both less than or equal to 1, they are determined to be adjacent units. For each high-proportion unit in the high-proportion unit set, count the number of its neighboring units in the high-proportion unit set; Add up the number of adjacent units of all spatial units and divide by 2 to get the total number of adjacent unit pairs. Calculate the ratio of the total number of adjacent unit pairs to the theoretical maximum number of adjacent unit pairs to get the proportion of adjacent units. Among them, the theoretical maximum number of adjacent unit pairs is the maximum possible number of adjacent pairs in the high proportion of unit set. For example, if there are n units in the high proportion of unit set, the maximum possible number of adjacent pairs is n×(n-1) / 2. If the proportion of adjacent units is greater than or equal to the concentration threshold, it means that the high proportion of units are mostly adjacent and the spatial distribution is concentrated; otherwise, the spatial distribution is dispersed. The concentration threshold refers to the minimum acceptable value of the proportion of adjacent units. Only when the proportion of adjacent units in a high proportion set is greater than or equal to this value is the high proportion unit spatial distribution determined to be concentrated (continuous and in patches). If the proportion is less than this value, the spatial distribution is determined to be scattered (isolated and discontinuous). The threshold is set according to the definition of continuity in the medical scenario. It should be noted that if the edge area is fixed in a few consecutive areas for a long time, the root cause of the problem is insufficient base station coverage in these areas. If the edge area changes with the movement of base stations, equipment, or temporary obstruction, the root cause of the problem is signal fluctuation caused by dynamic factors. Based on the positioning accuracy requirements in medical scenarios, the criteria for evaluating equipment are as follows: If the edge region of one or more spatial units has a proportion greater than or equal to a preset proportion, and the spatial distribution of these high-proportion spatial units is concentrated, then the edge region corresponding to the edge error positioning is fixed. If the proportion of edge regions of all spatial units is less than the preset proportion, and the spatial distribution of spatial units is scattered, then the edge region corresponding to the edge error positioning is not fixed. If the edge region corresponding to the edge error location is fixed, the spatial range formed by the high proportion of spatial units is marked as the fixed edge region. For any fixed edge region: Statistically analyze the historical positioning data of all devices within the fixed edge area, calculate the frequency of device occurrence in each spatial unit, and mark the spatial units where the frequency of device occurrence is greater than or equal to 20% of the total number of positioning times as the core activity area of ​​the devices; By comparing the fixed edge area grid with the real-time coordinates of the base station, the type of coverage defect is determined, and optimization measures are implemented for the base station deployment. Specifically: If the distance between the fixed edge area and the nearest base station is greater than the threshold and there are no other base stations in between, the defect type is insufficient base station density. In this case, additional edge micro base stations should be deployed between the fixed edge area and the nearest base station. If the base station is fixed in a non-core activity area for a long time, while the fixed edge area is at the edge of the core activity area, resulting in weak signal at the edge of the core area, the defect type is that the base station is deviated from the core area. The existing mobile base station should be adjusted to be near the geometric center of the core activity area to ensure that the distance between the edge of the core area and the base station is less than the threshold. For example, assuming the corridor is determined to be a fixed edge area, and there are currently two mobile low-energy Bluetooth base stations in the corridor, with the following real-time coordinates (corresponding to the grid): |Base Station ID|Real-time Coordinates (Grid)|Corresponding Physical Location|Signal Coverage Radius (Unobstructed)|; |BaseA|Grid-20-00|x=20~21m,y=0~1m|5m (typical coverage area of ​​Bluetooth Low Energy base station)|; |BaseB|Grid-15-00|x=15~16m,y=0~1m|5m|; Defect 1: Insufficient base station density (distance between the fixed edge area and the nearest base station exceeds the threshold): The nearest grid in the fixed edge area is Grid-28-00 (x=28~29m), and the nearest base station is BaseA (x=20~21m). Straight-line distance calculation: Center x-coordinate of Grid-28-00 = 28.5m, Center x-coordinate of BaseA = 20.5m, Distance = 28.5 - 20.5 = 8m; Set the base station coverage threshold to 5m (the upper limit of the unobstructed coverage radius of the Bluetooth base station), 8m > 5m, and there are no other base stations between BaseA and Grid-28-00 (x = 21~28m); Deploy one edge micro base station (BaseC) in the signal transition zone between BaseA and the fixed edge area (x=24~25m, y=0~1m, corresponding to Grid-24-00): BaseC parameters: power consumption <50mW, battery life 6 months, coverage radius 3~4m; Post-deployment results: BaseA (20.5m) → BaseC (24.5m) → Fixed edge area (28.5m), with spacing of less than 5m, forming relay coverage. The signal strength of the fixed edge area increased from -88dBm to -80dBm, and the single base station ranging error decreased from 1.2m to 0.6m. Defect Judgment 2: Base station location deviates from the core activity area (weak signal at the edge of the core area): Geometric center of core activity area: x=(22+24) / 2=23m, y=(0+2) / 2=1m, corresponding to Grid-23-01; The nearest base station, BaseA, is located at Grid-20-00 (x=20.5m). The distance from the edge of the core activity area (Grid-23-01, x=23.5m) to BaseA is 23.5-20.5=3m (<5m, signal is acceptable). However, the distance from the eastern edge of the core activity area (Grid-24-00, x=24.5m) to BaseA is 4m, which is close to the threshold. Moreover, this location is close to the fixed edge area (x=28~30m), where the signal is prone to attenuation. Move BaseA from Grid-20-00 (x=20.5m) to Grid-23-00 (x=23.5m, y=0~1m, north side of the entrance to Ward 303) near the geometric center of the core activity area. After adjustment, the distance from the edge of the core activity area (Grid-24-00, x=24.5m) to BaseA is 1m, the signal strength is improved from -82dBm to -75dBm, and the ranging error is <0.5m; The distance from the fixed edge area (Grid-28-00, x=28.5m) to BaseA is 5m (exactly equal to the threshold). With the coverage of BaseC superimposed, the signal strength is stable at -78dBm, and the multi-base station fusion error is reduced from 2.3m to 1.7m, meeting the accuracy requirement of ≤2m in medical scenarios. It should be noted that the purpose of edge region stability analysis is to accurately locate the root cause of edge errors and provide a basis for subsequent optimization strategies. Step 3: If not fixed, perform correlation analysis on the signal value and positioning error of edge error positioning, construct a correlation model between signal value and positioning error based on the correlation analysis results, and determine the edge area judgment conditions; In step three, the process of correlating the signal values ​​for edge error localization with the localization error includes: Extract the positioning records for edge error positioning, including: edge error positioning ID, device ID, signal value, positioning error value, signal variance, and positioning timestamp; After standardizing the signal values ​​and positioning errors, we obtain the signal value-positioning error dataset; Using the maximum signal value received by the device as the independent variable and the multi-base station fusion positioning error as the dependent variable, a linear fit is performed on the signal value and the positioning error, and the fitting equation is solved using the least squares method. Calculate the goodness of fit. If the goodness of fit satisfies the linear condition, then the signal value and the positioning error have a linear relationship; otherwise, the relationship is non-linear. If the signal value and the positioning error have a linear relationship, then the fitted equation is the correlation model between the signal value and the positioning error. If the relationship between the signal value and the positioning error is non-linear, clustering is performed based on the positioning error and the signal value according to the edge error positioning, specifically as follows: The signal value and positioning error are standardized and then used as the core features; The number of clusters K is determined using the elbow method, with K starting from 1. The sum of squared errors (SSE) corresponding to each K is calculated, and the K-SSE curve is plotted. The optimal value of K is the point where the SSE drops sharply and then flattens out. K core features are randomly selected as initial cluster centers. The Euclidean distance between each core feature and the K cluster centers is calculated, and the core features are assigned to the clusters with the closest Euclidean distance. After all core features are assigned, the center of each cluster is recalculated. The assignment and update are repeated until the change in the cluster center is less than or equal to a preset threshold. The clustering ends and multiple cluster groups are finally obtained. In step three, the edge region determination condition is determined as follows: If the location error of the correlation analysis result is linearly related to the signal value, then: Based on the medical scenario, an acceptable error threshold is set. The signal value corresponding to the error being equal to the acceptable threshold is obtained from the fitted equation. The resulting edge region threshold is the edge region judgment condition. If the location error of the correlation analysis result has a non-linear relationship with the signal value, then: By calculating the original feature mean of each core feature group, the signal-error pattern of each cluster is clarified and mapped to specific partitions in the medical scenario: Group the data by cluster label and statistically analyze the signal value range and corresponding positioning error range for each cluster group. The total range of signal values ​​for each cluster group is marked as the threshold interval of the edge region, i.e., the edge region judgment condition; It should be noted that the purpose of building the correlation model and determining the dynamic threshold is to: by quantifying the relationship between signal values ​​and positioning errors, build a model and determine the dynamic threshold, so as to provide accurate judgment basis and weight allocation rules for subsequent real-time positioning, replacing the traditional fixed threshold and fixed weight. Step 4: By comparing and analyzing the signal values ​​of each base station received by the device at the current positioning time with the edge area judgment conditions, determine whether the device is located in the edge area. If so, assign weights to each base station according to the correlation model between signal value and positioning error, and calculate the final device positioning coordinates according to the weighted least squares method. In step four, the method for determining whether the device is located in the edge area includes: If the location error of the correlation analysis result is linearly related to the signal value, then: The base station weights are assigned based on the fitted equation. The core logic is that the fitted equation can predict the expected error of each base station. The smaller the expected error, the higher the reliability of the base station, and the greater its weight. The specific allocation process is as follows: The signal values ​​of each participating base station are collected in real time. The expected error of each base station is calculated using a fitting equation. Weights are then allocated according to the inverse proportional relationship between weight and error. The weight formula is as follows: ,in, Let i be the weight of the i-th base station. Let be the expected error of the i-th base station, and k be the total number of participating base stations; Substitute the real-time coordinates, signal values, and fitting weights of each base station into the weighted least squares formula to solve for the device coordinates: The objective function is Where x and y are the device coordinates, For base station coordinates, This is the distance measurement value; Solving the objective function yields the final device positioning coordinates; If the location error of the correlation analysis result has a non-linear relationship with the signal value, then: When the device is being located, the signal values ​​of each base station are collected in real time. If the maximum signal value received by the device is less than or equal to the edge zone threshold, the device is located in the edge zone. The signal values ​​of each base station received by the current device are matched with each cluster group to determine the cluster group to which each base station signal value belongs; Real-time weight allocation is performed based on the error magnitude of the cluster groups, specifically as follows: Calculate the average localization error of each cluster group, and the weight ratio is the reciprocal ratio of the average localization error of each cluster group; Substitute the coordinates, ranging values, and weights of each base station into the weighted least squares formula to obtain the final device positioning coordinates; It should be noted that the role of real-time positioning and weighted calculation is to apply the correlation model and dynamic threshold to actual positioning through real-time signal acquisition, edge area judgment, and dynamic weight allocation, thereby solving the core problems of large single base station error and insufficient accuracy of multi-base station fusion, and ensuring that the real-time positioning accuracy meets medical needs.

[0021] The technical solution and advantages of this application embodiment are as follows: By performing edge area feature identification and positioning error analysis on multiple historical positioning, edge error positioning is identified, and statistical analysis is performed on edge error positioning to determine whether edge error phenomenon exists. If it exists, stability analysis is performed on the edge area of ​​edge error positioning to determine whether the edge area corresponding to edge error positioning is fixed. If it is fixed, base station deployment optimization is performed on the fixed edge area. If it is not fixed, correlation analysis is performed on the signal value of edge error positioning and positioning error. Based on the correlation analysis results, a correlation model between signal value and positioning error is constructed, and edge area judgment conditions are determined. By comparing and analyzing the signal values ​​of each base station received by the device at the current positioning time with the edge area judgment conditions, it is determined whether the device is located in the edge area. If so, weights are assigned to each base station according to the correlation model between signal value and positioning error, and the final device positioning coordinates are calculated according to the weighted least squares method. This invention acquires historical positioning data, filters and marks edge error locations, and determines whether edge error phenomena exist. It performs stability analysis on edge areas, optimizes base station deployment in fixed areas, and constructs a correlation model for non-fixed areas through linear fitting or clustering of positioning errors and signal values. During real-time positioning, it determines whether the device is located in an edge area, assigns base station weights based on the correlation model, and calculates the device location using the weighted least squares method. This solves the problem of low positioning accuracy at the edge of multi-base station coverage, adapts to the stability and signal relationships of different edge areas, improves positioning accuracy, reduces energy consumption, and meets the device positioning needs of medical scenarios.

[0022] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for locating external medical devices based on a mobile low-power Bluetooth base station, characterized in that: include: By performing edge area feature recognition and positioning error analysis on multiple historical positioning data, edge error positioning is identified, and statistical analysis of edge error positioning is performed to determine whether edge error phenomena exist. If it exists, perform stability analysis on the edge region of the edge error location to determine whether the edge region corresponding to the edge error location is fixed. If it is fixed, optimize the base station deployment for the fixed edge region. If not fixed, perform correlation analysis between the signal value and the positioning error for edge error localization, construct a correlation model between the signal value and the positioning error based on the correlation analysis results, and determine the edge region judgment conditions; By comparing and analyzing the signal values ​​of each base station received by the device at the current positioning time with the edge area judgment conditions, it is determined whether the device is located in the edge area. If so, weights are assigned to each base station according to the correlation model between signal value and positioning error, and the final device positioning coordinates are calculated according to the weighted least squares method.

2. The method for locating external medical devices based on a mobile low-power Bluetooth base station according to claim 1, characterized in that: The method for identifying the edge error localization is as follows: Obtain multiple historical location data from the positioning system. For any given historical location: The difference between the two strongest signal values ​​received by the calculation device is considered. If the difference between the two signal values ​​is close to a threshold, and the device receives the strongest signal value while the weak signal value meets the weak signal threshold, then it meets the characteristics of an edge region. Calculate the distance between the positioning coordinates of a single base station and the fused positioning coordinates of multiple base stations to obtain the positioning error of a single base station, and calculate the average error of a single base station. Calculate the distance between the fused coordinates of multiple base stations and the actual coordinates of the device to obtain the fused error of multiple base stations. If both the average error of a single station and the fusion error of multiple base stations meet the error determination criteria and conform to the characteristics of an edge area, then the historical positioning will be marked as edge error positioning.

3. The method for locating external medical devices based on a mobile low-power Bluetooth base station according to claim 1, characterized in that: The method for determining whether edge error exists is as follows: The proportion of edge error positioning in historical positioning is statistically analyzed. If the requirements are met, then edge error exists when positioning external medical devices based on mobile low-power Bluetooth base stations.

4. The method for locating external medical devices based on a mobile low-power Bluetooth base station according to claim 1, characterized in that: The method for determining whether the edge region corresponding to the edge error positioning is fixed is as follows: The hospital's location area is divided into spatial units. The number of times each spatial unit is marked as an edge area is counted, and the ratio of this to the total number of edge error locations is calculated to obtain the percentage of edge area occurrences. From all spatial units, select those with an edge region ratio greater than or equal to a preset ratio, and denot them as the high-ratio unit set; By statistically analyzing adjacent spatial units in a high-proportion unit set, it can be determined whether the spatial distribution of high-proportion spatial units is spatially concentrated. If there is one or more high-proportion spatial units in the high-proportion unit set, and the spatial distribution of high-proportion spatial units is concentrated, then the edge region corresponding to the edge error positioning is fixed.

5. The method for locating external medical devices based on a mobile low-power Bluetooth base station according to claim 4, characterized in that: The method for determining whether the spatial distribution of the high-proportion spatial units is spatially concentrated is as follows: If the difference between the horizontal coordinates and the difference between the vertical coordinates of two spatial units are both less than or equal to 1, they are determined to be adjacent units. For each high-proportion unit in the high-proportion unit set, count the number of adjacent units in the high-proportion unit set; Add up the number of adjacent units of all high-percentage units and divide by 2 to get the total number of adjacent unit pairs. Calculate the proportion of adjacent units by comparing the total number of adjacent unit pairs with the theoretical maximum number of adjacent unit pairs. If the proportion of adjacent units is greater than or equal to the concentration threshold, the spatial distribution is concentrated; otherwise, the spatial distribution is dispersed.

6. The method for locating external medical devices based on a mobile low-power Bluetooth base station according to claim 5, characterized in that: The process of optimizing base station deployment in fixed edge areas includes: The spatial extent formed by high-proportion spatial units is marked as a fixed edge zone; The frequency of device occurrence in each space within a fixed edge zone is counted. Spaces with a device occurrence frequency that meets the requirements are marked as core activity areas for the devices, while those with a frequency that does not meet the requirements are marked as non-core activity areas for the devices. If the distance between the fixed edge area and the nearest base station meets the requirements and there are no other base stations in between, the defect type is insufficient base station density. In this case, additional edge micro base stations should be deployed between the fixed edge area and the nearest base station. If the base station is fixed in a non-core activity area for a long time, the defect type is that the base station location is deviated from the core area. The existing mobile base station should be adjusted to be near the geometric center of the core activity area of ​​the equipment.

7. The method for locating external medical devices based on a mobile low-power Bluetooth base station according to claim 1, characterized in that: The process of constructing the correlation model between the signal value and the positioning error includes: The strongest signal value received by the device in each edge error positioning and the multi-base station fusion positioning error are screened and standardized to obtain the signal value and positioning error; Using the signal value as the independent variable and the multi-base station fusion positioning error as the dependent variable, a linear fit is performed on the signal value and the positioning error, and the fitting equation is solved using the least squares method. Calculate the goodness of fit. If the goodness of fit satisfies the linear condition, then the signal value and the positioning error have a linear relationship; otherwise, the relationship is non-linear. If the signal value and the positioning error have a linear relationship, then the fitted equation is the correlation model between the signal value and the positioning error. If the relationship between the signal value and the positioning error is non-linear, clustering can be performed based on the positioning error and the signal value according to the edge error positioning to obtain multiple cluster groups.

8. The method for locating external medical devices based on a mobile low-power Bluetooth base station according to claim 7, characterized in that: The method for determining the edge region judgment condition is as follows: If the positioning error is linearly related to the signal value, then the edge region threshold obtained by substituting the acceptable error threshold into the fitting equation is the edge region judgment condition. If the positioning error and the signal value have a non-linear relationship, then the signal value range and positioning error range of each cluster group are statistically analyzed. The total range of signal values ​​for each cluster group is used as the threshold range for the edge region, i.e., the edge region judgment condition.

9. The method for locating external medical devices based on a mobile low-power Bluetooth base station according to claim 1, characterized in that: The calculation process for the final device positioning coordinates includes: Real-time acquisition of signal values ​​received by the device from each base station when the device is currently positioned; If the positioning error is linearly related to the signal value, and if the maximum signal value received by the device is less than or equal to the edge zone threshold, then the device is located in the edge zone. Substitute the signal values ​​of each base station received by the device into the fitting equation to obtain the expected error of each base station participating in the positioning. For any participating positioning base station, calculate the proportion of the reciprocal of the expected error to the sum of the reciprocals of the expected errors of all participating positioning base stations to obtain the weight of each participating positioning base station; Substituting the real-time coordinates, signal values, and weights of each participating positioning base station into the weighted least squares formula yields the final device positioning coordinates.

10. The method for locating external medical devices based on a mobile low-power Bluetooth base station according to claim 9, characterized in that: The calculation process for the final equipment positioning coordinates also includes: If the positioning error has a non-linear relationship with the signal value, and if the maximum signal value received by the current device is not within the threshold range of the edge area, then the device is located in the edge area. The signal values ​​of each base station received by the current device are matched with the signal value range of each cluster group to determine the cluster group to which each base station signal value belongs. Calculate the average positioning error of each cluster group, and the weight of each base station is the reciprocal of the average positioning error of each cluster group; Substituting the coordinates and weights of each base station into the weighted least squares formula, the final device positioning coordinates are obtained.

Images