Method and system for generating hierarchical management strategies for electronic fences based on 5G base station towers

Abstract

An embodiment of the present invention discloses a method and system for generating a hierarchical management strategy for an electronic fence based on a 5G base station tower. The embodiment of the present invention collects a multi-source heterogeneous monitoring data set covering the electronic fence warning area through a sensor array deployed at different monitoring points of the 5G base station tower, including environmental perception data and target motion data; performs spatiotemporal correlation fusion processing on the data set to generate a basic analysis data set, based on which the spatial relative relationship between the target motion data and the electronic fence boundary and the potential impact of the environmental perception data on the movement behavior of the target object are determined, and risk assessment information is generated; a hierarchical management strategy is formulated based on the risk assessment information, and by continuously tracking the behavioral evolution process of the target object in the electronic fence warning area, the strategy execution parameters are adjusted to achieve optimization iteration, thereby improving the accuracy and adaptability of electronic fence management.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of computer data processing technology, and specifically to a method and system for generating a hierarchical management strategy for an electronic fence based on a 5G base station tower. Background Art

[0002] Electronic fence management is a technical means of using electronic technology to define the boundaries of a specific area and monitor and manage target objects that enter or attempt to enter the area. It has a wide range of applications in many fields such as security, logistics, and intelligent transportation.

[0003] Traditional geo-fence management relies primarily on single monitoring devices, such as infrared sensors and cameras. These devices provide limited information, typically only detecting whether a target object is approaching the geo-fence boundary, but failing to fully perceive the surrounding environmental conditions. During risk assessment, existing technologies often consider only the distance between the target object and the geo-fence, ignoring the potential impact of environmental factors on the target object's movement, resulting in inaccurate risk assessment results. Traditional management strategies often employ fixed models that are unable to dynamically adjust to actual risk conditions. This lack of flexibility and adaptability makes it difficult to cope with complex and changing real-world scenarios. Summary of the Invention

[0004] The embodiments of the present invention provide a method and system for generating a hierarchical management strategy for an electronic fence based on a 5G base station tower.

[0005] In a first aspect, an embodiment of the present invention provides a method for generating a hierarchical management strategy for an electronic fence based on a 5G base station tower, which is applied to a hierarchical management strategy generation system for an electronic fence based on a 5G base station tower. The method comprises: performing real-time environmental perception and target detection through a sensor array deployed at different monitoring points of the 5G base station tower, collecting a multi-source heterogeneous monitoring data set covering the electronic fence warning area, wherein the multi-source heterogeneous monitoring data set includes environmental perception data reflecting the environmental state and target motion data characterizing the dynamic properties of the target object; performing spatiotemporal correlation fusion processing on the multi-source heterogeneous monitoring data set to generate a basic analysis data set; based on the basic analysis data set, determining the spatial relative relationship between the target motion data and the electronic fence boundary and the potential impact of the environmental perception data on the movement behavior of the target object; generating risk assessment information of the target object entering the electronic fence warning area based on the spatial relative relationship and the potential impact; formulating a hierarchical management strategy based on the risk assessment information, and adjusting the execution parameters of the hierarchical management strategy to achieve optimization iteration by continuously tracking the behavior evolution process of the target object in the electronic fence warning area.

[0006] In a second aspect, an embodiment of the present invention provides a hierarchical management strategy generation system for an electronic fence based on a 5G base station tower, comprising:

[0007] processor;

[0008] a storage device having a computer program stored thereon,

[0009] When the computer program is executed by the processor, the processor implements any of the methods for generating a hierarchical management strategy for an electronic fence based on a 5G base station tower.

[0010] An embodiment of the present invention provides a readable storage medium, which stores a program or instruction. When the program or instruction is executed by a processor, the steps of the method for generating a hierarchical management strategy for an electronic fence based on a 5G base station tower are implemented.

[0011] The embodiment of the present invention realizes the effective generation and dynamic optimization of the electronic fence hierarchical management strategy, significantly improving the accuracy, adaptability and efficiency of electronic fence management. A sensor array deployed at different monitoring points on the 5G base station tower is used to collect a multi-source heterogeneous monitoring data set, covering environmental perception data and target motion data, to achieve accurate monitoring of the environment and target object status in the electronic fence warning area; the multi-source heterogeneous monitoring data set is subjected to spatiotemporal correlation fusion processing to generate a basic analysis data set, eliminating redundancy and inconsistency in the data; based on the basic analysis data set, the spatial relative relationship between the target motion data and the electronic fence boundary and the potential impact of the environmental perception data on the target object's motion behavior are determined, and risk assessment information is generated, fully considering environmental factors and target object dynamics, making risk assessment more scientific and accurate, and being able to predict the risk of the target object entering the electronic fence warning area in advance; a hierarchical management strategy is formulated based on the risk assessment information, and targeted management measures can be taken according to different risk situations, thereby improving the effectiveness and pertinence of the management strategy. By continuously tracking the behavioral evolution of target objects within the electronic fence warning area and adjusting the execution parameters of the hierarchical management strategy, dynamic optimization and iteration of the management strategy is achieved, enabling it to adapt to changes in the target object's behavior in a timely manner, further enhancing the adaptability and flexibility of electronic fence management, effectively reducing safety risks, and improving resource utilization efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 A flowchart of a method for generating a hierarchical management strategy for an electronic fence based on a 5G base station tower provided in an embodiment of the present invention.

[0013] Figure 2 A schematic diagram of the basic structure of a hierarchical management strategy generation system for electronic fences based on 5G base station towers provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0014] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the embodiments of the present invention are further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0015] See also Figure 1 As shown in FIG, this figure is a flow chart of a method for generating a hierarchical management strategy for an electronic fence based on a 5G base station tower according to an embodiment of the present invention. This method can be applied to a hierarchical management strategy generation system for an electronic fence based on a 5G base station tower. Figure 1 As shown, the method may include steps 110 to 140.

[0016] Step 110: Real-time environmental perception and target detection are performed through sensor arrays deployed at different monitoring points on the 5G base station tower to collect a multi-source heterogeneous monitoring data set covering the electronic fence warning area. The multi-source heterogeneous monitoring data set includes environmental perception data reflecting the environmental status and target motion data characterizing the dynamic properties of the target object.

[0017] In an embodiment of the present invention, in the scenario of generating a hierarchical management strategy for an electronic fence, 5G base station towers are deployed in a set area, and sensor arrays are installed at different monitoring points. These sensors work continuously to perform real-time environmental perception and target detection. For example, for environmental perception data, sensors can collect meteorological information such as temperature, humidity, wind speed, and wind direction, as well as data such as topography and electromagnetic interference intensity. These data can reflect the environmental status of the electronic fence warning area. For target motion data, sensors will capture dynamic properties such as the position, speed, and direction of movement of the target object. For example, if the target object is a drone, the sensor will record information such as the drone's position coordinates, driving speed, and driving direction at different times. In this way, the sensor array collects a multi-source heterogeneous monitoring data set covering the electronic fence warning area.

[0018] Step 120: Perform spatiotemporal correlation fusion processing on the multi-source heterogeneous monitoring data set to generate a basic analysis data set.

[0019] Next, the collected multi-source heterogeneous monitoring data set needs to be processed. Because this data comes from different sensors and has different temporal and spatial characteristics, spatiotemporal correlation fusion processing is required. For example, data collected by different sensors at the same time must be correlated based on their spatial position relationship. One sensor is located on the east side of the electronic fence, and the other is located on the west side. The environmental data and target object data collected by them at the same time need to be integrated based on their spatial location. At the same time, for data collected at different times, temporal continuity and change trends must be considered. Through spatiotemporal correlation fusion processing, the various data in the multi-source heterogeneous monitoring data set are integrated and optimized, and redundant and erroneous information is removed. Ultimately, a basic analysis data set is generated. This basic analysis data set contains processed environmental perception data and target motion data, which is more suitable for subsequent analysis and calculation.

[0020] Step 130: Based on the basic analysis data set, determine the spatial relative relationship between the target motion data and the electronic fence boundary and the potential impact of the environmental perception data on the movement behavior of the target object; generate risk assessment information of the target object entering the electronic fence warning area based on the spatial relative relationship and the potential impact.

[0021] As you can understand, the generated basic analysis dataset requires two aspects of analysis. First, the spatial relationship between the target's motion data and the geo-fence boundary is determined. For example, using the target's location coordinates and the preset geo-fence boundary coordinates, the vertical distance and horizontal offset between the target and the geo-fence boundary at different times are calculated to determine the target's positional changes relative to the geo-fence boundary. Second, the potential impact of environmental perception data on the target's motion behavior is analyzed. For example, strong winds may affect the target's speed, while complex terrain may alter its direction of movement. By comprehensively considering these spatial relationships and potential impacts, a risk assessment of the target's entry into the geo-fence warning area can be generated. For example, if the target is approaching the geo-fence boundary and environmental factors favor its proximity to the fence, the risk of the target entering the fence can be determined to be high.

[0022] In one embodiment, determining the spatial relative relationship between the target motion data and the electronic fence boundary and the potential impact of the environmental perception data on the motion behavior of the target object based on the basic analysis data set includes:

[0023] Step 131: performing trajectory feature extraction processing on the target motion data in the basic analysis data set to obtain a position coordinate sequence, a motion direction angle sequence, and a motion speed sequence of the target object at continuous time sampling points.

[0024] In this step, the target motion data in the basic analysis data set is processed. Trajectory features are extracted from the target motion data using a set algorithm. For example, a data mining algorithm is used to extract the position coordinate sequence of the target object at continuous time sampling points from the target motion data. This means that the specific position coordinates of the target object at each time sampling point are recorded. At the same time, a motion direction angle sequence is extracted, that is, the motion direction angle of the target object at each time sampling point, and a motion rate sequence is also extracted, that is, the motion speed of the target object at each time sampling point.

[0025] Step 132: Perform spatial distance calculation based on the position coordinate sequence and the preset coordinate set of the electronic fence boundary to generate the vertical distance value and horizontal offset between the target object and the electronic fence boundary at each time sampling point, and combine the vertical distance value and horizontal offset into a spatial relative relationship.

[0026] Then, the extracted target object position coordinate sequence and the preset coordinate set of the electronic fence boundary are used to calculate the spatial distance. For each time sampling point, the vertical distance value between the target object and the electronic fence boundary is calculated. This vertical distance value reflects the degree of proximity of the target object to the electronic fence boundary in the vertical direction. At the same time, the horizontal offset is calculated, which represents the horizontal offset of the target object relative to the electronic fence boundary. For example, if the target object gradually approaches one side of the electronic fence in the horizontal direction, the horizontal offset will change accordingly. By combining the vertical distance value and the horizontal offset, the spatial relative relationship between the target object and the electronic fence boundary is obtained. This relationship can intuitively show the change in the position of the target object relative to the electronic fence in space.

[0027] Step 133: performing environmental factor classification processing on the environmental perception data in the basic analysis data set to obtain an environmental impact factor set including meteorological factor data, terrain factor data, and electromagnetic interference factor data.

[0028] Specifically, environmental perception data is divided according to different factors to obtain a set of environmental influencing factors. This includes meteorological factors, such as the aforementioned temperature, humidity, wind speed, and direction; terrain factors, such as the slope and aspect of the terrain; and electromagnetic interference factors, such as the intensity and frequency of electromagnetic interference. This classification process enables a clearer analysis of the potential impact of different environmental factors on the target object's motion behavior.

[0029] Step 134: Based on the correlation between each environmental impact factor in the environmental impact factor set and the motion rate sequence and motion direction angle sequence in the target motion data, generate a correction coefficient sequence of the meteorological factor data for the motion rate, a deflection angle sequence of the terrain factor data for the motion direction angle, and a fluctuation amplitude sequence of the electromagnetic interference factor data for the position coordinates, and combine the correction coefficient sequence, the deflection angle sequence, and the fluctuation amplitude sequence into the potential impact of the environmental perception data on the motion behavior of the target object.

[0030] Further analysis is performed based on the correlation between each environmental factor in the set of environmental impact factors and the motion velocity and motion direction angle sequences in the target motion data. For meteorological factor data, its relationship with the motion velocity sequence is analyzed to generate a sequence of correction coefficients for the meteorological factor data to the motion velocity. For example, strong winds may affect the target object's motion velocity. Analysis can yield correction coefficients for the motion velocity at different wind speeds. For terrain factor data, its relationship with the motion direction angle sequence is studied to generate a sequence of deflection angles for the terrain factor data to the motion direction angle. For example, on sloping terrain, the target object's motion direction may deflect. Analysis can determine the deflection angles at different slopes and downslopes. For electromagnetic interference factor data, its relationship with position coordinates is analyzed to generate a sequence of fluctuation amplitudes for the electromagnetic interference factor data to the position coordinates. Because electromagnetic interference can cause errors in the sensor's measurement of the target object's position coordinates, analysis can yield the fluctuation amplitudes for the position coordinates at different electromagnetic interference intensities and frequencies. Combining these three sequences reveals the potential impact of environmental perception data on the target object's motion behavior.

[0031] In a preferred embodiment, step 134 includes:

[0032] Step 1341: Extract wind speed data and wind direction data from the meteorological factor data in the environmental influencing factor set, perform time series matching on the wind speed data and the motion rate sequence in the target motion data, calculate the ratio of the wind speed data to the motion rate at each time sampling point, and generate a wind speed ratio sequence.

[0033] In this step, wind speed data and wind direction data are extracted from the meteorological factor data of the environmental influence factor set.Then the wind speed data and the motion rate sequence in the target motion data are carried out time series matching, that is, the wind speed data and the motion rate at the same time sampling point are corresponding.Then, the ratio of wind speed data and motion rate on each time sampling point is calculated to generate a wind speed rate ratio sequence.For example, if at a certain time sampling point, the wind speed is a certain value and the motion rate of the target object is another value, the ratio of the two is calculated, and the ratio of all time sampling points is combined to form a wind speed rate ratio sequence, which can reflect the relative influence of wind speed on the motion rate of the target object.

[0034] Step 1342: Based on the angle difference between the wind speed ratio sequence and the wind direction data and the motion direction angle sequence, a meteorological impact correction model is constructed, and the motion rate sequence is weighted point by point through the meteorological impact correction model to generate a correction coefficient sequence of the meteorological factor data for the motion rate.

[0035] Based on the generated wind speed ratio sequence and the angle difference between the wind direction data and the motion direction angle sequence, a meteorological impact correction model is constructed. This model considers the impact of the ratio of wind speed to motion rate and the angle between wind direction and motion direction on the target object's motion rate. This model performs point-by-point weighting on the motion rate sequence, adjusting each value in the motion rate sequence based on the model's calculation results to generate a sequence of correction coefficients for the meteorological factor data on the motion rate. For example, if the angle between the wind direction and the motion direction is large and the wind speed is high, the model will make corresponding corrections to the motion rate based on these factors, resulting in a correction coefficient for each time sampling point.

[0036] Step 1343: Extract the slope data and slope direction data from the terrain factor data in the environmental impact factor set, perform spatial correlation processing on the slope data and the motion direction angle sequence in the target motion data, calculate the motion direction angle change corresponding to different slope intervals, and generate a slope direction change sequence.

[0037] Slope and aspect data are extracted from the terrain factor data in the environmental impact factor set. The slope data is spatially correlated with the motion direction angle sequence in the target motion data, taking into account the changes in the motion direction angle of the target object as it moves on terrain with different slopes. The changes in the motion direction angle corresponding to different slope intervals are calculated. For example, the slope is divided into different intervals, and for each interval, the changes in the motion direction angle of the target object as it moves within that slope interval are analyzed. These changes are combined to generate a slope direction change sequence, which can reflect the impact of terrain slope on the target object's motion direction.

[0038] Step 1344: Construct a terrain influence deflection model by combining the slope data with the azimuth deviation value of the motion direction angle sequence, perform angle compensation processing on the motion direction angle sequence through the terrain influence deflection model, and generate a deflection angle sequence of the terrain factor data to the motion direction angle.

[0039] By combining the aspect data with the azimuth deviation values ​​of the motion direction angle sequence, a terrain-influenced deflection model is constructed. This model considers the impact of the azimuth deviation between the aspect and the motion direction on the target object's motion direction. This model applies angle compensation to the motion direction angle sequence, adjusting the motion direction angle based on the model's calculations to generate a deflection angle sequence based on the terrain factor data. For example, if the azimuth deviation between the aspect and the motion direction is large, the model will compensate the motion direction angle accordingly, resulting in the deflection angle for each time sampling point.

[0040] Step 1345: Extract the interference intensity data and interference frequency data from the electromagnetic interference factor data in the environmental impact factor set, perform time domain correlation processing on the interference intensity data and the position coordinate sequence in the target motion data, calculate the position coordinate deviation when the interference intensity exceeds the preset threshold, and generate an interference position deviation sequence.

[0041] Interference intensity data and interference frequency data are extracted from the electromagnetic interference factor data in the environmental impact factor set. The interference intensity data is then correlated with the position coordinate sequence in the target motion data in the time domain, analyzing the impact of the interference intensity on the target object's position coordinates at different time points. When the interference intensity exceeds a preset threshold, the deviation of the position coordinates is calculated and combined to generate an interference position deviation sequence. For example, if the interference intensity exceeds the preset threshold at a certain time point, the target object's position coordinates measured by the sensor may deviate. This deviation is recorded, and the deviations at all time points constitute the interference position deviation sequence.

[0042] Step 1346: Construct an electromagnetic influence fluctuation model based on the ratio of the interference frequency data to the sampling frequency of the position coordinate sequence, perform amplitude modulation processing on the interference position deviation sequence through the electromagnetic influence fluctuation model, and generate a fluctuation amplitude sequence of the electromagnetic interference factor data to the position coordinates.

[0043] Based on the ratio of the interference frequency data to the sampling frequency of the position coordinate sequence, an electromagnetic influence fluctuation model is constructed. This model considers the impact of the relationship between the interference frequency and the sampling frequency on the position coordinate deviation. This model applies amplitude modulation to the interference position deviation sequence. Specifically, each value in the interference position deviation sequence is adjusted based on the model's calculation results to generate a sequence of fluctuation amplitudes of the electromagnetic interference factor data on the position coordinates. For example, if the interference frequency is high and its ratio to the sampling frequency reaches a certain level, the model will amplitude modulate the interference position deviation sequence accordingly, resulting in the fluctuation amplitude at each time sampling point.

[0044] In an optional embodiment, generating risk assessment information of the target object entering the electronic fence warning area according to the spatial relative relationship and the potential impact includes:

[0045] Step 135: Based on the vertical distance value and the horizontal offset in the spatial relative relationship, calculate the decay rate of the vertical distance value over time and the cumulative offset of the horizontal offset over time.

[0046] The calculation is performed based on the vertical distance value and horizontal offset in the spatial relative relationship obtained previously. For the vertical distance value, its decay rate over time is calculated. This decay rate reflects the speed at which the target object approaches the boundary of the electronic fence in the vertical direction. For example, if the vertical distance value gradually decreases over a period of time, its decay rate can be obtained through the set calculation method. For the horizontal offset, its cumulative offset over time is calculated, that is, the horizontal offset of each time sampling point is accumulated to obtain the total offset. This cumulative offset can show the degree of horizontal offset of the target object relative to the boundary of the electronic fence.

[0047] Step 136: Analyze the correction coefficient sequence, deflection angle sequence, and fluctuation amplitude sequence in the potential influence, and extract the minimum correction coefficient in the correction coefficient sequence, the maximum deflection angle in the deflection angle sequence, and the average fluctuation amplitude in the fluctuation amplitude sequence.

[0048] Analyze the potential impact correction coefficient sequence, deflection angle sequence, and fluctuation amplitude sequence. Find the minimum correction coefficient from the correction coefficient sequence. This minimum correction coefficient indicates the maximum impact of meteorological factors on the target object's motion velocity. Find the maximum deflection angle from the deflection angle sequence, which reflects the maximum impact of terrain factors on the target object's motion direction. Calculate the average fluctuation amplitude from the fluctuation amplitude sequence. This average fluctuation amplitude reflects the average level of electromagnetic interference influence on the target object's position coordinates.

[0049] Step 137: If the attenuation rate is greater than the preset attenuation threshold and the cumulative offset is less than the preset offset threshold, and the minimum correction coefficient is less than the preset correction threshold, the maximum deflection angle is greater than the preset deflection angle threshold, and the average fluctuation amplitude is greater than the preset fluctuation threshold, then it is determined that the target object has a first risk entry trend, and the first risk assessment algorithm is called to generate the first risk assessment sub-information.

[0050] The calculated decay rate, cumulative offset, minimum correction coefficient, maximum deflection angle, and average fluctuation amplitude are evaluated. If the decay rate exceeds a preset decay threshold, it indicates that the target object is rapidly approaching the geo-fence boundary in the vertical direction. If the cumulative offset is less than the preset offset threshold, it indicates that the target object's horizontal deviation is within an acceptable range. If the minimum correction coefficient is less than the preset correction threshold, it means that meteorological factors have a significant impact on the target object's movement rate. If the maximum deflection angle is greater than the preset deflection angle threshold, it indicates that terrain factors have significantly changed the target object's movement direction. If the average fluctuation amplitude is greater than the preset fluctuation threshold, it indicates that electromagnetic interference factors have a significant impact on the target object's position coordinates. When these conditions are met simultaneously, the target object is determined to have a first risk entry trend. At this point, the first risk assessment algorithm is invoked, which calculates and analyzes the relevant data to generate first risk assessment sub-information, which contains relevant risk information about the target object under this risk trend.

[0051] In a preferred technical solution, the calling of the first risk assessment algorithm to generate the first risk assessment sub-information includes:

[0052] Step 1371: Input the vertical distance value and the horizontal offset in the spatial relative relationship into the spatial feature processing layer of the first risk assessment algorithm, perform exponential decay function fitting processing on the vertical distance value, and obtain a vertical distance attenuation curve.

[0053] The vertical distance values ​​and horizontal offsets from the spatial relative relationship are input into the spatial feature processing layer of the first risk assessment algorithm. In this layer, an exponential decay function is fitted to the vertical distance values. Using a pre-defined fitting algorithm, the vertical distance values ​​are matched to the exponential decay function to produce a vertical distance decay curve, which visually demonstrates the changing trend of vertical distance values ​​over time.

[0054] Step 1372: Perform sliding window summation processing on the horizontal offset to obtain a horizontal offset accumulation curve.

[0055] The horizontal offset is summed using a sliding window. Using a sliding window approach, a certain number of adjacent horizontal offsets are selected at each time point and summed. As time passes, the position of the sliding window is continuously updated, resulting in a series of summed results. These results are then connected to form a cumulative horizontal offset curve, which reflects the cumulative horizontal offset of the target object.

[0056] Step 1373: Input the correction coefficient sequence, deflection angle sequence and fluctuation amplitude sequence in the potential impact into the environmental feature processing layer of the first risk assessment algorithm, perform minimum value filtering on the correction coefficient sequence, and obtain a correction coefficient minimum value sequence.

[0057] The correction coefficient sequence, deflection angle sequence, and fluctuation amplitude sequence from the potential impact are input into the environmental feature processing layer of the first risk assessment algorithm. In this layer, the correction coefficient sequence undergoes minimum filtering. This minimum filtering algorithm selects the minimum value in the correction coefficient sequence at each time point to form a minimum correction coefficient sequence. This sequence highlights the instances where meteorological factors have the greatest impact on the target object's motion velocity.

[0058] Step 1374: Perform peak detection processing on the deflection angle sequence to obtain a deflection angle peak sequence.

[0059] Perform peak detection on the deflection angle sequence. Using a peak detection algorithm, find all peak points in the deflection angle sequence and combine the deflection angle values ​​of these peak points to obtain a deflection angle peak sequence. This sequence reflects the time and degree of the greatest impact of terrain factors on the target object's motion direction.

[0060] Step 1375: Perform mean calculation on the fluctuation amplitude sequence to obtain the fluctuation amplitude mean.

[0061] Perform mean calculation on the fluctuation amplitude sequence. All values ​​in the fluctuation amplitude sequence are added together and then divided by the length of the sequence to obtain the mean fluctuation amplitude. This mean reflects the average level of impact of electromagnetic interference factors on the position coordinates of the target object, providing comprehensive information on the impact of electromagnetic interference for risk assessment.

[0062] Step 1376: Input the vertical distance attenuation curve, horizontal offset cumulative curve, correction coefficient minimum value sequence, deflection angle peak value sequence and fluctuation amplitude mean into the fusion layer of the first risk assessment algorithm for spatiotemporal feature association modeling to generate a first risk feature matrix containing a spatial risk vector and an environmental risk vector.

[0063] The vertical distance attenuation curve, horizontal offset accumulation curve, correction coefficient minimum value sequence, deflection angle peak value sequence, and fluctuation amplitude mean obtained above are input into the fusion layer of the first risk assessment algorithm. In this fusion layer, spatiotemporal feature correlation modeling is performed, which considers the temporal and spatial relationships between these features. Using a predefined modeling algorithm, these features are integrated and correlated to generate a first risk feature matrix consisting of a spatial risk vector and an environmental risk vector. The spatial risk vector reflects the spatial risk between the target object and the boundary of the electronic fence, while the environmental risk vector reflects the impact of environmental factors on the risk of the target object entering the electronic fence.

[0064] Step 1377: Call the output layer of the first risk assessment algorithm to perform risk probability prediction processing on the first risk feature matrix to generate first risk assessment sub-information including a first risk probability value, a risk occurrence time window, and coordinates of a risk coverage area.

[0065] The output layer of the first risk assessment algorithm is called to perform risk probability prediction processing on the first risk feature matrix. The output layer uses the risk probability prediction algorithm to calculate the probability that the target object has a first risk entry trend based on the information in the first risk feature matrix, thereby obtaining a first risk probability value. Simultaneously, a time window for risk occurrence is determined, representing the time period during which the target object may enter the electronic fence. The coordinates of the risk coverage area are also determined, representing the area potentially affected by the target object entering the electronic fence. This information is combined to generate first risk assessment sub-information, including the first risk probability value, the risk occurrence time window, and the coordinates of the risk coverage area.

[0066] Step 138: If the attenuation rate is less than or equal to the preset attenuation threshold or the cumulative offset is greater than or equal to the preset offset threshold, or the minimum correction coefficient is greater than or equal to the preset correction threshold, the maximum deflection angle is less than or equal to the preset deflection angle threshold, and the average fluctuation amplitude is less than or equal to the preset fluctuation threshold, it is determined that the target object has a second risk entry trend, and the second risk assessment algorithm is called to generate second risk assessment sub-information.

[0067] If the decay rate is less than or equal to the preset decay threshold, it indicates that the target object is approaching the geo-fence boundary slowly or not approaching it vertically. If the cumulative offset is greater than or equal to the preset offset threshold, it indicates that the target object's horizontal offset exceeds the acceptable range. If the minimum correction coefficient is greater than or equal to the preset correction threshold, it means that meteorological factors have little impact on the target object's movement rate. If the maximum deflection angle is less than or equal to the preset deflection angle threshold, it indicates that terrain factors have little impact on the target object's movement direction. If the average fluctuation amplitude is less than or equal to the preset fluctuation threshold, it indicates that electromagnetic interference factors have little impact on the target object's position coordinates. If any of these conditions are met, the target object is determined to have a second risk entry trend. At this point, the second risk assessment algorithm is invoked, which calculates and analyzes the relevant data to generate second risk assessment sub-information, which contains relevant risk information about the target object under this risk trend.

[0068] Step 139: The first risk assessment sub-information and the second risk assessment sub-information are integrated to generate risk assessment information including risk trend type, risk occurrence probability, and risk impact scope.

[0069] The first and second risk assessment sub-information are fused. The relevant information from the two sub-information is integrated and processed using a pre-defined fusion algorithm. The risk trend type is extracted, indicating whether the target object's risk trend for entering the electronic fence is the first or second risk entry trend. The risk occurrence probability is calculated, taking into account the likelihood of the target object entering the electronic fence under both risk trends. The risk impact range is determined, combining the areas potentially affected by the target object under both risk trends. This combination of information generates risk assessment information that includes the risk trend type, risk occurrence probability, and risk impact range.

[0070] Step 140: Formulate a hierarchical management strategy based on the risk assessment information, and adjust the execution parameters of the hierarchical management strategy to achieve optimization iteration by continuously tracking the behavior evolution process of the target object within the electronic fence warning area.

[0071] Develop a hierarchical management strategy based on the generated risk assessment information. Develop different levels of management strategies based on the risk trend type, risk probability, and risk impact scope. For example, if the risk probability is high, develop a stricter management strategy; if the risk impact scope is large, increase corresponding management resources. After developing the strategy, continuously track the behavioral evolution of the target object within the electronic fence warning area. Use monitoring equipment deployed within the electronic fence warning area to obtain real-time behavioral data of the target object. Based on this data, adjust the execution parameters of the hierarchical management strategy, such as the strategy execution frequency and resource allocation ratio, to achieve optimized iteration of the hierarchical management strategy, enabling it to better respond to behavioral changes of the target object.

[0072] In one design approach, formulating a hierarchical management strategy based on the risk assessment information includes:

[0073] Step 141: parse the risk trend type, risk occurrence probability and risk impact range in the risk assessment information, and extract the policy template identifier corresponding to the risk trend type, the policy strength parameter corresponding to the risk occurrence probability and the policy coverage parameter corresponding to the risk impact range.

[0074] Parse the risk assessment information. Extract the risk trend type from the risk assessment information. Based on the different risk trend types, search for the corresponding policy template identifier. The policy template identifier points to the preset policy template. Different risk trend types correspond to different policy templates. These templates contain the basic management policy framework. At the same time, extract the policy strength parameter corresponding to the risk probability. The policy strength parameter determines the strictness of the management policy. The higher the risk probability, the greater the policy strength parameter. Also extract the policy coverage parameter corresponding to the risk impact range. The policy coverage parameter indicates the area that the management policy needs to cover. The larger the risk impact range, the larger the policy coverage parameter.

[0075] Step 142: Based on the policy template identifier, a corresponding basic management policy framework is retrieved from a preset policy template library. The basic management policy framework includes a policy execution subject, a policy execution process, and policy resource configuration items.

[0076] Based on the extracted policy template identifier, the corresponding basic management policy framework is retrieved from the preset policy template library. The preset policy template library stores a variety of different policy templates, each corresponding to a risk trend type. The basic management policy framework includes the policy execution entity (i.e., the department or personnel responsible for executing the management policy); the policy execution process, which specifies the steps and sequence for executing the management policy; and the policy resource configuration items, including the required human and material resources.

[0077] Step 143: adjusting the step interval and execution frequency of the policy execution process in the basic management policy framework according to the policy strength parameter, and generating a management policy framework after strength adjustment.

[0078] Adjust the policy execution process within the basic management policy framework based on the policy strength parameter. Regarding the step intervals in the policy execution process, a higher policy strength parameter results in shorter step intervals, meaning more compact management policy execution. Regarding execution frequency, a higher policy strength parameter results in a higher execution frequency, meaning the management policy will be executed more frequently. These adjustments create a management policy framework with adjusted strengths that better aligns with current risk conditions.

[0079] Step 144: Generate a management policy framework after coverage adjustment according to the resource allocation ratio and resource scheduling priority of the policy resource configuration items in the management policy framework after the policy coverage parameter adjustment strength is adjusted.

[0080] Adjust the policy resource configuration items in the strength-adjusted management policy framework based on the policy coverage parameter. Regarding resource allocation ratio, a larger policy coverage parameter indicates a higher proportion of resources allocated to areas with a larger risk impact. Regarding resource scheduling priority, a larger policy coverage parameter indicates a higher priority for resource scheduling in areas with a larger risk impact. These adjustments generate a coverage-adjusted management policy framework, ensuring that the management policy covers the appropriate areas and allocates resources appropriately.

[0081] Step 145: Perform policy verification on the management policy framework after the overlay adjustment, and identify policy conflict items and policy missing items in the management policy framework after the overlay adjustment.

[0082] Conduct policy validation on the adjusted management policy framework. By analyzing and examining the policy execution process and resource configuration items, identify any policy conflicts and gaps. Policy conflicts may include circular dependencies between execution steps or conflicting resource requirements. Policy gaps may include the lack of corresponding management measures in certain areas or the lack of appropriate response strategies for certain risk situations.

[0083] In one technical solution, the performing of policy verification on the management policy framework after the overlay adjustment to identify policy conflict items and policy missing items in the management policy framework after the overlay adjustment includes:

[0084] Step 1451: extract all policy execution steps in the management policy framework after coverage adjustment, and construct a policy step directed graph, where the nodes of the policy step directed graph represent policy execution steps, and the directed edges represent the execution order relationship between the steps.

[0085] Extract all policy execution steps from the adjusted management policy framework and construct a directed graph of policy steps. In this directed graph, each node represents a policy execution step, and directed edges indicate the execution order between steps. For example, if step A must be executed before step B, there is a directed edge from node A to node B. By constructing a directed graph of policy steps, the relationships between policy execution steps can be intuitively displayed.

[0086] Step 1452: Determine whether there is a circular dependency path in the policy step directed graph. If there is a circular dependency path, mark the policy execution steps constituting the circular dependency path as policy conflict items.

[0087] The constructed directed graph of policy steps is examined to see if there are any circular dependency paths. A circular dependency path means that some policy execution steps form a closed loop, preventing them from executing in the normal order. If a circular dependency path exists, the policy execution steps that constitute this path are marked as policy conflicts, as this circular dependency prevents the management policy from executing properly.

[0088] Step 1453: Extract the resource requirement parameters in the policy execution step, compare the resource requirement parameters of different policy execution steps within the same time window, and if the sum of the resource requirement parameters exceeds the preset resource threshold, mark the corresponding policy execution step as a policy conflict item.

[0089] Resource requirement parameters are extracted from the policy execution steps. These parameters represent the amount of resources required for each policy execution step. The resource requirement parameters of different policy execution steps within the same time window are compared and summed. If the sum of the resource requirement parameters exceeds the preset resource threshold, it indicates that the resource demand within the time window is too high to meet the requirements of all policy execution steps. The corresponding policy execution step is marked as a policy conflict.

[0090] Step 1454: parse the spatial coordinate set corresponding to the risk impact range in the risk assessment information, check whether the policy execution steps in the management policy framework after coverage adjustment include execution actions covering all sub-areas in the spatial coordinate set, and if there are uncovered sub-areas, determine that there are policy missing items.

[0091] Parse the spatial coordinate set corresponding to the risk impact range in the risk assessment information. This set represents the area potentially affected by the target object entering the geo-fence. Check the policy execution steps in the adjusted management policy framework to see if they include actions covering all sub-areas in the spatial coordinate set. If any sub-areas are uncovered, the management policy does not manage all potentially affected areas, and a policy entry is determined to be missing.

[0092] Step 1455: Check whether the policy execution steps in the management policy framework after the adjustment include all preset response measures for the risk trend type. If there are preset response measures that are not included, it is determined that there are policy missing items.

[0093] Check the policy execution steps in the adjusted management policy framework to see if they include all pre-defined responses for the risk trend type. Different risk trend types may have different pre-defined responses. For example, high-risk trends may have more stringent monitoring and interception measures. If the policy execution steps include any pre-defined responses that are not included, the management policy is incomplete and is considered to have policy omissions.

[0094] Step 146: Prioritize the policy conflict items to obtain a priority ranking result; supplement the policy missing items by adding policy execution steps that match the risk impact range to obtain a policy missing item supplement result.

[0095] Prioritize the marked policy conflict items. Determine their priority order based on their impact on management policy execution, and obtain the prioritization results. For policy missing items, supplementary processing is performed. Based on the risk impact scope, matching policy execution steps are added. For example, for uncovered sub-areas, corresponding monitoring and management steps are added to obtain the policy missing item supplementation results.

[0096] Step 147: Generate a hierarchical management policy by combining the priority sorting result and the policy missing item supplementation result.

[0097] Combine the results of the prioritization process with the results of the policy gap supplementation process. Based on the prioritization process, adjust and optimize policy conflicts to ensure the proper execution of the management policy. Based on the results of the policy gap supplementation process, refine the implementation steps of the management policy to ensure that it covers all potentially affected areas and addresses all risk scenarios. Combine this adjusted and supplemented information to generate a hierarchical management strategy.

[0098] In the next step, the execution parameters of the hierarchical management strategy are adjusted to achieve optimization iteration by continuously tracking the behavior evolution of the target object within the electronic fence warning area, including:

[0099] Step 1481: Continuously collect real-time behavior data of the target object through a dynamic monitoring sensor array deployed in the electronic fence warning area, wherein the real-time behavior data includes real-time position coordinates, real-time movement direction and real-time movement speed.

[0100] A dynamic monitoring sensor array deployed within the electronic fence's warning area continuously collects real-time behavioral data on the target object. These sensor arrays monitor the target object's position, direction of movement, and speed in real time. For example, a high-precision positioning sensor acquires the target object's real-time position coordinates, a direction sensor measures the target object's real-time direction of movement, and a speed sensor records the target object's real-time speed. Combining this data yields real-time behavioral data including real-time position coordinates, real-time direction of movement, and real-time speed.

[0101] Step 1482: Compare the real-time behavior data with the preset behavior benchmark data in the hierarchical management strategy, and calculate the behavior deviation characteristics between the real-time behavior data and the preset behavior benchmark data.

[0102] The collected real-time behavior data is compared with the preset behavioral baseline data in the hierarchical management strategy. The preset behavioral baseline data represents the normal behavior range of the target object, which is set when the hierarchical management strategy is formulated. It includes a baseline position coordinate sequence, a baseline motion direction sequence, and a baseline motion rate sequence. Using a predefined comparison algorithm, the difference between the real-time behavior data and the preset behavioral baseline data is calculated to generate a behavioral deviation signature. This behavioral deviation signature reflects the degree of deviation between the target object's actual behavior and the expected behavior.

[0103] As a design idea, the real-time behavior data is compared with the preset behavior benchmark data in the hierarchical management strategy, and the behavior deviation characteristics between the real-time behavior data and the preset behavior benchmark data are calculated, including:

[0104] Step 14821: Perform data optimization processing on the real-time position coordinates, real-time movement direction and real-time movement rate in the real-time behavior data to obtain optimized real-time position coordinates, optimized real-time movement direction and optimized real-time movement rate.

[0105] Data optimization processing is performed on the real-time location coordinates, real-time movement direction, and real-time movement rate in the real-time behavior data. Data optimization processing can include operations such as noise removal and data smoothing. For example, for real-time location coordinates, a filtering algorithm is used to remove possible measurement errors to obtain more accurate optimized real-time location coordinates. For real-time movement direction and real-time movement rate, a smoothing algorithm is used to reduce data fluctuations to obtain the optimized real-time movement direction and optimized real-time movement rate.

[0106] Step 14822: Extract the reference position coordinate sequence, reference movement direction sequence and reference movement rate sequence from the preset behavior reference data from the hierarchical management strategy, and perform the same optimization processing on the reference position coordinate sequence, reference movement direction sequence and reference movement rate sequence as that on the optimized real-time data to obtain an optimized reference position coordinate sequence, an optimized reference movement direction sequence and an optimized reference movement rate sequence.

[0107] The hierarchical management strategy extracts the baseline position coordinate sequence, baseline motion direction sequence, and baseline motion rate sequence from the pre-set behavioral baseline data. These baseline data undergo the same optimization processing as the real-time data, including noise removal and data smoothing. This process yields optimized baseline position coordinate sequences, optimized baseline motion direction sequences, and optimized baseline motion rate sequences, ensuring that the real-time data and baseline data are compared under the same optimization conditions.

[0108] Step 14823: Calculate the Euclidean distance between the optimized real-time position coordinates and the optimized reference position coordinate sequence at the corresponding time sampling points to generate a position deviation sequence; calculate the angular difference between the optimized real-time motion direction and the optimized reference motion direction sequence at the corresponding time sampling points to generate a direction deviation sequence; calculate the absolute difference between the optimized real-time motion rate and the optimized reference motion rate sequence at the corresponding time sampling points to generate a rate deviation sequence.

[0109] For the optimized real-time position coordinates and optimized reference position coordinate sequences, their Euclidean distances at corresponding time sampling points are calculated. The Euclidean distance reflects the spatial distance between two positions. Combining the Euclidean distances at each time sampling point generates a position deviation sequence. For the optimized real-time motion direction and optimized reference motion direction sequences, their angular differences at corresponding time sampling points are calculated. This angular difference reflects the degree of deviation in motion direction, forming a direction deviation sequence. For the optimized real-time motion rate and optimized reference motion rate sequences, their absolute differences at corresponding time sampling points are calculated to obtain a rate deviation sequence, which represents the difference in motion rate.

[0110] Step 14824: Perform feature fusion processing on the position deviation sequence, direction deviation sequence, and speed deviation sequence to generate a behavioral deviation feature including a deviation mean, a deviation variance, and a deviation change trend.

[0111] The position deviation sequence, direction deviation sequence, and velocity deviation sequence are subjected to feature fusion processing. The information from these three sequences is integrated using a predefined fusion algorithm. The mean deviation is calculated, representing the average level of behavioral deviation; the variance is calculated, reflecting the degree of dispersion of behavioral deviation; and the deviation trend is analyzed to understand how behavioral deviation changes over time. This combination of information generates a behavioral deviation signature that includes the mean deviation, variance, and trend.

[0112] Step 1483: Based on the behavior deviation feature, sensitive execution parameters are screened out from the execution parameter set of the hierarchical management strategy, where the sensitive execution parameters are execution parameters whose impact on the behavior deviation feature exceeds a preset impact threshold.

[0113] Based on the behavioral deviation characteristics obtained, sensitive execution parameters are screened from the hierarchical management policy's execution parameter set. The execution parameter set includes parameters such as the policy activation threshold, resource scheduling period, monitoring frequency, and response delay. Using a predefined screening algorithm, the impact of each execution parameter on the behavioral deviation characteristics is calculated. If the impact of a particular execution parameter on the behavioral deviation characteristics exceeds a preset impact threshold, it is marked as a sensitive execution parameter. Adjustment of these sensitive execution parameters has a significant impact on the target object's behavioral deviation.

[0114] As another design idea, the method of screening out sensitive execution parameters from the execution parameter set of the hierarchical management strategy based on the behavioral deviation characteristics includes:

[0115] Step 14831: extract all execution parameters in the execution parameter set of the hierarchical management policy, wherein the execution parameters include a policy start threshold, a resource scheduling period, a monitoring frequency parameter, and a response delay parameter.

[0116] All execution parameters are extracted from the execution parameter set of the hierarchical management strategy. These execution parameters include the policy start threshold, which determines the conditions for the start of management policy execution; the resource scheduling cycle, which stipulates the time interval between resource allocation and scheduling; the monitoring frequency parameter, which determines the frequency of data collection by the monitoring equipment; and the response delay parameter, which represents the response time of the management strategy to changes in the behavior of the target object.

[0117] Step 14832: Calculate the Spearman rank correlation coefficient between the deviation mean in the behavioral deviation feature and the parameter value of each execution parameter to generate a first correlation coefficient set; calculate the Kendall rank correlation coefficient between the deviation variance in the behavioral deviation feature and the parameter value of each execution parameter to generate a second correlation coefficient set; perform weighted summation on the correlation coefficients in the first correlation coefficient set and the second correlation coefficient set to generate a comprehensive correlation coefficient set.

[0118] Calculate the Spearman rank correlation coefficient between the deviation mean in the behavioral deviation feature and the parameter value of each execution parameter, which reflects the rank correlation between the deviation mean and the execution parameter value. Combine the correlation coefficients of each execution parameter to generate a first correlation coefficient set. At the same time, calculate the Kendall rank correlation coefficient between the deviation variance in the behavioral deviation feature and the parameter value of each execution parameter, which reflects the rank correlation between the deviation variance and the execution parameter value to form a second correlation coefficient set. Perform weighted summation on the correlation coefficients in the first correlation coefficient set and the second correlation coefficient set, that is, assign a certain weight to each correlation coefficient, and then add them together to generate a comprehensive correlation coefficient set, which comprehensively considers the correlation between the deviation mean and the deviation variance and the execution parameters.

[0119] Step 14833: Filter out target execution parameters whose absolute values ​​of comprehensive correlation coefficients are greater than a preset correlation threshold from the comprehensive correlation coefficient set, and mark the target execution parameters as candidate sensitive execution parameters.

[0120] Target execution parameters whose absolute value of the comprehensive correlation coefficient exceeds a preset correlation threshold are selected from the set of comprehensive correlation coefficients. The preset correlation threshold is a pre-set value used to determine whether the correlation between the execution parameter and the behavioral deviation signature is sufficiently strong. If the absolute value of the comprehensive correlation coefficient of an execution parameter exceeds the preset correlation threshold, it is marked as a candidate sensitive execution parameter, which may have a significant impact on the behavioral deviation signature.

[0121] Step 14834: Analyze the sensitivity of the parameter value change of the candidate sensitive execution parameter to the behavioral deviation feature, calculate the behavioral deviation feature change corresponding to the unit change of the parameter value, and generate a sensitivity index.

[0122] Further analysis is performed on candidate sensitive execution parameters, and the change in the behavioral deviation characteristic corresponding to a unit change in their parameter value is calculated to obtain a sensitivity index. For example, when the value of a candidate sensitive execution parameter increases by one unit, the change in the behavioral deviation characteristic is observed and the change amount is calculated. This sensitivity index can reflect the degree of influence of the candidate sensitive execution parameter on the behavioral deviation characteristic. A larger sensitivity index indicates that the execution parameter is more sensitive to the behavioral deviation characteristic.

[0123] Step 14835: Filter out candidate sensitive execution parameters whose sensitivity index is greater than a preset sensitivity threshold as sensitive execution parameters.

[0124] The sensitivity indicators of candidate sensitive execution parameters are screened, and those with sensitivity indicators exceeding a preset sensitivity threshold are selected as sensitive execution parameters. The preset sensitivity threshold is a pre-set standard used to determine whether the impact of candidate sensitive execution parameters is significant enough. Screened sensitive execution parameters are those whose impact on behavioral deviation characteristics exceeds the preset impact threshold. Adjusting these parameters can more effectively optimize the hierarchical management strategy.

[0125] Step 1484: Adjust the values ​​of sensitive execution parameters according to the deviation direction and deviation degree of the behavioral deviation feature to generate an adjusted execution parameter set.

[0126] Adjust the values ​​of sensitive execution parameters based on the direction and degree of deviation of the behavioral deviation signature. If the direction of the behavioral deviation signature indicates that the target object's behavior deviates significantly from the normal range and the degree of deviation is high, then adjust the sensitive execution parameters accordingly. For example, increase the policy activation threshold, shorten the resource scheduling period, increase the monitoring frequency parameter, or reduce the response delay parameter. Through these adjustments, a set of adjusted execution parameters is generated, enabling the hierarchical management strategy to better respond to changes in the target object's behavior.

[0127] Step 1485: Update the adjusted execution parameter set to the hierarchical management strategy, complete the optimization iteration of the hierarchical management strategy, and continue to track the behavioral evolution process of the target object, repeat the step of continuously collecting the real-time behavior data of the target object through the dynamic monitoring sensor array deployed in the electronic fence warning area until the value of the sensitive execution parameter is adjusted according to the deviation direction and deviation degree of the behavioral deviation characteristics to generate the adjusted execution parameter set.

[0128] The adjusted execution parameter set is updated into the hierarchical management strategy, achieving iterative optimization of the hierarchical management strategy. The updated hierarchical management strategy enables more effective management based on the latest behavioral data of the target object. The target object's behavioral evolution is then continuously tracked, repeating the steps from collecting real-time behavioral data to adjusting the values ​​of sensitive execution parameters. Through this iterative cycle, the hierarchical management strategy is continuously optimized to adapt to changes in the target object's behavior.

[0129] As a non-limiting embodiment, it also includes: collecting the actual behavior data of the target object after the hierarchical management strategy optimization iteration and the electronic fence boundary state data through the feedback monitoring sensor array deployed in the electronic fence warning area, the actual behavior data includes the optimized position coordinate sequence, the optimized motion direction angle sequence and the optimized motion rate sequence; performing spatiotemporal difference analysis on the actual behavior data and the target motion data before the optimization iteration, calculating the deviation convergence rate of the optimized position coordinate sequence and the original position coordinate sequence, the angular stability of the optimized motion direction angle sequence and the original motion direction angle sequence, and the rate fluctuation coefficient of the optimized motion rate sequence and the original motion rate sequence, to generate a strategy execution effect feature set; and analyzing the boundary state. A correlation analysis is performed between the boundary alert intensity sequence and the boundary response delay sequence in the data to generate boundary response time-efficiency characteristics; the strategy execution effect feature set and the boundary response time-efficiency characteristics are integrated to construct a strategy effect evaluation model, and the strategy effect evaluation model is used to comprehensively score the hierarchical management strategy after optimization iteration to generate a strategy effect evaluation result; the strategy effect evaluation result is compared with the preset effect benchmark, and the difference between the effect score value and the benchmark score value, the ratio of the risk control efficiency to the benchmark efficiency, and the ratio of the resource consumption rate to the benchmark consumption rate are calculated to generate a strategy effect deviation vector; based on the strategy effect deviation vector, the spatial relative relationship calculation weight and potential impact analysis coefficient in the risk assessment information generation process are adjusted.

[0130] In this embodiment, a feedback monitoring sensor array deployed within the electronic fence warning area collects actual target object behavior data and electronic fence boundary status data after optimization iterations of the hierarchical management strategy. The actual behavior data includes an optimized position coordinate sequence, an optimized motion direction angle sequence, and an optimized motion rate sequence. These data reflect the actual behavior of the target object after the optimization iterations. A spatiotemporal difference analysis is performed between the actual behavior data and the target motion data before the optimization iterations. The deviation convergence rate between the optimized position coordinate sequence and the original position coordinate sequence is calculated. This rate indicates how quickly the target object's position converges to its normal position after the optimization iterations. The angular stability between the optimized motion direction angle sequence and the original motion direction angle sequence is calculated, reflecting the stability of the target object's motion direction after the optimization iterations. The rate fluctuation coefficient between the optimized motion rate sequence and the original motion rate sequence is calculated, reflecting the fluctuation of the target object's motion rate after the optimization iterations. These indicators are combined to generate a strategy execution effect feature set. A correlation analysis is performed between the boundary warning intensity sequence and the boundary response delay sequence in the boundary status data. Using a pre-defined analysis algorithm, the correlation between the two is identified to generate a boundary response time-effect feature. The strategy execution effect feature set is integrated with the boundary response timeliness feature to construct a strategy effectiveness evaluation model. This model comprehensively considers strategy execution effect and boundary response timeliness, and comprehensively scores the optimized hierarchical management strategy after iteration to generate a strategy effectiveness evaluation result. The strategy effectiveness evaluation result is compared with the preset performance benchmark, and the difference between the performance score and the benchmark score is calculated. This difference reflects the difference between the actual and expected performance of the hierarchical management strategy. The ratio of risk control efficiency to benchmark efficiency is calculated, which represents the difference between the actual and expected performance of the hierarchical management strategy in controlling risk. The ratio of resource consumption rate to benchmark consumption rate is calculated, which reflects the difference between the actual and expected performance of the hierarchical management strategy in resource consumption. These ratios are combined to generate a strategy effectiveness deviation vector. Based on the strategy effectiveness deviation vector, the spatial relative relationship calculation weights and potential impact analysis coefficients in the risk assessment information generation process are adjusted to improve the accuracy of the risk assessment information, thereby further optimizing the hierarchical management strategy.

[0131] As a non-limiting embodiment, after adjusting the execution parameters of the hierarchical management strategy to achieve optimization iteration, it also includes: collecting the optimized hierarchical management strategy set of multiple target objects existing simultaneously in the electronic fence warning area, the real-time behavior evolution data of each target object and the resource capacity data of the electronic fence warning area through the multi-target collaborative monitoring module deployed on the 5G base station tower, the optimized hierarchical management strategy set includes the risk assessment information and execution parameters corresponding to each target object, and the real-time behavior evolution data includes the current position coordinates, motion trend vector and risk status label of each target object; performing spatial clustering processing on the real-time behavior evolution data of multiple target objects, identifying target object groups with behavioral interactions, and generating target association groups and risk coupling coefficients of each target in the group; and combining the execution parameters of each target object in the optimized hierarchical management strategy set. The matching degree between the row parameters and the resource capacity data is calculated, the resource demand vector of each strategy is extracted, the Euclidean distance between the resource demand vector and the resource capacity data is calculated, and a resource supply and demand deviation matrix is ​​generated; based on the risk coupling coefficient of the target association group and the resource supply and demand deviation matrix, a multi-objective collaborative scheduling model is constructed and the hierarchical management strategy execution parameters of each target object are dynamically adjusted to generate collaborative scheduling parameters; the collaborative scheduling parameters are injected into the hierarchical management strategy of each target object, and the resource scheduling cycle and response delay parameters in the strategy execution parameter set are updated to generate a multi-objective collaborative optimization management strategy; the real-time behavior evolution data and resource capacity data of each target object during the execution of the multi-objective collaborative optimization management strategy are continuously tracked, and the steps of spatial clustering the real-time behavior evolution data of multiple target objects are repeatedly performed to generate a multi-objective collaborative optimization management strategy.

[0132] In this embodiment, a multi-target collaborative monitoring module deployed on a 5G base station tower collects data related to multiple target objects simultaneously within the electronic fence warning area. The collected data includes an optimized hierarchical management strategy set, which contains risk assessment information and execution parameters corresponding to each target object; real-time behavioral evolution data for each target object, including its current location coordinates, motion trend vector, and risk status label; and resource capacity data for the electronic fence warning area. Spatial clustering is performed on the real-time behavioral evolution data of multiple target objects. Using a spatial clustering algorithm, target objects with behavioral interactions are grouped together based on their locations and motion trends, generating target-related groups. Simultaneously, a risk coupling coefficient is calculated for each target within the group. This coefficient represents the degree of mutual risk influence between the target objects within the group. The matching degree between the execution parameters in the optimized hierarchical management strategy set and the resource capacity data is calculated, and the resource requirement vector for each strategy is extracted, representing the amount of resources required by the hierarchical management strategy for each target object. The Euclidean distance between the resource demand vector and the resource capacity data is calculated and combined to generate a resource supply and demand deviation matrix, which reflects the difference between the resource demand of each target object and the resource capacity of the electronic fence warning area. Based on the risk coupling coefficient of the target association group and the resource supply and demand deviation matrix, a multi-objective collaborative scheduling model is constructed. This model comprehensively considers the risk coupling relationship and resource supply and demand between target objects, dynamically adjusts the execution parameters of the hierarchical management strategy for each target object, and generates collaborative scheduling parameters. The collaborative scheduling parameters are injected into the hierarchical management strategy of each target object, and the resource scheduling cycle and response delay parameters in the strategy execution parameter set are updated to generate a multi-objective collaborative optimization management strategy. The real-time behavior evolution data and resource capacity data of each target object are continuously tracked during the execution of the multi-objective collaborative optimization management strategy. The steps from spatial clustering processing to generating the multi-objective collaborative optimization management strategy are repeatedly executed to continuously optimize the multi-objective collaborative management strategy to enable it to better cope with the situation of multiple target objects.

[0133] The embodiment of the present invention realizes the effective generation and dynamic optimization of the electronic fence hierarchical management strategy, significantly improving the accuracy, adaptability and efficiency of electronic fence management. A sensor array deployed at different monitoring points on the 5G base station tower is used to collect a multi-source heterogeneous monitoring data set, covering environmental perception data and target motion data, to achieve accurate monitoring of the environment and target object status in the electronic fence warning area; the multi-source heterogeneous monitoring data set is subjected to spatiotemporal correlation fusion processing to generate a basic analysis data set, eliminating redundancy and inconsistency in the data; based on the basic analysis data set, the spatial relative relationship between the target motion data and the electronic fence boundary and the potential impact of the environmental perception data on the target object's motion behavior are determined, and risk assessment information is generated, fully considering environmental factors and target object dynamics, making risk assessment more scientific and accurate, and being able to predict the risk of the target object entering the electronic fence warning area in advance; a hierarchical management strategy is formulated based on the risk assessment information, and targeted management measures can be taken according to different risk situations, thereby improving the effectiveness and pertinence of the management strategy. By continuously tracking the behavioral evolution of target objects within the electronic fence warning area and adjusting the execution parameters of the hierarchical management strategy, dynamic optimization and iteration of the management strategy is achieved, enabling it to adapt to changes in the target object's behavior in a timely manner, further enhancing the adaptability and flexibility of electronic fence management, effectively reducing safety risks, and improving resource utilization efficiency.

[0134] See also Figure 2 As shown in FIG. 1 , the figure is a schematic diagram of the basic structure of a hierarchical management strategy generation system 200 for an electronic fence based on a 5G base station tower provided by an embodiment of the present invention. The hierarchical management strategy generation system 200 for an electronic fence based on a 5G base station tower includes:

[0135] Processor 201;

[0136] a storage device 202 having a computer program 2020 stored thereon;

[0137] When the computer program 2020 is executed by the processor 201, the processor 201 implements any of the methods for generating a hierarchical management strategy for an electronic fence based on a 5G base station tower.

[0138] Based on the above, a readable storage medium is provided, on which a program or instruction is stored. When the program or instruction is executed by a processor, the steps of the above method are implemented.

[0139] It should be noted that the various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Reference can be made to the common and similar parts between the various embodiments. For the systems or devices disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple, and the relevant parts can be referred to the method description.

Claims

1. A method for generating a hierarchical management strategy for an electronic fence based on a 5G base station tower, characterized in that: The method comprises: Real-time environmental perception and target detection are performed through sensor arrays deployed at different monitoring points on 5G base station towers, collecting a multi-source heterogeneous monitoring data set covering the electronic fence warning area. The multi-source heterogeneous monitoring data set includes environmental perception data reflecting the environmental state and target motion data representing the dynamic properties of the target object; Performing spatiotemporal correlation fusion processing on the multi-source heterogeneous monitoring data set to generate a basic analysis data set; Based on the basic analysis data set, determining the spatial relative relationship between the target motion data and the electronic fence boundary and the potential impact of the environmental perception data on the target object's motion behavior; generating risk assessment information of the target object entering the electronic fence warning area based on the spatial relative relationship and the potential impact; A hierarchical management strategy is formulated based on the risk assessment information, and by continuously tracking the behavioral evolution of the target object in the electronic fence warning area, the execution parameters of the hierarchical management strategy are adjusted to achieve optimization iteration: the real-time behavior data of the target object is continuously collected by a dynamic monitoring sensor array deployed in the electronic fence warning area, and the real-time behavior data includes real-time position coordinates, real-time movement direction and real-time movement speed; the real-time behavior data is compared with the preset behavior benchmark data in the hierarchical management strategy, and the behavior deviation characteristics between the real-time behavior data and the preset behavior benchmark data are calculated; based on the behavior deviation characteristics, sensitive execution targets are screened out from the execution parameter set of the hierarchical management strategy. The sensitive execution parameter is an execution parameter whose influence on the behavior deviation feature exceeds a preset influence threshold; the value of the sensitive execution parameter is adjusted according to the deviation direction and deviation degree of the behavior deviation feature to generate an adjusted execution parameter set; the adjusted execution parameter set is updated to the hierarchical management strategy to complete the optimization iteration of the hierarchical management strategy, and the behavior evolution process of the target object is continuously tracked, and the steps of continuously collecting the real-time behavior data of the target object through the dynamic monitoring sensor array deployed in the electronic fence warning area to adjusting the value of the sensitive execution parameter according to the deviation direction and deviation degree of the behavior deviation feature to generate an adjusted execution parameter set are repeatedly executed.

2. The method for generating a hierarchical management strategy for an electronic fence based on a 5G base station tower according to claim 1, wherein: The determining, based on the basic analysis data set, the spatial relative relationship between the target motion data and the electronic fence boundary and the potential impact of the environmental perception data on the motion behavior of the target object includes: Performing trajectory feature extraction processing on the target motion data in the basic analysis data set to obtain a position coordinate sequence, a motion direction angle sequence, and a motion rate sequence of the target object at continuous time sampling points; Performing spatial distance calculation based on the position coordinate sequence and a preset coordinate set of the electronic fence boundary to generate a vertical distance value and a horizontal offset between the target object and the electronic fence boundary at each time sampling point, and combining the vertical distance value and the horizontal offset into a spatial relative relationship; Performing environmental factor classification processing on the environmental perception data in the basic analysis data set to obtain an environmental impact factor set including meteorological factor data, terrain factor data, and electromagnetic interference factor data; Based on the correlation between each environmental impact factor in the environmental impact factor set and the motion rate sequence and motion direction angle sequence in the target motion data, a correction coefficient sequence of the meteorological factor data on the motion rate, a deflection angle sequence of the terrain factor data on the motion direction angle, and a fluctuation amplitude sequence of the electromagnetic interference factor data on the position coordinates are generated. The correction coefficient sequence, the deflection angle sequence, and the fluctuation amplitude sequence are combined into the potential impact of the environmental perception data on the motion behavior of the target object.

3. The method for generating a hierarchical management strategy for an electronic fence based on a 5G base station tower according to claim 2, wherein: The method of generating, based on the correlation between each environmental impact factor in the set of environmental impact factors and the motion rate sequence and motion direction angle sequence in the target motion data, a correction coefficient sequence of the meteorological factor data for the motion rate, a deflection angle sequence of the terrain factor data for the motion direction angle, and a fluctuation amplitude sequence of the electromagnetic interference factor data for the position coordinates, includes: Extracting wind speed data and wind direction data from the meteorological factor data in the set of environmental influencing factors, performing time series matching on the wind speed data and the motion rate sequence in the target motion data, calculating the ratio of the wind speed data to the motion rate at each time sampling point, and generating a wind speed ratio sequence; Based on the wind speed ratio sequence and the angle difference between the wind direction data and the motion direction angle sequence, a meteorological influence correction model is constructed, and the motion rate sequence is weighted point by point using the meteorological influence correction model to generate a correction coefficient sequence of the meteorological factor data for the motion rate; Extracting slope data and slope direction data from the terrain factor data in the set of environmental impact factors, spatially correlating the slope data with a motion direction angle sequence in the target motion data, calculating motion direction angle changes corresponding to different slope intervals, and generating a slope direction change sequence; Combining the slope data with the azimuth deviation value of the motion direction angle sequence, constructing a terrain influence deflection model, performing angle compensation processing on the motion direction angle sequence using the terrain influence deflection model, and generating a deflection angle sequence of the motion direction angle based on the terrain factor data; Extracting interference intensity data and interference frequency data from the electromagnetic interference factor data in the set of environmental impact factors, performing time domain correlation processing on the interference intensity data and the position coordinate sequence in the target motion data, calculating the position coordinate deviation when the interference intensity exceeds a preset threshold, and generating an interference position deviation sequence; An electromagnetic influence fluctuation model is constructed according to the ratio of the interference frequency data to the sampling frequency of the position coordinate sequence, and the interference position deviation sequence is amplitude modulated by the electromagnetic influence fluctuation model to generate a fluctuation amplitude sequence of the electromagnetic interference factor data to the position coordinates.

4. The method for generating a hierarchical management strategy for an electronic fence based on a 5G base station tower according to claim 1, wherein: Generating risk assessment information of the target object entering the electronic fence warning area according to the spatial relative relationship and the potential impact includes: Based on the vertical distance value and the horizontal offset in the spatial relative relationship, calculating the decay rate of the vertical distance value over time and the cumulative offset of the horizontal offset over time; Analyzing the correction coefficient sequence, deflection angle sequence, and fluctuation amplitude sequence in the potential impact, and extracting the minimum correction coefficient in the correction coefficient sequence, the maximum deflection angle in the deflection angle sequence, and the average fluctuation amplitude in the fluctuation amplitude sequence; If the decay rate is greater than a preset decay threshold and the cumulative offset is less than a preset offset threshold, and the minimum correction coefficient is less than a preset correction threshold, the maximum deflection angle is greater than a preset deflection angle threshold, and the average fluctuation amplitude is greater than a preset fluctuation threshold, it is determined that the target object has a first risk entry trend, and a first risk assessment algorithm is called to generate first risk assessment sub-information; If the decay rate is less than or equal to the preset decay threshold, or the cumulative offset is greater than or equal to the preset offset threshold, or the minimum correction coefficient is greater than or equal to the preset correction threshold, the maximum deflection angle is less than or equal to the preset deflection angle threshold, and the average fluctuation amplitude is less than or equal to the preset fluctuation threshold, it is determined that the target object has a second risk entry trend, and a second risk assessment algorithm is called to generate second risk assessment sub-information; The first risk assessment sub-information and the second risk assessment sub-information are integrated to generate risk assessment information including risk trend type, risk occurrence probability and risk impact scope.

5. The method for generating a hierarchical management strategy for an electronic fence based on a 5G base station tower according to claim 4, wherein: The calling of the first risk assessment algorithm to generate first risk assessment sub-information includes: Inputting the vertical distance value and the horizontal offset in the spatial relative relationship into the spatial feature processing layer of the first risk assessment algorithm, performing exponential decay function fitting processing on the vertical distance value to obtain a vertical distance decay curve; Perform sliding window summation on the horizontal offset to obtain a horizontal offset accumulation curve; Inputting the correction coefficient sequence, deflection angle sequence and fluctuation amplitude sequence in the potential impact into the environmental feature processing layer of the first risk assessment algorithm, performing minimum value filtering on the correction coefficient sequence, and obtaining a correction coefficient minimum value sequence; Performing peak detection processing on the deflection angle sequence to obtain a deflection angle peak sequence; Perform mean calculation on the fluctuation amplitude series to obtain the mean of the fluctuation amplitude; Inputting the vertical distance attenuation curve, the horizontal offset accumulation curve, the correction coefficient minimum value sequence, the deflection angle peak value sequence, and the fluctuation amplitude mean into the fusion layer of the first risk assessment algorithm to perform spatiotemporal feature correlation modeling, thereby generating a first risk feature matrix including a spatial risk vector and an environmental risk vector; The output layer of the first risk assessment algorithm is called to perform risk probability prediction processing on the first risk feature matrix to generate first risk assessment sub-information including a first risk probability value, a risk occurrence time window, and coordinates of a risk coverage area.

6. The method for generating a hierarchical management strategy for an electronic fence based on a 5G base station tower according to claim 1, wherein: The formulating of a hierarchical management strategy based on the risk assessment information includes: Analyze the risk trend type, risk occurrence probability, and risk impact range in the risk assessment information, and extract the policy template identifier corresponding to the risk trend type, the policy strength parameter corresponding to the risk occurrence probability, and the policy coverage parameter corresponding to the risk impact range; Retrieving a corresponding basic management policy framework from a preset policy template library based on the policy template identifier, wherein the basic management policy framework includes a policy execution subject, a policy execution process, and policy resource configuration items; Adjusting the step interval and execution frequency of the policy execution process in the basic management policy framework according to the policy strength parameter to generate a management policy framework after strength adjustment; generating a management policy framework after coverage adjustment according to the resource allocation ratio and resource scheduling priority of the policy resource configuration items in the management policy framework after the policy coverage parameter adjustment strength is adjusted; Performing policy verification on the management policy framework after the overlay adjustment, and identifying policy conflict items and policy missing items in the management policy framework after the overlay adjustment; Prioritizing the policy conflict items to obtain a priority sorting result; supplementing the policy missing items by adding policy execution steps that match the risk impact range to obtain a policy missing item supplementation result; A hierarchical management strategy is generated by combining the priority sorting processing result and the strategy missing item supplementation result.

7. The method for generating a hierarchical management strategy for an electronic fence based on a 5G base station tower according to claim 6, wherein: The performing policy verification on the management policy framework after the overlay adjustment to identify policy conflict items and policy missing items in the management policy framework after the overlay adjustment includes: Extracting all policy execution steps in the management policy framework after the coverage adjustment, and constructing a policy step directed graph, wherein the nodes of the policy step directed graph represent policy execution steps, and the directed edges represent the execution order relationship between the steps; Determine whether there is a circular dependency path in the directed graph of the policy steps, and if there is a circular dependency path, mark the policy execution steps constituting the circular dependency path as policy conflict items; Extracting resource requirement parameters from the policy execution steps, comparing resource requirement parameters of different policy execution steps within the same time window, and marking the corresponding policy execution step as a policy conflict item if the sum of the resource requirement parameters exceeds a preset resource threshold; Parsing the spatial coordinate set corresponding to the risk impact range in the risk assessment information, checking whether the policy execution steps in the coverage-adjusted management policy framework include execution actions covering all sub-areas in the spatial coordinate set, and if there are uncovered sub-areas, determining that there are policy missing items; Check whether the policy execution steps in the management policy framework after coverage adjustment include all preset response measures for the risk trend type. If there are preset response measures that are not included, it is determined that there are policy missing items.

8. The method for generating a hierarchical management strategy for an electronic fence based on a 5G base station tower according to claim 1, wherein: The comparing the real-time behavior data with the preset behavior benchmark data in the hierarchical management strategy and calculating the behavior deviation characteristics between the real-time behavior data and the preset behavior benchmark data include: Performing data optimization processing on the real-time position coordinates, real-time movement direction, and real-time movement speed in the real-time behavior data to obtain optimized real-time position coordinates, optimized real-time movement direction, and optimized real-time movement speed; Extracting a reference position coordinate sequence, a reference motion direction sequence, and a reference motion rate sequence from the preset behavior reference data from the hierarchical management strategy, and performing the same optimization processing as that for optimizing the real-time data on the reference position coordinate sequence, the reference motion direction sequence, and the reference motion rate sequence to obtain an optimized reference position coordinate sequence, an optimized reference motion direction sequence, and an optimized reference motion rate sequence; Calculate the Euclidean distance between the optimized real-time position coordinates and the optimized reference position coordinate sequence at the corresponding time sampling points to generate a position deviation sequence; Calculate the angular difference between the optimized real-time motion direction and the optimized reference motion direction sequence at the corresponding time sampling points to generate a direction deviation sequence; Calculate the absolute difference between the optimized real-time motion rate and the optimized reference motion rate sequence at the corresponding time sampling points to generate a rate deviation sequence; Perform feature fusion processing on the position deviation sequence, direction deviation sequence and speed deviation sequence to generate a behavioral deviation feature including deviation mean, deviation variance and deviation change trend; The step of screening out sensitive execution parameters from the execution parameter set of the hierarchical management strategy based on the behavioral deviation characteristics includes: Extracting all execution parameters of the execution parameter set of the hierarchical management policy, wherein the execution parameters include a policy startup threshold, a resource scheduling period, a monitoring frequency parameter, and a response delay parameter; Calculating the Spearman rank correlation coefficient between the deviation mean in the behavioral deviation feature and the parameter value of each execution parameter to generate a first correlation coefficient set; Calculating the Kendall rank correlation coefficient between the deviation variance in the behavioral deviation feature and the parameter value of each execution parameter to generate a second correlation coefficient set; Performing weighted sum processing on the correlation coefficients in the first correlation coefficient set and the second correlation coefficient set to generate a comprehensive correlation coefficient set; Filtering target execution parameters whose absolute values ​​of comprehensive correlation coefficients are greater than a preset correlation threshold from the comprehensive correlation coefficient set, and marking the target execution parameters as candidate sensitive execution parameters; Analyzing the sensitivity of the impact of parameter value changes of the candidate sensitive execution parameters on the behavioral deviation characteristics, calculating the amount of change in the behavioral deviation characteristics corresponding to a unit change in the parameter value, and generating a sensitivity index; Candidate sensitive execution parameters whose sensitivity index is greater than a preset sensitivity threshold are screened out as sensitive execution parameters.

9. A hierarchical management strategy generation system for electronic fences based on 5G base station towers, characterized in that: include: processor; A storage device having a computer program stored thereon, wherein when the computer program is executed by the processor, the processor implements the method for generating a hierarchical management strategy for an electronic fence based on a 5G base station tower as described in any one of claims 1 to 8.

Images