An edge-computing-based intelligent early warning method for sports venue internet-of-things equipment

Abstract

The application relates to the technical field of Internet of Things early warning, and discloses a stadium Internet of Things equipment intelligent early warning method based on edge computing, which comprises the following steps: calculating a structure vibration energy value based on instantaneous vibration speed; calculating an antenna offset angle by using a reference speed constant, and calculating a dynamic coverage radius by combining a basic coverage radius; calculating a spatial overlap area by using the dynamic coverage radius and a physical straight-line distance, calculating an exclusion overlap occupancy density index based on the spatial overlap area and an original number of occupants; calculating instantaneous transmission path loss according to a crowd attenuation coefficient and the exclusion overlap occupancy density index; calculating a compensation time stamp by using a path loss normalization constant, an original receiving time stamp and the instantaneous transmission path loss; and calculating an early warning index based on a reference alignment time, the compensation time stamp and a normalized sensor reading.

Description

Technical Field

[0001] This invention relates to the field of IoT early warning technology, specifically to an intelligent early warning method for IoT devices in sports venues based on edge computing. Background Technology

[0002] With the application of IoT technology, intelligent early warning systems for IoT devices in sports stadium edge computing scenarios have become commonplace. Existing technologies typically collect environmental data by deploying various IoT sensors within the stadium. Edge nodes receive data based on internal clocks. Data processing modules compare multi-source data according to set thresholds to achieve stadium safety early warnings. However, sports stadium environments possess physical and spatial complexity. The stadium's steel truss structure exhibits objective low-frequency micro-vibrations. These vibrations are transmitted along the truss to the IoT device bases. Conventional methods often assume the devices are in an ideal, stationary state, failing to quantify the energy of structural vibrations. Physical micro-biases can cause geometric deflections in antenna beams, distorting the actual detection boundaries. Simultaneously, wireless crowd occupancy sensors deployed within the stadium exhibit overlapping coverage. Existing technologies typically directly accumulate data when assessing crowd density. This approach cannot analyze the degree of spatial interference in the sensing area and fails to eliminate duplicate counts caused by overlapping spatial areas. This leads to an overestimation of the calculated occupancy density index. Furthermore, sports stadiums are characterized by dynamic crowd gatherings. The human body has electromagnetic absorption effects on specific radio frequency bands. Traditional algorithms often employ fixed path loss models. This model fails to map dynamic crowd density to a dynamic RF attenuation factor. This fails to reflect RF shielding effects, leading to errors in communication loss assessment. Furthermore, increased path loss increases retransmission time at lower levels. Heterogeneous IoT data packets experience time deviations during network transmission. Existing systems often directly read the original received timestamp, failing to extrapolate expected delays and perform compensation. Time-series data cannot achieve baseline alignment in this situation. Deviations in synchronization data can cause system warnings to fail.

[0003] In summary, an intelligent early warning method for IoT devices in sports venues based on edge computing is needed to solve the problem of early warning judgment failure caused by physical interference, spatial interference and communication delay. Summary of the Invention

[0004] This invention provides an intelligent early warning method for IoT devices in sports venues based on edge computing, which helps to solve the problems mentioned in the background art.

[0005] This invention provides the following technical solution: an intelligent early warning method for IoT devices in sports venues based on edge computing, comprising: The structural vibration energy value is calculated based on the total number of samplings and the instantaneous vibration velocity. The antenna offset angle is calculated using the reference velocity constant and the vibration energy value of the structure. The dynamic coverage radius is calculated by combining the basic coverage radius with the antenna offset angle; The spatial overlap area is calculated using the dynamic coverage radius, the basic coverage radius, and the physical straight-line distance. The rejection overlap occupancy density index is calculated based on the spatial overlap area, the dynamic coverage radius, and the original number of occupants. Instantaneous transmission path loss is calculated based on the base attenuation value, the crowd attenuation coefficient, the straight-line transmission distance, the system loss constant, and the rejection overlap occupancy density index. The compensation timestamp is calculated using the base delay time, the path loss normalization constant, the original reception timestamp, and the instantaneous transmission path loss. The early warning index is calculated based on the reference alignment time, normalized sensor readings, maximum range constant of sensor readings, total number of sensors, the basic delay time, and the compensation timestamp.

[0006] Optionally, the calculation of the structural vibration energy value based on the total number of samples and the instantaneous vibration velocity includes: Five vibration sensors were deployed at the key load-bearing nodes of the steel truss structure of the stadium. The key load-bearing nodes include the tops of the four main diagonal load-bearing columns around the stadium and the central cantilevered mid-span node of the roof. Multiple wireless crowd occupancy sensors are deployed in a grid pattern of square two-dimensional topology arrays with equal horizontal and vertical spacing in the stadium's grandstand area. In the venue's low-voltage control room, a standalone server is deployed as an edge node to serve as a data processing module, and is connected to all the vibration sensors and the wireless crowd occupancy sensors. Set the total number of samples; Acquire the instantaneous vibration velocity within a sliding time window that covers at least one complete cycle of the lowest-order principal vibration response of the stadium's steel structure; The instantaneous vibration velocity is squared. The instantaneous vibration velocity results after squaring are summed together. The summation result is divided by the total number of samplings to obtain the structural vibration energy value.

[0007] Optionally, calculating the antenna offset angle using the reference velocity constant and the structural vibration energy value includes: Obtain the maximum elastic deformation recovery rate of the stadium truss steel and use it as a reference velocity constant; Perform a square root operation on the vibration energy value of the structure; Divide the result of the square root operation by the reference velocity constant; The result of the division is subjected to an arcsine function operation to obtain the antenna offset angle.

[0008] Optionally, the calculation of the dynamic coverage radius by combining the basic coverage radius and the antenna offset angle includes: Obtain the maximum effective transmission distance of the wireless antenna in theoretical free space and use it as the basic coverage radius; Calculate the cosine value of the antenna offset angle; The dynamic coverage radius is obtained by multiplying the basic coverage radius by the cosine value.

[0009] Optionally, the calculation of the spatial overlap area using the dynamic coverage radius, the basic coverage radius, and the physical straight-line distance includes: Obtain the physical straight-line distance between adjacent sensors; Add the dynamic coverage radii of two adjacent sensors together; Subtract the physical straight-line distance from the sum; Divide the result of the subtraction by twice the base coverage radius to obtain the overlap scaling factor; Square the overlap scaling factor; The spatial overlap area is obtained by sequentially multiplying the squared result of the overlap scaling factor, the pi constant, and the dynamic coverage radius of the two adjacent sensors.

[0010] Optionally, the calculation of the rejection overlap occupancy density index based on the spatial overlap area, the dynamic coverage radius, and the original number of occupants includes: Get the original number of occupants; Squaring the dynamic coverage radius; Multiply the result of squaring the dynamic coverage radius by the constant pi to obtain the first intermediate result; The spatial overlap areas of each adjacent sensor are summed to obtain the area summation result; Divide the sum of the areas by the first intermediate result; Subtract the result of the division from the number 1; The difference obtained by subtracting is multiplied by the original number of occupants to obtain the corrected number of occupants; Subtract the sum of the areas from the first intermediate result to obtain the second intermediate result; Divide the corrected number of occupants by the second intermediate result to obtain the rejection overlap occupancy density index.

[0011] Optionally, the step of calculating the instantaneous transmission path loss based on the base attenuation value, the crowd attenuation coefficient, the straight-line transmission distance, the system loss constant, and the rejection overlap occupancy density index includes: Obtain the baseline of fixed shielding loss measured under the unloaded venue building structure as the basic attenuation value; The electromagnetic absorption rate of the human body in the radio frequency band is used as the population attenuation coefficient. Obtain the straight-line transmission distance from the sensor to the edge node; Obtain the fixed insertion loss values ​​of the antenna feed line and connector as the system loss constant; The population attenuation coefficient is multiplied by the rejection overlap density index; Add the result of multiplication to the base attenuation value to obtain the dynamic population attenuation factor; Calculate the base-10 logarithmic value of the straight-line transmission distance; Multiplying the logarithmic value by the number 20 yields the third intermediate result; The instantaneous transmission path loss is obtained by adding the third intermediate result, the dynamic crowd attenuation factor, and the system loss constant together.

[0012] Optionally, the step of calculating the compensation timestamp using the base delay time, the path loss normalization constant, the original reception timestamp, and the instantaneous transmission path loss includes: The shortest instruction processing cycle at the edge gateway's underlying layer is used as the base latency time; The inherent attenuation response redundancy of the receiver is obtained as the path loss normalization constant. Get the original received timestamp; Divide the instantaneous transmission path loss by the path loss normalization constant to obtain the exponential value. Calculate the result of a power operation with the number 10 as the base and the power value as the exponent; The expected delay time is obtained by multiplying the result of the exponentiation operation by the base delay time. The compensation timestamp is obtained by subtracting the expected delay time from the original reception timestamp.

[0013] Optionally, the calculation of the early warning index based on the reference alignment time, normalized sensor readings, maximum range constant of sensor readings, total number of sensors, the basic delay time, and the compensation timestamp includes: Set the absolute system base time of the current processing frame as the reference alignment time; The safe measurement limit range of the sensor hardware is obtained as the maximum range constant of the sensor reading. Acquire normalized sensor readings; Get the total number of sensors; Calculate the difference between the compensation timestamp and the reference alignment time; calculate the absolute value of the difference; Divide the absolute value by the base delay time, and take the opposite of the result of the division; The synchronization weight is obtained by calculating the result of the power operation with the natural constant as the base and the opposite number as the exponent; The synchronization weight of each sensor is multiplied by its corresponding normalized sensor reading. The summation of all the multiplication results yields the warning feature value. The fourth intermediate result is obtained by multiplying the maximum range constant of the sensor reading by the total number of sensors. Divide the warning feature value by the fourth intermediate result to obtain the warning index.

[0014] The present invention has the following beneficial effects:

[0015] 1. This solution utilizes vibration sensors deployed at key load-bearing nodes of the steel truss structure in sports stadiums and wireless crowd occupancy sensors deployed in the stands. Based on edge computing, it provides intelligent early warning for IoT devices. Sports stadiums, as specific application environments, exhibit complex physical and spatial structures. Micro-vibrations in the steel structure are transmitted to the device bases, causing antenna offset and geometric distortion of the actual detection range of the wireless sensors. Furthermore, the densely deployed sensor sensing area is prone to spatial interference. In addition, the dynamic gathering of crowds within the stadium can shield and absorb specific radio frequency bands, increasing transmission path loss and leading to anticipated delays in the network transmission of underlying IoT data packets, resulting in timing deviations. To address these issues, this solution sequentially calculates the structural vibration energy value and antenna offset angle using instantaneous vibration velocity, converting this into the dynamic coverage radius and spatial overlap area. It further isolates the spatial interference area to calculate the rejection overlap occupancy density index. Subsequently, the solution converts the density index into instantaneous transmission path loss, derives network transmission delay to calculate the compensation timestamp, and finally combines parameters such as reference alignment time to obtain the early warning index. This scheme maps microstructural vibrations across domains into corrections for spatial and temporal errors, correcting for detection boundary distortions caused by physical vibrations and repetitive counting errors due to spatial overlap. It also quantifies and eliminates transmission delays caused by radio frequency attenuation from dynamic crowds. In venue environments with structural vibrations, spatial interference, and communication fading, this scheme calibrates and synchronizes the underlying heterogeneous time-series data, outputting an early warning index with objective criteria.

[0016] 2. By deploying a specific number of vibration sensors at the top of the main diagonal load-bearing columns and the mid-span node of the cantilevered roof of the stadium's steel structure truss, and by deploying wireless crowd occupancy sensors in a square two-dimensional topological array in the stands area, and by extracting instantaneous vibration velocities within a sliding time window covering at least one complete cycle of the lowest-order principal vibration response of the stadium's steel structure, and then squaring and summing the instantaneous vibration velocities to obtain the structural vibration energy value, this method structurally binds the physical characteristics of the macroscopic stadium space with the microscopic sampling nodes. This ensures that the collected vibration physical quantities cover the low-frequency structural deformation characteristics of the building's foundation, overcomes the limitations of local response caused by conventional single-node sampling, quantifies the actual physical intensity of minor structural vibrations, and provides subsequent compensation calculations with benchmark judgment conditions that conform to the actual physical environment, thereby improving the objectivity and reliability of the underlying physical interference assessment.

[0017] 3. By obtaining the maximum elastic deformation recovery rate of the stadium truss steel as a reference velocity constant, and combining it with the square root result of the structural vibration energy value, the antenna offset angle is calculated using the arcsine function. This method utilizes the inherent physical recovery characteristics of structural materials to establish a reference benchmark, converting and normalizing the abstract physical vibration energy into a specific beam geometric offset radian change, establishing a positive analytical mapping relationship between kinetic energy and spatial geometric deflection angle. This reflects the degree of beam pointing change of wireless equipment caused by the excited vibration of the base, avoiding spatial detection angle deviation caused by ignoring the deformation of the physical load-bearing structure, and providing an angle reference that conforms to the physical deformation law for subsequent correction of the actual detection boundary of the sensor, enabling the system to adapt to environmental perturbations of the underlying structure.

[0018] 4. By obtaining the maximum effective transmission distance of the wireless antenna in theoretical free space as the basic coverage radius, and calculating the cosine value of the antenna offset angle obtained in the previous step, the two are multiplied to obtain the dynamic coverage radius. This step utilizes the spatial geometric transformation characteristics of cosine projection attenuation to convert the antenna viewing angle deviation caused by the underlying physical vibration into the reduction of the vertical component on the horizontal detection profile. This realizes the operation of reducing the three-dimensional vibration deflection angle into a two-dimensional planar geometric detection radius, dynamically restoring the true spatial coverage boundary of the wireless sensor under excited disturbance state, eliminating the spatial applicability defects of the static theoretical detection radius under specific stress environment, and enhancing the rationality of the device's sensing boundary definition.

[0019] 5. By introducing the physical straight-line distance between adjacent sensors and combining the addition and subtraction of dynamic coverage radii, an overlap scaling coefficient reflecting the degree of spatial interference is constructed. Then, the square of this coefficient is multiplied continuously with the constant pi and the dynamic coverage radius to obtain the spatial overlap area. Based on the relative proportional relationship between the center distance of the physical topology layout and their respective sensing radii, this approach mathematically constructs a geometric segmentation model for the interference and overlap of adjacent sensing areas. Without introducing complex intersecting surface analysis, it quantifies the objectively existing cross-detection redundancy area when multiple sensors work together, quantifies the degree of spatial interference in the sensing area, and establishes a spatial calibration basis for subsequent evaluation of the actual IoT sensing element density in a specific area.

[0020] 6. By subtracting the original number of occupants from the proportion of the spatial overlap area in the total theoretical detection area, the corrected number of occupants is obtained. The actual net area is then obtained by subtracting the sum of the areas from the total area. Finally, the ratio of the corrected number of occupants to the actual net area is calculated to obtain the rejection overlap occupancy density index. This process takes into account the phenomenon of repeated counting of sensors in densely deployed environments. The personnel counting base in the overlapping interference area in the theoretical detection space is stripped and calibrated proportionally. The difference in occupancy density assessment caused by the cross-coverage of multiple sensors is eliminated. The geometric dimension interference area data is transformed into the density characteristics of the environmental IoT sensing dimension, reflecting the degree of crowd gathering in the objective physical space after rejection overlap interference.

[0021] 7. By obtaining the electromagnetic absorption rate of the human body under the radio frequency band as the crowd attenuation coefficient, and combining it with the rejection overlap occupancy density index and the basic attenuation value of empty venues, the dynamic crowd attenuation factor is calculated. Then, it is added to the loss based on the logarithmic mapping of the straight transmission distance and the system loss constant to obtain the instantaneous transmission path loss. This design converts the change in the density of the dynamic crowd into the physical absorption equivalent of the signal attenuation in a specific radio frequency band, making up for the deficiency that the conventional static communication loss model cannot reflect the physical shielding effect of the dynamic crowd. It breaks down the mapping relationship between the crowd density in the physical environment and the electromagnetic communication frequency band obstruction characteristics. On the basis of free space path fading, it superimposes dynamic obstruction characteristics that conform to the actual operating conditions, improving the objectivity of communication quality assessment in complex environments.

[0022] 8. By obtaining the shortest instruction processing cycle at the bottom layer of the edge gateway as the basic delay time, and using the ratio of instantaneous transmission path loss to the path loss normalization constant to construct an exponential mapping law, the expected delay time is solved. Finally, the compensation timestamp is obtained by stripping from the original receiving timestamp. This method quantizes and maps the physical attenuation characteristics of the spatial radio frequency link to the retransmission and data packet queuing digital consumption of the underlying communication protocol, establishes a cross-domain conversion mechanism from communication physical loss to network transmission time deviation, removes the additional transmission delay component induced by environmental and channel changes from the original receiving node time recorded by the device, purifies the timing characteristics of heterogeneous IoT data packets, and enables the data collected by different nodes to have the technical conditions for alignment on the same reference time axis.

[0023] 9. By calculating the absolute value of the difference between the compensation timestamp and the reference alignment time of the control system, and using the natural constant to construct an exponential decay model to obtain the synchronization weight, the early warning characteristic value is obtained by multiplying it with the normalized sensor readings and accumulating the results. Finally, the early warning index is obtained by combining the maximum range constant of the sensor readings and the total number of sensors. This processing method constructs a nonlinear smooth weighting mechanism for data confidence based on the degree of time sequence deviation, assigns corresponding weights to high-quality synchronization data close to the reference frame, suppresses the data characteristics of time sequence deviation, and thus integrates the time characteristics after the previous calibration into the final safety judgment system, outputting a judgment label that comprehensively considers the underlying physical microseismic, spatial interference and communication fading multi-mode correction factors. Attached Figure Description

[0024] Figure 1 This is a schematic diagram of the basic process of the present invention.

[0025] Figure 2 This is a schematic diagram illustrating the relationship between vibration correction, coverage overlap, and occupancy density in this invention.

[0026] Figure 3 This is a schematic diagram illustrating the relationship between path loss, timestamp compensation, and early warning index in this invention. Detailed Implementation

[0027] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0028] Example 1, refer to Figure 1 A smart early warning method for IoT devices in sports venues based on edge computing, comprising: The structural vibration energy value is calculated based on the total number of samplings and the instantaneous vibration velocity. The antenna offset angle is calculated using the reference velocity constant and the vibration energy value of the structure. The dynamic coverage radius is calculated by combining the basic coverage radius with the antenna offset angle; The spatial overlap area is calculated using the dynamic coverage radius, the basic coverage radius, and the physical straight-line distance. The rejection overlap occupancy density index is calculated based on the spatial overlap area, the dynamic coverage radius, and the original number of occupants. Instantaneous transmission path loss is calculated based on the base attenuation value, the crowd attenuation coefficient, the straight-line transmission distance, the system loss constant, and the rejection overlap occupancy density index. The compensation timestamp is calculated using the base delay time, the path loss normalization constant, the original reception timestamp, and the instantaneous transmission path loss. The early warning index is calculated based on the reference alignment time, normalized sensor readings, maximum range constant of sensor readings, total number of sensors, the basic delay time, and the compensation timestamp.

[0029] The calculation of structural vibration energy based on the total number of samples and instantaneous vibration velocity includes: The purpose of this step is to quantify the instantaneous micro-vibration intensity of the venue structure. The principle is to establish the baseline data of the underlying physical disturbance by performing energy integration calculation on the instantaneous velocity within a sliding time window that covers at least one complete cycle of the lowest-order principal vibration response of the venue's steel structure. Five vibration sensors are deployed at key load-bearing nodes of the stadium's steel truss structure. These key load-bearing nodes include the tops of the four main diagonal load-bearing columns around the stadium and one cantilevered mid-span node in the center of the roof. Multiple wireless crowd occupancy sensors are deployed in a gridded square two-dimensional topology array with equal horizontal and vertical spacing in the stadium's grandstand area. An independent server is deployed as an edge node in the stadium's low-voltage control room as a data processing module, and is connected to all the vibration sensors and wireless crowd occupancy sensors. Before the calculation begins, the total number of samples is set manually as the baseline parameter for the subjective adjustment system; Simultaneously, the instantaneous vibration velocity of each sensor is acquired during each sampling within a sliding time window that covers at least one complete cycle of the lowest-order principal vibration response of the stadium's steel structure. First, the structural vibration energy value is calculated by summing the squares of the instantaneous vibration velocities and averaging them, following these steps: The instantaneous vibration velocity is squared; the results of the squared instantaneous vibration velocities are summed; the summation result is divided by the total number of samplings to obtain the structural vibration energy value.

[0030] By deploying a specific number of vibration sensors at the top of the main diagonal load-bearing columns and the mid-span node of the cantilevered roof of the stadium's steel truss, and by deploying wireless crowd occupancy sensors in a square two-dimensional topological grid in the stands area, and by extracting instantaneous vibration velocities within a sliding time window covering at least one complete cycle of the lowest-order principal vibration response of the stadium's steel structure, and then squaring and summing these instantaneous vibration velocities to obtain the structural vibration energy value, this method structurally binds the physical characteristics of the macroscopic stadium space with the microscopic sampling nodes. This allows the collected vibration physical quantities to cover the low-frequency structural deformation characteristics of the building's foundation, overcoming the limitations of local response caused by conventional single-node sampling, quantifying the actual physical intensity of minor structural vibrations, and providing subsequent compensation calculations with benchmark judgment conditions that conform to the actual physical environment, thereby improving the objectivity and reliability of the underlying physical interference assessment.

[0031] The calculation of the antenna offset angle using the reference velocity constant and the structural vibration energy value includes: The purpose of this step is to convert the microscopic physical vibration energy into the geometric offset angle of the antenna beam. The principle is to use inverse sine mapping to normalize the kinetic energy into the change in radians, thereby solving the problem of wireless beam pointing distortion caused by base vibration. Before the calculation begins, the maximum elastic deformation recovery rate of the stadium truss steel is obtained. This rate is an objectively existing physical and chemical constant of the medium and is used as a reference rate constant. Based on the ratio of the structural vibration energy value obtained in the previous step to this constant, the antenna offset angle is calculated using inverse trigonometric functions through the following steps: The square root of the structural vibration energy value is performed; the result of the square root operation is divided by the reference velocity constant; the result of the division is then subjected to an arcsine function operation to obtain the antenna offset angle.

[0032] By obtaining the maximum elastic deformation recovery rate of the stadium truss steel as a reference velocity constant and combining it with the square root result of the structural vibration energy value, the antenna offset angle is calculated using the arcsine function. This method utilizes the inherent physical recovery characteristics of structural materials to establish a reference benchmark, converting and normalizing the abstract physical vibration energy into a specific beam geometric offset radian change, establishing a positive analytical mapping relationship between kinetic energy and spatial geometric deflection angle. This reflects the degree of beam pointing change of wireless equipment caused by the excited vibration of the base, avoiding spatial detection angle deviation caused by ignoring the deformation of the physical load-bearing structure. It provides an angle reference that conforms to the physical deformation law for subsequent correction of the actual detection boundary of the sensor, enabling the system to adapt to environmental perturbations of the underlying structure.

[0033] The calculation of the dynamic coverage radius by combining the basic coverage radius and the antenna offset angle includes: The purpose of this step is to dynamically correct the actual spatial detection range of the wireless sensor; the principle is to calculate the vertical component of the basic transmission radius through the cosine attenuation model of the offset angle; this step converts the vibration angle into the spatial geometric radius. Before the calculation begins, the maximum effective transmission distance of the wireless antenna in theoretical free space is obtained. This distance is an inherent structural parameter of the device and is used as the basic coverage radius. Combining the base coverage radius with the cosine of the angle calculated in the previous step, the dynamic coverage radius of the k-th sensor is calculated using the following steps: Calculate the cosine value of the antenna offset angle; multiply the basic coverage radius by the cosine value to obtain the dynamic coverage radius.

[0034] By obtaining the maximum effective transmission distance of the wireless antenna in theoretical free space as the basic coverage radius, and calculating the cosine value of the antenna offset angle obtained in the previous step, the two are multiplied to obtain the dynamic coverage radius. This step utilizes the spatial geometric transformation characteristics of cosine projection attenuation to convert the antenna viewing angle deviation caused by the underlying physical vibration into the reduction of the vertical component on the horizontal detection profile. This realizes the operation of reducing the three-dimensional vibration deflection angle into a two-dimensional planar geometric detection radius, dynamically restoring the true spatial coverage boundary of the wireless sensor under excited disturbance state, eliminating the spatial applicability defects of the static theoretical detection radius under specific stress environment, and enhancing the rationality of the device's sensing boundary definition.

[0035] The calculation of the spatial overlap area using the dynamic coverage radius, the basic coverage radius, and the physical straight-line distance includes: The purpose of this step is to quantify the degree of spatial interference between the sensing areas of adjacent sensors; the principle is to analyze the overlap ratio by the relative ratio between the physical center distance and their respective dynamic radii. Before the calculation begins, the physical straight-line distance between adjacent sensors is obtained, which is an objective spatial layout parameter of the venue. First, combining the dynamic coverage radius and the physical straight-line distance, the overlap scaling factor between adjacent sensors is calculated through the following steps: Add the dynamic coverage radii of two adjacent sensors; subtract the physical straight-line distance from the sum; divide the result of the subtraction by twice the basic coverage radius to obtain the overlap scaling factor; Next, based on this coefficient, the one-dimensional scale is scaled and extended to a two-dimensional region, and the spatial overlap area is calculated through the following steps: The overlap scaling factor is squared; the result of the squared overlap scaling factor, the constant of pi, and the dynamic coverage radius of the two adjacent sensors are multiplied sequentially to obtain the spatial overlap area.

[0036] By introducing the physical straight-line distance between adjacent sensors and combining the addition and subtraction of dynamic coverage radii, an overlap scaling coefficient reflecting the degree of spatial interference is constructed. Then, the square of this coefficient is multiplied continuously by the constant pi and the dynamic coverage radius to obtain the spatial overlap area. Based on the relative proportional relationship between the center distance of the physical topology layout and their respective sensing radii, this approach mathematically constructs a geometric segmentation model for the interference and overlap of adjacent sensing areas. Without introducing complex intersecting surface analysis, it quantifies the objectively existing cross-detection redundancy area when multiple sensors work together, quantifies the degree of spatial interference in the sensing area, and establishes a spatial calibration basis for subsequent evaluation of the actual IoT sensing element density in a specific area.

[0037] Reference Figure 2 , Figure 2 The diagram shows the relationship between the structural vibration energy value, corrected to the antenna offset angle by a reference velocity constant, and further correlated with the dynamic coverage radius, spatial overlap area, and rejection overlap occupancy density index. The calculation of the rejection overlap occupancy density index based on the spatial overlap area, the dynamic coverage radius, and the original number of occupants includes: The purpose of this step is to eliminate duplicate data to assess the population density in the real physical space; the principle is to calibrate the actual population count base by proportionally subtracting the overlapping interference area from the total theoretical detection area. Before the calculation begins, the original number of occupants is obtained by direct measurement using equipment. First, the number of people measured is deducted based on the proportion of the overlapping area in the total area. The corrected number of people is then calculated through the following steps: The dynamic coverage radius is squared; the result of the squared dynamic coverage radius is multiplied by the constant pi to obtain a first intermediate result; the spatial overlap areas of each adjacent sensor are summed to obtain an area summation result; the area summation result is divided by the first intermediate result; the division result is subtracted from the number 1; the difference obtained by subtraction is multiplied by the original number of occupants to obtain the corrected number of occupants. Subsequently, using the corrected ratio of the number of people to the actual net area, the rejection overlap density index is calculated through the following steps: Subtract the sum of the areas from the first intermediate result to obtain the second intermediate result; divide the corrected number of people occupying the space by the second intermediate result to obtain the rejection overlap density index.

[0038] The corrected number of occupants is obtained by subtracting the original number of occupants from the proportion of the spatial overlap area in the total theoretical detection area. The actual net area is then obtained by subtracting the sum of the areas from the total area. Finally, the ratio of the corrected number of occupants to the actual net area is calculated to obtain the rejection overlap occupancy density index. This process takes into account the phenomenon of repeated counting of sensors in densely deployed environments. The personnel counting base is stripped and calibrated proportionally in the overlapping interference area in the theoretical detection space. This eliminates the difference in occupancy density assessment caused by the cross-coverage of multiple sensors. The geometric interference area data is transformed into density characteristics in the environmental IoT sensing dimension, reflecting the degree of crowd gathering in the objective physical space after rejection overlap interference.

[0039] Reference Figure 3 , Figure 3 The relationship between the rejection overlap density index and the communication attenuation factor in forming the instantaneous transmission path loss, and the early warning index generated after timestamp compensation and synchronization weight processing, is shown. The calculation of instantaneous transmission path loss based on the base attenuation value, the crowd attenuation coefficient, the straight-line transmission distance, the system loss constant, and the rejection overlap occupancy density index includes: The purpose of this step is to deduce the physical shielding effect of dynamic crowds on radio frequency signals and quantify communication loss; its principle is to use the density index to map the absorption attenuation of a specific radio frequency band and establish a real-time channel obstruction model; this step transfers the environmental density into the physical domain of the communication frequency band. The fixed shielding loss baseline measured under the empty venue building structure is obtained as the objectively existing basic attenuation value; the human electromagnetic absorption rate under this radio frequency band is obtained as the objectively existing crowd attenuation coefficient; the straight-line transmission distance from the sensor to the edge node is obtained; and the fixed insertion loss value of the antenna feed line and connector is obtained as the system loss constant of the inherent structural parameters of the equipment. First, combining the baseline decay value and density index, the dynamic population decay factor is calculated through the following steps: The population attenuation coefficient is multiplied by the rejection overlap density index; the result of the multiplication is added to the base attenuation value to obtain the dynamic population attenuation factor. Next, the instantaneous transmission path loss is further calculated by taking into account the spatial distance through the following steps: Calculate the logarithm of the straight-line transmission distance to base 10; multiply the logarithm by 20 to obtain a third intermediate result; add the third intermediate result, the dynamic crowd attenuation factor, and the system loss constant together to obtain the instantaneous transmission path loss.

[0040] By obtaining the electromagnetic absorption rate of the human body in the radio frequency band as the crowd attenuation coefficient, and combining it with the rejection overlap occupancy density index and the basic attenuation value of empty venues, the dynamic crowd attenuation factor is calculated. Then, it is added to the loss based on the logarithmic mapping of the straight transmission distance and the system loss constant to obtain the instantaneous transmission path loss. This design converts the change in the density of the dynamic crowd into the physical absorption equivalent of the signal attenuation in a specific radio frequency band, making up for the deficiency that the conventional static communication loss model cannot reflect the physical shielding effect of the dynamic crowd. It breaks down the mapping relationship between the crowd density in the physical environment and the electromagnetic communication frequency band obstruction characteristics, and superimposes dynamic obstruction characteristics that conform to the actual operating conditions on the basis of free space path fading, thereby improving the objectivity of communication quality assessment in complex environments.

[0041] The calculation of the compensation timestamp using the base delay time, the path loss normalization constant, the original reception timestamp, and the instantaneous transmission path loss includes: The purpose of this step is to calibrate the time deviation of the underlying heterogeneous IoT data packets in network transmission; its principle is to map path loss to retransmission and queuing time and separate it from the original timestamp; this step transforms the physical communication loss into a time digital feature. Before the calculation begins, the shortest instruction processing cycle at the bottom layer of the edge gateway is obtained as the inherent base latency of the device; the inherent attenuation response redundancy of the receiver is obtained as the inherent path loss normalization constant of the device; and the original reception timestamp is measured and obtained at the same time. First, based on the exponential mapping law of path loss, the expected delay time is calculated through the following steps: Divide the instantaneous transmission path loss by the path loss normalization constant to obtain the exponent value; calculate the result of the power operation with the number 10 as the base and the exponent value as the exponent; multiply the result of the power operation by the base delay time to obtain the expected delay time. Subsequently, the stripping delay is calculated using the following steps to determine the final compensation timestamp: The compensation timestamp is obtained by subtracting the expected delay time from the original reception timestamp.

[0042] By obtaining the shortest instruction processing cycle at the bottom layer of the edge gateway as the basic delay time, and constructing an exponential mapping law using the ratio of instantaneous transmission path loss to the path loss normalization constant, the expected delay time is solved. Finally, the compensation timestamp is obtained by stripping and calculating from the original receiving timestamp. This method quantizes and maps the physical attenuation characteristics of the spatial radio frequency link to the retransmission and data packet queuing digital consumption of the underlying communication protocol, establishes a cross-domain conversion mechanism from communication physical loss to network transmission time deviation, removes the additional transmission delay components induced by environmental and channel changes from the original receiving node time recorded by the device, purifies the timing characteristics of heterogeneous IoT data packets, and enables the data collected by different nodes to have the technical conditions for alignment on the same reference time axis.

[0043] The calculation of the early warning index based on the reference alignment time, normalized sensor readings, maximum sensor reading range constant, total number of sensors, the basic delay time, and the compensation timestamp includes: The purpose of this step is to integrate multimodal time-series aligned data to output a highly reliable venue safety assessment indicator. The principle is to construct an exponentially decaying weight by the deviation between the compensated timestamp and the system's core reference frame, thus preserving high-quality synchronization data. This step converges the calibration time into an early warning result. Before the calculation begins, the absolute system reference time of the current processing frame is set as the reference alignment time of the control system reference parameters; the safe measurement limit range of the sensor hardware is obtained as the inherent maximum range constant of the sensor reading; and the normalized sensor reading is measured and obtained at the same time. Calculate the difference between the compensation timestamp and the reference alignment time; calculate the absolute value of the difference; divide the absolute value by the base delay time and take the opposite of the result; calculate the result of the power operation with the natural constant as the base and the opposite as the exponent to obtain the synchronization weight; The synchronization weight of each sensor is multiplied by its corresponding normalized sensor reading; the results of all multiplications are summed to obtain the warning feature value. Finally, the limit range mapping is performed, and the final warning index is calculated through the following steps: The fourth intermediate result is obtained by multiplying the maximum range constant of the sensor reading by the total number of sensors; the warning characteristic value is then divided by the fourth intermediate result to obtain the warning index.

[0044] By calculating the absolute value of the difference between the compensation timestamp and the reference alignment time of the control system, and using the natural constant to construct an exponential decay model to obtain the synchronization weight, the early warning feature value is obtained by multiplying it with the normalized sensor readings and accumulating the results. Finally, the early warning index is obtained by combining the maximum range constant of the sensor readings and the total number of sensors. This processing method constructs a nonlinear smooth weighting mechanism for data confidence based on the degree of time sequence deviation, assigns corresponding weights to high-quality synchronization data close to the reference frame, suppresses the data features of time sequence deviation, and thus integrates the time features after the previous calibration into the final safety judgment system, outputting a judgment label that comprehensively considers the underlying physical microseismic, spatial interference and communication fading multi-mode correction factors.

[0045] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0046] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A smart early warning method for IoT devices in sports venues based on edge computing, characterized in that, include: The structural vibration energy value is calculated based on the total number of samplings and the instantaneous vibration velocity. The antenna offset angle is calculated using the reference velocity constant and the vibration energy value of the structure. The dynamic coverage radius is calculated by combining the basic coverage radius with the antenna offset angle; The spatial overlap area is calculated using the dynamic coverage radius, the basic coverage radius, and the physical straight-line distance. The rejection overlap occupancy density index is calculated based on the spatial overlap area, the dynamic coverage radius, and the original number of occupants. Instantaneous transmission path loss is calculated based on the base attenuation value, the crowd attenuation coefficient, the straight-line transmission distance, the system loss constant, and the rejection overlap occupancy density index. The compensation timestamp is calculated using the base delay time, the path loss normalization constant, the original reception timestamp, and the instantaneous transmission path loss. The early warning index is calculated based on the reference alignment time, normalized sensor readings, maximum range constant of sensor readings, total number of sensors, the basic delay time, and the compensation timestamp.

2. The intelligent early warning method for IoT devices in sports venues based on edge computing according to claim 1, characterized in that, The calculation of structural vibration energy based on the total number of samples and instantaneous vibration velocity includes: Five vibration sensors were deployed at the key load-bearing nodes of the steel truss structure of the stadium. The key load-bearing nodes include the tops of the four main diagonal load-bearing columns around the stadium and the central cantilevered mid-span node of the roof. Multiple wireless crowd occupancy sensors are deployed in a grid pattern of square two-dimensional topology arrays with equal horizontal and vertical spacing in the stadium's grandstand area. In the venue's low-voltage control room, a standalone server is deployed as an edge node to serve as a data processing module, and is connected to all the vibration sensors and the wireless crowd occupancy sensors. Set the total number of samples; Acquire the instantaneous vibration velocity within a sliding time window that covers at least one complete cycle of the lowest-order principal vibration response of the stadium's steel structure; The instantaneous vibration velocity is squared. The instantaneous vibration velocity results after squaring are summed together. The summation result is divided by the total number of samplings to obtain the structural vibration energy value.

3. The intelligent early warning method for IoT devices in sports venues based on edge computing according to claim 2, characterized in that, The calculation of the antenna offset angle using the reference velocity constant and the structural vibration energy value includes: Obtain the maximum elastic deformation recovery rate of the stadium truss steel and use it as a reference velocity constant; Perform a square root operation on the vibration energy value of the structure; Divide the result of the square root operation by the reference velocity constant; The result of the division is subjected to an arcsine function operation to obtain the antenna offset angle.

4. The intelligent early warning method for IoT devices in sports venues based on edge computing according to claim 3, characterized in that, The calculation of the dynamic coverage radius by combining the basic coverage radius and the antenna offset angle includes: Obtain the maximum effective transmission distance of the wireless antenna in theoretical free space and use it as the basic coverage radius; Calculate the cosine value of the antenna offset angle; The dynamic coverage radius is obtained by multiplying the basic coverage radius by the cosine value.

5. The intelligent early warning method for IoT devices in sports venues based on edge computing according to claim 4, characterized in that, The calculation of the spatial overlap area using the dynamic coverage radius, the basic coverage radius, and the physical straight-line distance includes: Obtain the physical straight-line distance between adjacent sensors; Add the dynamic coverage radii of two adjacent sensors together; Subtract the physical straight-line distance from the sum; Divide the result of the subtraction by twice the base coverage radius to obtain the overlap scaling factor; Square the overlap scaling factor; The spatial overlap area is obtained by sequentially multiplying the squared result of the overlap scaling factor, the pi constant, and the dynamic coverage radius of the two adjacent sensors.

6. The intelligent early warning method for IoT devices in sports venues based on edge computing according to claim 5, characterized in that, The calculation of the rejection overlap occupancy density index based on the spatial overlap area, the dynamic coverage radius, and the original number of occupants includes: Get the original number of occupants; Squaring the dynamic coverage radius; Multiply the result of squaring the dynamic coverage radius by the constant pi to obtain the first intermediate result; The spatial overlap areas of each adjacent sensor are summed to obtain the area summation result; Divide the sum of the areas by the first intermediate result; Subtract the result of the division from the number 1; The difference obtained by subtracting is multiplied by the original number of occupants to obtain the corrected number of occupants; Subtract the sum of the areas from the first intermediate result to obtain the second intermediate result; Divide the corrected number of occupants by the second intermediate result to obtain the rejection overlap occupancy density index.

7. The intelligent early warning method for IoT devices in sports venues based on edge computing according to claim 6, characterized in that, The calculation of instantaneous transmission path loss based on the base attenuation value, the crowd attenuation coefficient, the straight-line transmission distance, the system loss constant, and the rejection overlap occupancy density index includes: Obtain the baseline of fixed shielding loss measured under the unloaded venue building structure as the basic attenuation value; The electromagnetic absorption rate of the human body in the radio frequency band is used as the population attenuation coefficient. Obtain the straight-line transmission distance from the sensor to the edge node; Obtain the fixed insertion loss values ​​of the antenna feed line and connector as the system loss constant; The population attenuation coefficient is multiplied by the rejection overlap density index; Add the result of multiplication to the base attenuation value to obtain the dynamic population attenuation factor; Calculate the base-10 logarithmic value of the straight-line transmission distance; Multiplying the logarithmic value by the number 20 yields the third intermediate result; The instantaneous transmission path loss is obtained by adding the third intermediate result, the dynamic crowd attenuation factor, and the system loss constant together.

8. The intelligent early warning method for IoT devices in sports venues based on edge computing according to claim 7, characterized in that, The calculation of the compensation timestamp using the base delay time, the path loss normalization constant, the original reception timestamp, and the instantaneous transmission path loss includes: The shortest instruction processing cycle at the edge gateway's underlying layer is used as the base latency time; The inherent attenuation response redundancy of the receiver is obtained as the path loss normalization constant. Get the original received timestamp; Divide the instantaneous transmission path loss by the path loss normalization constant to obtain the exponential value. Calculate the result of a power operation with the number 10 as the base and the power value as the exponent; The expected delay time is obtained by multiplying the result of the exponentiation operation by the base delay time. The compensation timestamp is obtained by subtracting the expected delay time from the original reception timestamp.

9. The intelligent early warning method for IoT devices in sports venues based on edge computing according to claim 8, characterized in that, The calculation of the early warning index based on the reference alignment time, normalized sensor readings, maximum sensor reading range constant, total number of sensors, the basic delay time, and the compensation timestamp includes: Set the absolute system base time of the current processing frame as the reference alignment time; The safe measurement limit range of the sensor hardware is obtained as the maximum range constant of the sensor reading. Acquire normalized sensor readings; Get the total number of sensors; Calculate the difference between the compensation timestamp and the reference alignment time; calculate the absolute value of the difference; Divide the absolute value by the base delay time, and take the opposite of the result of the division; The synchronization weight is obtained by calculating the result of the power operation with the natural constant as the base and the opposite number as the exponent; The synchronization weight of each sensor is multiplied by its corresponding normalized sensor reading. The summation of all the multiplication results yields the warning feature value. The fourth intermediate result is obtained by multiplying the maximum range constant of the sensor reading by the total number of sensors. Divide the warning feature value by the fourth intermediate result to obtain the warning index.

Images