Distributed roof near-edge person approaching induction alarm system based on wireless networking

Abstract

The invention relates to the technical field of safety monitoring of people near a roof, and discloses a distributed sensing and warning system for people near the roof based on wireless networking. The system comprises a node deployment module, a signal synchronization module, a field domain division module, an approaching judgment module and an alarm triggering module. The node deployment module obtains related data and generates a node deployment set; the signal synchronization module synchronizes adjacent node signals and obtains a signal consistency synchronization labeling set; the field domain division module generates a field domain coverage overlay analysis result; the approaching discrimination module forms a personnel approaching abnormal point set; and the alarm triggering module generates an adjacent person approaching detection and alarm triggering result. The system can optimize the approaching monitoring of people near the edge of the roof, reduces the false alarm, improves the monitoring precision and reliability, and guarantees the safety of people in related regions.

Description

Technical Field

[0001] The present invention relates to the technical field of rooftop safety monitoring for personnel, and in particular to a distributed rooftop proximity sensing alarm system based on wireless networking. Background Art

[0002] With the accelerating pace of urbanization, the construction and use of high-rise buildings is becoming increasingly common. As a crucial component of buildings, rooftops pose an increasingly prominent safety challenge. Rooftop edges typically refer to areas with falling risks, such as the edge of the roof or near rooftop railings. If people approach these areas, accidents are highly likely to occur, posing a serious threat to life and safety. Currently, safety monitoring of people near rooftops primarily relies on manual inspections and traditional security equipment, but these methods have numerous limitations.

[0003] Manual inspections require dedicated personnel to regularly inspect rooftop edges, which is not only labor-intensive and time-consuming, but also limited in frequency and coverage, making it difficult to achieve round-the-clock, seamless monitoring. At night or in inclement weather, the efficiency and accuracy of manual inspections decrease significantly, making it impossible to detect dangerous situations involving people approaching the edge. Furthermore, manual inspections carry the risk of oversight, potentially leading to missed inspections due to fatigue or inattention, creating hidden dangers for safety management.

[0004] While traditional security equipment, such as surveillance cameras, can provide real-time monitoring of rooftop areas, they are often limited by factors such as camera angles and lighting conditions, resulting in blind spots. This is particularly true in complex rooftop environments, where obstructions from various facilities and equipment can prevent cameras from clearly capturing activity near adjacent areas. Furthermore, surveillance footage requires real-time human oversight, making it difficult to respond promptly when someone approaches the edge, resulting in significant alert lags.

[0005] Furthermore, some existing sensing monitoring systems lack wireless networking capabilities and lack scientific planning for node deployment, resulting in uneven signal coverage and the risk of blind spots. During signal synchronization, signals between adjacent nodes are susceptible to electromagnetic interference, environmental changes, and other factors, leading to frequent signal fluctuations that affect the consistency and accuracy of monitoring data. Furthermore, these systems fail to fully consider the impact of signal fluctuations on sensing range when demarcating the sensing field, causing deviations between field coverage analysis results and actual conditions, leading to inaccurate identification of proximity.

[0006] Furthermore, existing monitoring systems are poor at identifying anomalies within signal fluctuations when determining proximity, often misinterpreting normal signal fluctuations as signs of approaching people, resulting in numerous false alarms. Furthermore, when triggering alarms, these systems are unable to effectively distinguish locations with a high risk of false alarms, reducing the reliability of the alarms. This not only increases the workload of safety managers, but can also lead to a desensitization to the alarms due to frequent false alarms. Consequently, they fail to address real dangers promptly, potentially causing accidents.

[0007] The current field of personnel safety monitoring on rooftops faces problems such as low monitoring efficiency, limited coverage, poor signal stability, insufficient judgment accuracy, and low alarm reliability. There is an urgent need for a technical solution that can achieve accurate, efficient, and all-weather monitoring to meet the actual needs of rooftop safety management. Summary of the Invention

[0008] The purpose of the present invention is to provide a distributed rooftop edge personnel approaching sensing alarm system based on wireless networking to solve the problems raised in the above background technology.

[0009] To achieve the above objectives, the present invention provides a distributed rooftop edge personnel approach sensing alarm system based on wireless networking, the system comprising:

[0010] The node deployment module obtains the coordinates of the roof edge boundary and the node deployment time, collects the signal strength and electromagnetic interference value of the corresponding location, associates the deployed nodes, calculates the corresponding coverage matching value, and generates a node deployment set;

[0011] The signal synchronization module extracts the coverage matching values ​​and corresponding coordinates in the node deployment set, synchronizes the signals of adjacent nodes in chronological order, marks the stable and fluctuating sections in the adjacent nodes, and obtains a signal consistency synchronization annotation set;

[0012] The field division module obtains the deployment nodes located in the signal stable section of the signal consistency synchronization annotation set, extracts the node location and sensing range, compares the coverage range based on the coverage overlap, evaluates the sensing impact strength under signal fluctuations, and generates field coverage overlay analysis results;

[0013] The proximity discrimination module identifies abnormal points in the deployment nodes whose sensing response values ​​are greater than the average response reference value and are in the signal fluctuation section in the field coverage overlay analysis results, and forms a person proximity abnormal point set;

[0014] The alarm trigger module obtains all points where people are close to abnormal points and corresponding point information, marks points with false alarm risks, and generates edge person approach detection and alarm triggering results.

[0015] Preferably, the node deployment set includes coverage matching value, deployment node spatial coordinates, and normalized deployment factor; the signal consistency synchronization annotation set specifically includes signal stable segment annotation, signal fluctuation segment annotation, and adjacent deployment node coverage matching value difference rate; the field coverage overlay analysis results include the degree of influence of node position offset on coverage, the degree of influence of sensing range expansion on coverage, and coverage response comparison under each signal fluctuation condition; the personnel approaching abnormal point set includes the spatial position of the abnormal point, the abnormal point signal strength variation characteristics, and the abnormal point coverage and sensing fluctuation ratio; the edge personnel approaching detection and alarm triggering results include a list of detected abnormal points and a personnel approaching abnormal point set.

[0016] Preferably, the node deployment module includes:

[0017] The boundary information collection submodule obtains the coordinate position of the roof edge boundary and the node deployment time, collects the signal strength data corresponding to the coordinate position and the electromagnetic interference data of the day, and records the collection results as two deployment factors, signal factor and interference factor, to obtain the deployment node environmental factor data set;

[0018] The node parameter configuration submodule performs normalization processing on the signal factor and interference factor data in the deployment node environmental factor data group, establishes a correspondence between the normalized results and the deployment node coordinate positions, calculates the average value of the normalized signal value and the normalized interference value as the coverage matching value, and generates a node deployment set.

[0019] Preferably, the signal synchronization module includes:

[0020] The synchronization benchmark extraction submodule obtains the coverage matching values ​​and corresponding coordinate data in the node deployment set, identifies the time synchronization relationship of all deployed nodes based on the coordinate information, calls the deployment node time set, and uses the adjacent time threshold as a benchmark to perform signal synchronization measurement and sorting on the time deployed nodes to generate a time sorting sequence of adjacent deployment nodes;

[0021] The deviation calibration submodule uses a stable synchronization reference value and a fluctuating synchronization reference value to compare the time sequence of the adjacent deployment nodes, and integrates and generates a signal synchronization deviation sequence;

[0022] The consistency annotation submodule extracts the signal change direction and interference change direction between adjacent deployment nodes based on the signal synchronization deviation sequence, classifies and annotates each pair of deployment nodes according to whether the change trends in the two directions are consistent, records and groups the signal stable and signal fluctuating sections respectively, and obtains the signal consistency synchronization annotation set.

[0023] Preferably, the field division module includes:

[0024] The field feature extraction submodule selects the segments marked as signal stable according to the signal consistency synchronization annotation set, detects the position data and sensing range data of each deployment node within the coverage period, arranges them in spatial order to form a position space sequence and a sensing range sequence, and generates a field coverage space sequence set;

[0025] The coverage overlay calculation submodule calculates the position offset rate and sensing range expansion rate between consecutive spatial nodes in each deployment node spatial sequence based on the field coverage spatial sequence set, compares the position offset rate and sensing range expansion rate in parallel under the same coverage conditions, and identifies the numerical relationship between the change amplitude under the coverage range by jointly analyzing the two types of rate indicators. It integrates the influence value sequence of each deployment node and establishes the field coverage overlay analysis result.

[0026] Preferably, the proximity determination module includes:

[0027] The sequence extraction submodule selects deployment nodes whose sensing response values ​​are greater than the average response reference value and deployment nodes in the signal fluctuation section based on the field coverage superposition analysis results, extracts continuous coverage records of the deployment nodes in spatial order, collects coverage range and sensing sensitivity data corresponding to each spatial node, and generates a continuous coverage record set;

[0028] The fluctuation change ratio calculation submodule calls the continuous coverage record set, extracts the coverage value and sensing value of the deployment node at two consecutive spatial nodes, calculates the coverage change ratio and sensing sensitivity change ratio, integrates them into the coverage change ratio sequence and the sensing sensitivity change ratio sequence, and establishes a coverage fluctuation change data set;

[0029] Based on the coverage fluctuation change data set, the anomaly identification submodule extracts the signal strength variation data and interference fluctuation data of the corresponding spatial segment, determines whether the coverage change ratio and the sensing sensitivity change ratio both exceed the set fluctuation identification threshold, and determines whether the signal strength variation and the interference fluctuation simultaneously exceed the anomaly judgment threshold. The spatial nodes that meet the conditions are marked as anomaly points, and a set of people approaching anomaly points is generated.

[0030] Preferably, the alarm triggering module includes:

[0031] The indicator joint judgment submodule obtains all the points where the person is close to the abnormal point concentration and the corresponding coordinates and identification information, uses the stable response threshold and the fluctuation response threshold to compare them respectively, and establishes a joint response judgment value sequence;

[0032] Based on the joint response judgment value sequence, the abnormal output sorting submodule screens points whose sensing response values ​​are greater than the sensing response risk threshold, whose coverage matching values ​​are lower than the coverage reference value, and whose signal consistency labels are fluctuation segments. The corresponding point numbers, location identifiers, and segments to which they belong are extracted, and marked as detection anomalies with false alarm risks. The points that meet the joint conditions are segmented and output in a structured format to generate edge person approach detection and alarm triggering results.

[0033] Preferably, the system further includes a parameter calibration module, and the parameter calibration module includes:

[0034] The historical data retrieval submodule obtains the node deployment set, signal consistency synchronization annotation set and field coverage overlay analysis results within the historical period, extracts the historical operating parameters and actual detection results of each module, and generates a historical parameter comparison data set;

[0035] The calibration coefficient calculation submodule calculates the calibration weights of the signal factor and the interference factor based on the historical parameter comparison data set, performs weighted adjustment on the calibration weights and the normalized signal value and normalized interference value of the currently deployed node, generates a calibrated coverage matching value sequence, and updates the node deployment set.

[0036] Preferably, the system further includes an abnormality tracing module, which includes:

[0037] The trajectory tracking submodule obtains the spatial position and time node information of the person approaching the abnormal point in the abnormal point set, retrieves the continuous coverage record set and coverage fluctuation change data set within the corresponding time period, restores the signal strength change trajectory and interference fluctuation trajectory of the abnormal point in chronological order, and generates abnormal point trajectory tracing data;

[0038] The responsible section marking submodule extracts the boundary markers of the signal stable section and the signal fluctuating section based on the abnormal point trajectory tracing data, determines the specific location where the abnormal point trajectory crosses the section, marks the responsibility for the main impact section causing the abnormality, and generates the abnormal point responsible section marking result.

[0039] Preferably, the system further includes a result output module, and the result output module includes:

[0040] The multi-dimensional integration submodule obtains the edge personnel approach detection and alarm triggering results, the calibrated node deployment set of the parameter calibration module, and the abnormal point responsibility section marking results of the abnormality tracing module, extracts the detection abnormal point list, calibration coverage matching value sequence and responsibility section marking information, and integrates the data according to the three dimensions of detection results, calibration parameters, and tracing conclusions to generate a comprehensive output data set;

[0041] Based on the comprehensive output data set, the structured output submodule classifies the list of detected abnormal points by location segment, arranges the calibration coverage matching value sequence by time period, sorts the responsible section marking information by the degree of influence, and outputs it in a format combining tables and text to generate the final comprehensive results of edge personnel approach detection and alarm.

[0042] Compared with the prior art, the present invention has the following beneficial effects:

[0043] This distributed wireless networking-based rooftop proximity sensing alarm system uses a node deployment module to obtain the coordinates of the rooftop edge boundary and the node deployment time. It also collects the signal strength and daily electromagnetic interference values ​​at the corresponding locations, associates the deployed nodes, and calculates coverage matching values ​​to generate a node deployment set. This approach allows node deployment to better match the actual rooftop environment. Different rooftop edge areas vary in signal strength and electromagnetic interference. By accurately collecting this data and calculating coverage matching values, deployed nodes can achieve more optimal signal coverage, avoiding monitoring blind spots caused by improper node deployment and ensuring that each node functions within its optimal coverage range.

[0044] The signal synchronization module extracts coverage match values ​​and corresponding coordinates, synchronizes adjacent node signals in chronological order, and marks stable and fluctuating segments to create a signal consistency synchronization annotation set. This process effectively resolves the issue of signal asynchrony between adjacent nodes. In wireless networking, signals from adjacent nodes are easily affected by various factors and may become asynchronous. Marking stable and fluctuating segments allows the system to clearly understand signal variation patterns, providing a more reliable signal foundation for subsequent field demarcation and human proximity identification, reducing monitoring errors caused by signal asynchrony.

[0045] The field segmentation module identifies deployed nodes within stable signal zones, extracts node locations and sensing ranges, compares coverage ranges based on coverage overlap, assesses the impact of signal fluctuations, and generates a field coverage overlay analysis, fully accounting for the impact of signal fluctuations on sensing range. Signal fluctuations are inevitable in wireless networking, and these fluctuations can alter the actual sensing range of nodes. By assessing their impact and performing field coverage overlay analysis, field segmentation can be more precise, ensuring that the system's sensing coverage of rooftop areas is more realistic and avoiding missed or misjudged human presence due to inappropriate field segmentation.

[0046] The proximity identification module identifies outliers within deployment nodes whose sensing response values ​​exceed the average response baseline value and are within signal fluctuation ranges within the field coverage overlay analysis results. This creates a set of outlier points, improving the accuracy of proximity identification. In signal fluctuation ranges, misjudgments are more likely to occur. By comparing the sensing response values ​​with the average response baseline value, it effectively identifies true outliers, reducing the misinterpretation of normal signal fluctuations as human proximity and making the outlier point set more valuable for reference.

[0047] The alarm trigger module captures all points and corresponding location information for people approaching abnormal points, marks those with false alarm risks, and generates adjacent person approach detection and alarm triggering results, significantly improving alarm reliability. Marking false alarm risk points allows safety managers to more effectively address received alarms, avoiding the potential for numerous false alarms to disrupt their judgment. This allows managers to quickly identify true dangers and take timely measures to prevent accidents.

[0048] Furthermore, through the collaborative work of various modules, the entire system implements an integrated process from node deployment, signal synchronization, field division, proximity detection, to alarm triggering, adapting to the complex environmental conditions of rooftops. Whether it's structural differences between different rooftops or various electromagnetic interference and signal fluctuations, the system can effectively respond, eliminating the need for manual inspections and reducing labor costs. It also enables real-time, 24 / 7 monitoring, making rooftop safety management more efficient and intelligent. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 This is a working principle diagram of the distributed rooftop edge personnel approaching sensing alarm system based on wireless networking according to the present invention;

[0050] Figure 2 Flowcharts generated for the data set;

[0051] Figure 3 Flowchart of the module deployment process for a node;

[0052] Figure 4 Flowchart of the signal synchronization module;

[0053] Figure 5 Flowchart of the module work for field partitioning. DETAILED DESCRIPTION

[0054] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0055] See also Figure 1-Figure 5 The present invention provides a distributed rooftop proximity sensing alarm system based on wireless networking. The system includes a node deployment module, a signal synchronization module, a field division module, a proximity determination module, and an alarm triggering module. The specific implementation is as follows:

[0056] The node deployment module obtains the coordinates of the roof edge boundary and the node deployment time, collects the signal strength and electromagnetic interference value of the corresponding location, associates the deployed nodes, calculates the corresponding coverage matching value, and generates a node deployment set;

[0057] The signal synchronization module extracts the coverage matching values ​​and corresponding coordinates in the node deployment set, synchronizes the signals of adjacent nodes in chronological order, marks the stable and fluctuating segments in the adjacent nodes, and obtains the signal consistency synchronization annotation set;

[0058] The field division module obtains signal consistency and simultaneously marks the deployment nodes located in the signal stable section, extracts the node location and sensing range, compares the coverage range based on the coverage overlap, evaluates the sensing impact strength under signal fluctuations, and generates field coverage overlay analysis results;

[0059] The proximity discrimination module identifies abnormal points in the deployment nodes whose sensing response values ​​are greater than the average response benchmark value and are in the signal fluctuation section in the field coverage overlay analysis results, forming a set of abnormal points where people are close to each other;

[0060] The alarm trigger module obtains all points where people are close to abnormal points and the corresponding point information, marks the points with false alarm risks, and generates the results of edge person approach detection and alarm triggering.

[0061] Example 1:

[0062] The node deployment set includes coverage matching values, spatial coordinates of deployed nodes, and normalized deployment factors. The coverage matching value reflects the degree of compatibility between the signal coverage and environmental interference of the deployed node at the corresponding location. The spatial coordinates of the deployed nodes precisely identify the specific location of each node in the rooftop area. The normalized deployment factor is a parameter obtained by normalizing the signal factor and interference factor to eliminate the impact of different data levels. The signal consistency synchronization annotation set specifically consists of stable signal segment annotations, fluctuating signal segment annotations, and the coverage matching value difference ratio of adjacent deployed nodes. The stable signal segment annotations mark the time periods and spatial ranges where the signal transmission of adjacent nodes is stable. The fluctuating signal segment annotations correspond to areas and periods where the signal transmission is unstable. The coverage matching value difference ratio of adjacent deployed nodes quantifies the degree of difference in coverage matching values ​​between adjacent nodes, thereby determining the consistency of signal coordination between nodes. The results of the field coverage overlay analysis include the impact of node position offset on coverage, the impact of sensing range expansion on coverage, and a comparison of coverage responses under each signal fluctuation condition. The impact of node position offset on coverage is used to assess the effect of deviations between the node's actual installation location and the preset coordinates on signal coverage. The impact of sensing range expansion on coverage analyzes the impact of expanding or contracting the sensing range under different conditions on coverage. The coverage response comparison under each signal fluctuation condition records the changes in the coverage area's response under different signal fluctuation intensities. The person approach anomaly point set includes the anomaly's spatial location, the anomaly's signal strength variation characteristics, and the anomaly's coverage-to-sensing fluctuation ratio. The anomaly's spatial location specifies the anomaly's specific location on the roof edge. The anomaly's signal strength variation characteristics describe the amplitude and trend of signal strength variations at the anomaly. The anomaly's coverage-to-sensing fluctuation ratio reflects the proportional relationship between coverage fluctuations and sensor sensitivity fluctuations. The person approach detection and alarm triggering results include a list of detected anomaly points and a person approach anomaly point set. The list of detected anomaly points summarizes the numbers and basic information of all identified anomalies, while the person approach anomaly point set serves as the core basis for triggering the alarm.

[0063] The boundary information collection submodule in the node deployment module is responsible for obtaining the coordinate position of the roof edge boundary and the node deployment time. When obtaining the coordinate position, the edge lines of the roof edge are sampled at multiple points through high-precision positioning equipment to generate a continuous coordinate sequence to ensure the accuracy of the boundary range. The node deployment time is accurate to the minute, recording the specific time when each node is installed. At the same time, this submodule collects the signal strength data and the electromagnetic interference data of the corresponding coordinate position. The signal strength data is obtained in real time through the node's own signal receiving device. The electromagnetic interference data of the day is continuously collected through a special interference detection device. The sampling interval is set according to the complexity of the environment, generally once every 10-30 seconds. The collected signal strength data and electromagnetic interference data are recorded as two deployment factors, signal factor and interference factor, respectively, and together constitute the deployment node environment factor data group. This data group will be dynamically updated over time to reflect real-time changes in the environment.

[0064] The node parameter configuration submodule performs normalization processing based on the signal factor and interference factor data in the deployment node environmental factor data group. During the normalization process, the value range of the signal factor and interference factor is first determined, and the actual collected data is mapped to the interval of 0-1 to eliminate the influence of different physical dimensions. For example, if the actual value range of the signal factor is [-80dBm, -30dBm], each signal factor value is converted to a corresponding value between 0-1 through linear transformation. The result after normalization is established in a one-to-one correspondence with the coordinate position of the deployment node, that is, each coordinate point has a corresponding normalized signal factor and normalized interference factor. Subsequently, the average value of the normalized signal value and the normalized interference value is calculated as the coverage matching value. The specific calculation method is to add the normalized signal value and the normalized interference value of the same node and divide it by 2. The result is the coverage matching value of the node. Finally, the coverage matching value, deployment node spatial coordinates, and normalized deployment factors (i.e., normalized signal value and normalized interference value) are integrated together to generate a node deployment set, which will be stored in the system database to provide basic data for the processing of subsequent modules.

[0065] The synchronization benchmark extraction submodule in the signal synchronization module retrieves the coverage matching value and corresponding coordinate data from the node deployment set, determines the spatial distribution of all deployed nodes based on the coordinate information, and then identifies the synchronization relationship of all deployed nodes in time. Call the deployment node time set, which contains the deployment time and running timestamp of each node. Based on the adjacent time threshold, the adjacent time threshold is set according to the communication frequency of the node, usually 50-100 milliseconds, and the signal synchronization of the deployed nodes in time is measured. During the measurement process, the signal transmission delay and phase difference of different nodes at the same time point are compared, and then the nodes are sorted according to the measurement results to generate a time sorting sequence of adjacent deployment nodes. This sequence reflects the order of signal synchronization of adjacent nodes in the time dimension.

[0066] The deviation calibration submodule compares the time-ordered sequence of adjacent deployed nodes using a stable synchronization benchmark and a fluctuating synchronization benchmark. The stable synchronization benchmark is the system's preset standard deviation of signal synchronization between adjacent nodes under ideal conditions, while the fluctuating synchronization benchmark is the average deviation under signal fluctuations, calculated based on historical data. The actual measured synchronization deviation is compared with these two benchmarks to calculate the deviation difference. These deviation differences are then combined to generate a signal synchronization deviation sequence, which records the signal synchronization deviation between each pair of adjacent nodes.

[0067] The consistency annotation submodule extracts the direction of signal and interference changes between adjacent deployed nodes based on the signal synchronization deviation sequence. Signal change directions include signal enhancement, signal weakening, or signal stability, while interference change directions include interference enhancement, interference weakening, or interference stability. Each pair of deployed nodes is classified and annotated based on whether the change trends in both directions are consistent. If the signal change direction is the same as the interference change direction, such as simultaneous enhancement or weakening, it is marked as a signal stable segment. If the change directions are opposite or one direction changes dramatically while the other remains stable, it is marked as a signal fluctuating segment. Segments of stable and fluctuating signals are recorded and grouped separately. The records include the segment's start time, end time, and the node pairs involved. Grouping classifies adjacent segments according to their spatial location, ultimately obtaining a signal consistency synchronization annotation set. This annotation set provides a basis for signal state division for subsequent field division and proximity discrimination.

[0068] Example 2:

[0069] The field segmentation module consists of a field feature extraction submodule and a coverage overlay calculation submodule. The field feature extraction submodule receives a synchronized signal consistency annotation set, which includes information such as stable signal segments, fluctuating signal segments, and the difference rate of coverage matching values ​​between adjacent deployed nodes. The submodule first filters the annotation set, retaining only segments marked as stable. These segments indicate stable signal transmission at the node and are suitable as the basis for field segmentation. Next, the submodule detects the location and sensing range data of each deployed node within its coverage period. Location data is obtained through the node's built-in positioning component and includes latitude, longitude, and altitude information. Sensing range data is determined based on the node's sensing parameters and environmental conditions, typically presenting a fan-shaped or circular coverage area centered on the node. This location data is spatially arranged to form a position spatial sequence. The sensing range data is also spatially arranged to form a sensing range sequence. These two sequences are combined to form a field coverage spatial sequence set, which fully records the spatial distribution and sensing range of nodes within each stable segment.

[0070] The coverage overlay calculation submodule processes the set of spatial sequences of field coverage and calculates the position drift rate and sensing range expansion rate between consecutive nodes in each deployment node spatial sequence. The position drift rate refers to the change in position of two adjacent spatial nodes per unit time and is calculated by dividing the difference in their position coordinates by the time interval. The sensing range expansion rate is the difference in the sensing range radius of two adjacent spatial nodes by the time interval. The position drift rate and sensing range expansion rate are compared side by side under the same coverage conditions, such as signal strength and electromagnetic interference levels. By jointly analyzing these two rate metrics, the numerical relationship between them and the magnitude of coverage change is identified. For example, whether an increase in the position drift rate leads to a decrease in coverage, or whether an increase in the sensing range expansion rate is accompanied by a nonlinear expansion of coverage, is determined. The impact values ​​of each deployment node at different time periods and spatial locations are integrated to comprehensively reflect the effects of position drift and sensing range expansion on coverage, ultimately generating the field coverage overlay analysis results.

[0071] The proximity discrimination module consists of a sequence extraction submodule, a fluctuation change ratio calculation submodule, and anomaly identification submodule. The sequence extraction submodule receives the results of the field coverage overlay analysis, which includes information such as the impact of node position offset on coverage, the impact of sensing range expansion on coverage, and a comparison of coverage responses under different signal fluctuation conditions. The submodule then selects two types of deployed nodes: nodes whose sensing response values ​​exceed the average response benchmark value (the average response benchmark value is calculated based on historical sensing data); and nodes within the signal fluctuation range. Continuous coverage records for these deployed nodes are extracted in spatial order. These records include information such as the coverage boundary coordinates and sensing signal strength at each time point. Coverage data and sensing sensitivity data corresponding to each spatial node are also collected. The sensing sensitivity data reflects the node's sensitivity to the proximity of people. These data are integrated to generate a continuous coverage record set.

[0072] The Fluctuation Change Ratio Calculation submodule uses a continuous coverage record set to extract the coverage and sensing values ​​of the deployed node at two consecutive spatial nodes. The coverage value quantifies the coverage range, while the sensing value reflects the strength of the sensing signal. The coverage change ratio is calculated as the difference between the coverage value of the subsequent spatial node and the coverage value of the previous spatial node divided by the coverage value of the previous spatial node. The sensing sensitivity change ratio is also calculated as the difference between the sensing value of the subsequent spatial node and the sensing value of the previous spatial node divided by the sensing value of the previous spatial node. The coverage change ratios of all consecutive spatial nodes are arranged in sequence to form a coverage change ratio sequence. Similarly, the sensing sensitivity change ratios are arranged in sequence to form a sensing sensitivity change ratio sequence. Together, these two sequences constitute the coverage fluctuation change dataset.

[0073] The anomaly identification submodule operates based on the coverage fluctuation data set, extracting signal strength variation and interference fluctuation data for the corresponding spatial segments. Signal strength variation refers to the maximum change in signal strength within a given spatial segment, while interference fluctuation refers to the amplitude of electromagnetic interference fluctuation within that segment. The submodule first determines whether both the coverage variation ratio and the sensor sensitivity variation ratio exceed the set fluctuation identification threshold. This threshold is determined based on the system's definition of a normal fluctuation range. It then determines whether both the signal strength variation and the interference fluctuation exceed the anomaly determination threshold. The anomaly determination threshold, which is higher than the fluctuation identification threshold, distinguishes normal from abnormal fluctuations. If all four conditions are met simultaneously—that is, if both the coverage variation ratio and the sensor sensitivity variation ratio exceed the fluctuation identification threshold, and both the signal strength variation and the interference fluctuation exceed the anomaly determination threshold—the spatial node is marked as an anomaly. The information from all marked anomaly points is integrated, including the spatial coordinates, corresponding timestamps, and signal variation characteristics of the anomaly points, to generate a set of anomaly points where people have approached.

[0074] Example 3:

[0075] The alarm trigger module includes an indicator joint determination submodule and an abnormal output collation submodule. The indicator joint determination submodule receives a set of abnormal point locations where people are approaching and extracts detailed information about all points, including the three-dimensional spatial coordinates of each point, a unique identification code, the corresponding signal strength record, the induction response value, and the signal segment label to which it belongs. For each point, the submodule compares its induction response value with the stable response threshold and the fluctuation response threshold. The stable response threshold is the benchmark value used by the system to determine the proximity of people in the stable signal segment, while the fluctuation response threshold is an adjusted benchmark value set for the fluctuating signal segment, and its value is usually higher than the stable response threshold. The difference between each point and the two thresholds is calculated by comparison, and the two differences are then weighted and summed according to preset weights to establish a joint response determination value sequence. Each value in the sequence corresponds to the comprehensive determination result of an abnormal point.

[0076] The anomaly output consolidation submodule performs subsequent processing based on the joint response determination value sequence. First, a sensing response risk threshold is set, which is higher than the fluctuation response threshold, to filter out high-risk sensing response signals. A coverage benchmark is also set as a baseline for measuring the coverage effectiveness of deployed nodes. This submodule selects three types of points from the joint response determination value sequence: points with sensing response values ​​greater than the sensing response risk threshold, points with coverage matching values ​​lower than the coverage benchmark, and points with signal consistency labels clearly indicating a fluctuation segment. For points that meet all three criteria, the module extracts their point number, specific location identifier (such as the corresponding roof edge area number and straight-line distance from the edge), and signal segment information. These points are then marked as having detection anomalies and a risk of false alarms. These high-risk points are then grouped according to the spatial segment to which they belong. Each group contains details of all eligible points within that segment. The results are ultimately output in a structured format, forming the edge person approach detection and alarm triggering results, which can be directly used to activate subsequent alarm devices.

[0077] The system also includes a parameter calibration module, which consists of a historical data retrieval submodule and a calibration coefficient calculation submodule. The main function of the historical data retrieval submodule is to extract the operating data within the historical period from the system database, including the node deployment set generated in the past period, the signal consistency synchronization annotation set, and the field coverage overlay analysis results. At the same time, the actual detection results corresponding to these historical data are collected, such as the actual situation of the personnel at that time, the accuracy record of the alarm triggering, etc. The historical operating parameters are associated with the actual detection results to generate a historical parameter comparison data set. The data set is arranged in chronological order, and each data unit contains the system parameters at a certain moment and the corresponding actual situation record.

[0078] The calibration coefficient calculation submodule calculates the calibration coefficient based on the historical parameter comparison data set. First, the deviation relationship between the signal factor in the historical data and the actual detection results is analyzed to determine the calibration weight of the signal factor. Similarly, the deviation relationship between the interference factor and the actual detection results is analyzed to determine the calibration weight of the interference factor. The calculation of the calibration weight needs to take into account the influence of data from different historical periods. The weight of recent data is higher than that of long-term data. After obtaining the signal factor calibration weight (denoted as α) and the interference factor calibration weight (denoted as β), the coverage matching value of the currently deployed node is calibrated using the formula as follows:

[0079] C'=α×S+β×I

[0080] Where C' represents the calibrated coverage matching value, S represents the normalized signal value of the currently deployed node, and I represents the normalized interference value of the currently deployed node. α and β are calibration coefficients, and α + β = 1, ensuring that the calibrated value remains within a reasonable range. This formula is used to calculate the calibrated coverage matching value for each deployed node, forming a calibrated coverage matching value sequence. Finally, the coverage matching values ​​in the original node deployment set are replaced with the newly generated coverage matching value sequence to complete the update of the node deployment set. The updated set will serve as the basic data for subsequent system operations.

[0081] Example 4:

[0082] The system includes an anomaly tracing module, which consists of a trajectory tracking submodule and a responsibility segment marking submodule. The trajectory tracking submodule receives data on a set of anomaly points approaching a person and extracts the spatial location information and corresponding time node information for each anomaly point. The spatial location information includes the anomaly point's three-dimensional coordinates, accurate to the centimeter level, and the time node information, accurate to the second, recording the specific moment the anomaly point was identified. Based on this information, the submodule retrieves the continuous coverage record set and the coverage fluctuation data set for the corresponding time period from the system's stored historical data. The continuous coverage record set contains real-time records of coverage changes and signal strength fluctuations for each deployment node during the period. The coverage fluctuation data set records the specific values ​​of the coverage change ratio and the sensor sensitivity change ratio. The trajectory tracking submodule chronologically organizes this data to reconstruct the signal strength and interference fluctuation trajectory of the anomaly point from the time the signal anomaly is first detected to the time it is marked as an anomaly. The signal strength change trajectory forms a continuous curve with time as the horizontal axis and signal strength value as the vertical axis, visually demonstrating the rise and fall of signal strength over time. The interference fluctuation trajectory, similarly with time as the horizontal axis and interference value as the vertical axis, displays the fluctuations of electromagnetic interference. These two trajectories are integrated together to generate outlier trajectory tracing data. The data contains the signal strength and interference value corresponding to each time point, as well as the slope change of the trajectory, which reflects the rate of signal change.

[0083] The responsible segment marking submodule operates based on the outlier trajectory tracing data. It first extracts the boundary markers between stable and fluctuating signal segments from the data. Boundary markers include spatial coordinates and time nodes. The spatial coordinates mark the physical dividing point between stable and fluctuating segments, while the time nodes mark the specific moment when the signal state changes from stable to fluctuating or vice versa. The submodule analyzes the specific locations where the outlier trajectory crosses these boundaries, determining whether the signal change at the outlier occurs within the stable segment, within the fluctuating segment, or across both segments. If the outlier trajectory lies entirely within the stable signal segment, the submodule analyzes whether the node parameters within the stable segment are abnormal. If it lies entirely within the fluctuating signal segment, the submodule examines whether the amplitude and frequency of the fluctuations exceed the normal range. If the trajectory spans two segments, the submodule records the signal change characteristics at the time of boundary crossing, such as the sudden change in signal strength and the peak value of the interference fluctuation. Based on these analyses, the primary impacting segment causing the anomaly is determined. For example, if the signal amplitude of the anomaly point is the largest after crossing the boundary between the stable and fluctuating segments, the fluctuating segment is marked as the primary impacting segment. If the signal exhibits sustained abnormal changes within the stable segment, the stable segment is marked as the primary impacting segment. The primary impacting segment is then labeled with responsibility, including the segment number, the involved deployment node numbers, the segment's start and end times, and the spatial range. This ultimately generates the labeling results for the responsible segment for the anomaly point.

[0084] During the node deployment module's operation, the boundary information collection submodule continuously updates the coordinate position data of the rooftop edge boundary. Because the rooftop edge area may undergo subtle changes due to environmental fluctuations (such as temporary obstacles moved by wind or subtle changes in the boundary's physical form caused by rain erosion), the submodule periodically activates the positioning device to resample and obtain the latest boundary coordinates, ensuring the accuracy of the coordinate data. Furthermore, the module integrates real-time node deployment time adjustments. For example, if a deployment node is temporarily moved for maintenance, its deployment time record is promptly updated. Furthermore, signal strength and electromagnetic interference data are dynamically collected, with the collection frequency adjusted based on the severity of environmental changes. The frequency can be appropriately reduced during periods of stable electromagnetic conditions and increased during periods of potentially drastic electromagnetic changes, such as thunderstorms or the operation of nearby large equipment. This dynamic collection method ensures that the deployment node environmental factor data set reflects the current signal and interference status in real time. Each data point in the data set is timestamped, making it easy to trace the data collection moment.

[0085] The node parameter configuration submodule uses a sliding window approach to normalize signal factor and interference factor data in real time. The size of the sliding window is set based on the data collection frequency. For example, if data is collected every 10 seconds, the window size can be set to include the most recent 10 data collections. Within each window, the maximum and minimum values ​​of the signal factor and interference factor are recalculated and used as the reference range for normalization. Normalization is then performed on the most recent data within the window. This approach avoids the issue of normalization standards becoming outdated due to long-term environmental changes and ensures the accuracy of coverage match value calculations. For example, if electromagnetic interference in a particular area remains high for a long time, the sliding window gradually adapts to this change, ensuring that the normalized interference factor more accurately reflects the actual interference situation. By updating the normalization results in real time, parameters such as the coverage match value and normalized deployment factor in the node deployment set can promptly reflect the current actual environmental conditions, providing reliable data support for subsequent system processing.

[0086] Example 5:

[0087] The system includes a result output module, which consists of a multi-dimensional integration sub-module and a structured output sub-module. The multi-dimensional integration sub-module receives the results of the detection and alarm triggering of the approaching personnel at the edge, the calibrated node deployment set generated by the parameter calibration module, and the anomaly point responsibility section labeling results obtained by the anomaly tracing module. A list of detection anomaly points is extracted from the results of the detection and alarm triggering of the approaching personnel at the edge. The list contains information such as the number, spatial coordinates, and sensing response value of all points marked as having detection anomalies and with the risk of false alarms; a calibration coverage matching value sequence is extracted from the calibrated node deployment set. The sequence is sorted by the deployment node number and records the calibrated coverage matching value of each node; and the responsibility section labeling information is extracted from the anomaly point responsibility section labeling results, including the main affected section number corresponding to each anomaly point, the nodes involved in the section, and the spatiotemporal range of the section. The multi-dimensional integration sub-module associates and integrates these three types of information according to the three dimensions of test results, calibration parameters, and traceability conclusions. For example, a certain abnormal detection point is bound to its corresponding calibration coverage matching value and the information of the responsible section to form a complete comprehensive record. All comprehensive records are summarized to generate a comprehensive output data set, which is stored in a table format. Each record contains the basic information of the abnormal detection point, the calibrated parameter value, and the details of the responsible section obtained by tracing.

[0088] The structured output submodule processes the comprehensive output dataset. It first categorizes the list of detected anomaly points by location segment. Location segments are divided based on the physical structure of the roof edge, such as the east edge segment, south edge segment, and corner segment. Detected anomaly points belonging to the same location segment are grouped together and sorted within each group by their proximity to the boundary, with closer points appearing first. The calibration coverage match value sequence is sorted by time period, which can be set to hours, days, or weeks. The calibration coverage match values ​​within each time period are arranged in order of deployment node number, forming the basis for a time series curve, facilitating the observation of how coverage match values ​​at different nodes change over time. Responsible segment marker information is sorted by impact, determined based on factors such as the number of anomalies within the segment and the magnitude of the anomaly signal intensity. Responsible segments with higher impact are ranked higher. Each responsible segment lists the anomalies it contains and their corresponding detection information.

[0089] The output format uses a combination of tables and text. The table section contains three main tables: a classification table of detected anomaly points, with headers including location segment, point number, spatial coordinates, sensor response value, and false alarm risk level; a time series table of calibration coverage match values, with headers including time period, node number, and calibration coverage match value; and a ranking table of responsible sections, with headers including the responsible section number, impact level, number of included anomalies, and affected nodes. The text section provides supplementary explanations of the table content, such as the reasons for the concentration of detected anomaly points in a particular location segment, the overall trend of calibration coverage match value changes, and the basis for determining the impact level of the responsible section. The explanations are concise and clear, avoiding redundant phrasing. This approach presents the information in the comprehensive output dataset in a clear and organized manner, generating the final comprehensive results of edge person approach detection and alarms. These results can be directly provided to rooftop safety managers for the development of appropriate safety management measures.

[0090] When synchronizing signals between adjacent nodes, the signal synchronization module dynamically adjusts the adjacent time threshold based on electromagnetic environment characteristics at different times of day. These characteristics include electromagnetic interference intensity and signal transmission stability. For example, during daytime on weekdays, when electromagnetic interference is high due to the presence of numerous electrical devices, signal transmission is easily affected. In this case, the adjacent time threshold is lowered, such as to 50 milliseconds, to improve signal synchronization accuracy. During nighttime and holidays, when the electromagnetic environment is relatively stable, the adjacent time threshold can be increased, such as to 100 milliseconds, to reduce the system's computational load. The consistency annotation submodule optimizes the classification and annotation algorithm by combining segment characteristics from historical data to label stable and fluctuating signal segments. Segment characteristics in historical data include the duration of stable and fluctuating signal segments and the amplitude of signal fluctuations across different seasons and weather conditions. The algorithm learns these characteristics and adjusts the threshold for determining whether the direction of signal change is consistent with that of interference. For example, in rainy environments prone to signal fluctuations, the threshold for determining stable signal segments is appropriately raised to reduce false positives. This optimization improves the reliability of the signal consistency synchronization annotation set, ensuring that the annotation results are more consistent with actual electromagnetic environment variations.

[0091] The coverage overlay calculation submodule in the field division module introduces a spatial weight factor when calculating the position offset rate and the sensing range expansion rate. The spatial weight factor is set according to the importance of different areas. For areas with high importance, such as corners at the edge of the roof and dangerous areas where people are easily approached, the spatial weight factor value is larger. When calculating the rate, the position offset and sensing range expansion of these areas are given higher weights, so that the calculation results can better reflect the coverage changes in key areas. For areas with lower importance, such as flat areas far from the edge, the spatial weight factor value is smaller, reducing its impact in the overall coverage analysis. By introducing the spatial weight factor, the results of the field coverage overlay analysis are more in line with the actual safety management needs of the roof edge, highlighting the coverage changes in key areas, and providing more targeted data support for subsequent proximity judgment and alarm triggering.

[0092] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0093] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A distributed rooftop edge personnel approaching sensing alarm system based on wireless networking, characterized by: The system comprises: The node deployment module obtains the coordinates of the roof edge boundary and the node deployment time, collects the signal strength and electromagnetic interference value of the corresponding location, associates the deployed nodes, calculates the corresponding coverage matching value, and generates a node deployment set; The signal synchronization module extracts the coverage matching values ​​and corresponding coordinates in the node deployment set, synchronizes the signals of adjacent nodes in chronological order, marks the stable and fluctuating sections in the adjacent nodes, and obtains a signal consistency synchronization annotation set; The field division module obtains the deployment nodes located in the signal stable section of the signal consistency synchronization annotation set, extracts the node location and sensing range, compares the coverage range based on the coverage overlap, evaluates the sensing impact strength under signal fluctuations, and generates field coverage overlay analysis results; The proximity discrimination module identifies abnormal points in the deployment nodes whose sensing response values ​​are greater than the average response reference value and are in the signal fluctuation section in the field coverage overlay analysis results, and forms a person proximity abnormal point set; The alarm trigger module obtains all points where people are close to abnormal points and corresponding point information, marks points with false alarm risks, and generates edge person approach detection and alarm triggering results.

2. The distributed rooftop edge personnel approaching sensing alarm system based on wireless networking according to claim 1 is characterized in that: The node deployment set includes coverage matching value, deployment node spatial coordinates, and normalized deployment factor. The signal consistency synchronization annotation set specifically includes signal stable segment annotation, signal fluctuation segment annotation, and adjacent deployment node coverage matching value difference rate. The field coverage overlay analysis results include the degree of influence of node position offset on coverage, the degree of influence of sensing range expansion on coverage, and the comparison of coverage response under each signal fluctuation condition. The personnel approaching abnormal point set includes the spatial position of the abnormal point, the abnormal point signal strength variation characteristics, and the abnormal point coverage and sensing fluctuation ratio. The edge personnel approach detection and alarm triggering results include a list of detected abnormal points and a personnel approaching abnormal point set.

3. The distributed rooftop edge personnel approaching sensing alarm system based on wireless networking according to claim 1 is characterized in that: The node deployment module includes: The boundary information collection submodule obtains the coordinate position of the roof edge boundary and the node deployment time, collects the signal strength data corresponding to the coordinate position and the electromagnetic interference data of the day, and records the collection results as two deployment factors, signal factor and interference factor, to obtain the deployment node environmental factor data set; The node parameter configuration submodule performs normalization processing on the signal factor and interference factor data in the deployment node environmental factor data group, establishes a correspondence between the normalized results and the deployment node coordinate positions, calculates the average value of the normalized signal value and the normalized interference value as the coverage matching value, and generates a node deployment set.

4. The distributed rooftop edge personnel approaching sensing alarm system based on wireless networking according to claim 3 is characterized in that: The signal synchronization module includes: The synchronization benchmark extraction submodule obtains the coverage matching values ​​and corresponding coordinate data in the node deployment set, identifies the time synchronization relationship of all deployed nodes based on the coordinate information, calls the deployment node time set, and uses the adjacent time threshold as a benchmark to perform signal synchronization measurement and sorting on the time deployed nodes to generate a time sorting sequence of adjacent deployment nodes; The deviation calibration submodule uses a stable synchronization reference value and a fluctuating synchronization reference value to compare the time sequence of the adjacent deployment nodes, and integrates and generates a signal synchronization deviation sequence; The consistency annotation submodule extracts the signal change direction and interference change direction between adjacent deployment nodes based on the signal synchronization deviation sequence, classifies and annotates each pair of deployment nodes according to whether the change trends in the two directions are consistent, records and groups the signal stable and signal fluctuating sections respectively, and obtains the signal consistency synchronization annotation set.

5. The distributed rooftop edge personnel approaching sensing alarm system based on wireless networking according to claim 4 is characterized in that: The field division module includes: The field feature extraction submodule selects the segments marked as signal stable according to the signal consistency synchronization annotation set, detects the position data and sensing range data of each deployment node within the coverage period, arranges them in spatial order to form a position space sequence and a sensing range sequence, and generates a field coverage space sequence set; The coverage overlay calculation submodule calculates the position offset rate and sensing range expansion rate between consecutive spatial nodes in each deployment node spatial sequence based on the field coverage spatial sequence set, compares the position offset rate and sensing range expansion rate in parallel under the same coverage conditions, and identifies the numerical relationship between the change amplitude under the coverage range by jointly analyzing the two types of rate indicators. It integrates the influence value sequence of each deployment node and establishes the field coverage overlay analysis result.

6. The distributed rooftop edge personnel approaching sensing alarm system based on wireless networking according to claim 5 is characterized in that: The proximity discrimination module includes: The sequence extraction submodule selects deployment nodes whose sensing response values ​​are greater than the average response reference value and deployment nodes in the signal fluctuation section based on the field coverage superposition analysis results, extracts continuous coverage records of the deployment nodes in spatial order, collects coverage range and sensing sensitivity data corresponding to each spatial node, and generates a continuous coverage record set; The fluctuation change ratio calculation submodule calls the continuous coverage record set, extracts the coverage value and sensing value of the deployment node at two consecutive spatial nodes, calculates the coverage change ratio and sensing sensitivity change ratio, integrates them into the coverage change ratio sequence and the sensing sensitivity change ratio sequence, and establishes a coverage fluctuation change data set; Based on the coverage fluctuation change data set, the anomaly identification submodule extracts the signal strength variation data and interference fluctuation data of the corresponding spatial segment, determines whether the coverage change ratio and the sensing sensitivity change ratio both exceed the set fluctuation identification threshold, and determines whether the signal strength variation and the interference fluctuation simultaneously exceed the anomaly judgment threshold. The spatial nodes that meet the conditions are marked as anomaly points, and a set of people approaching anomaly points is generated.

7. The distributed rooftop edge personnel approaching sensing alarm system based on wireless networking according to claim 6 is characterized in that: The alarm triggering module includes: The indicator joint judgment submodule obtains all the points where the person is close to the abnormal point concentration and the corresponding coordinates and identification information, uses the stable response threshold and the fluctuation response threshold to compare them respectively, and establishes a joint response judgment value sequence; Based on the joint response judgment value sequence, the abnormal output sorting submodule screens points whose sensing response values ​​are greater than the sensing response risk threshold, whose coverage matching values ​​are lower than the coverage reference value, and whose signal consistency labels are fluctuation segments. The corresponding point numbers, location identifiers, and segments to which they belong are extracted, and marked as detection anomalies with false alarm risks. The points that meet the joint conditions are segmented and output in a structured format to generate edge person approach detection and alarm triggering results.

8. The distributed rooftop edge personnel approaching sensing alarm system based on wireless networking according to claim 7 is characterized in that: The system further includes a parameter calibration module, which includes: The historical data retrieval submodule obtains the node deployment set, signal consistency synchronization annotation set and field coverage overlay analysis results within the historical period, extracts the historical operating parameters and actual detection results of each module, and generates a historical parameter comparison data set; The calibration coefficient calculation submodule calculates the calibration weights of the signal factor and the interference factor based on the historical parameter comparison data set, performs weighted adjustment on the calibration weights and the normalized signal value and normalized interference value of the currently deployed node, generates a calibrated coverage matching value sequence, and updates the node deployment set.

9. The distributed rooftop edge personnel approaching sensing alarm system based on wireless networking according to claim 8 is characterized in that: The system further includes an abnormality tracing module, which includes: The trajectory tracking submodule obtains the spatial position and time node information of the person approaching the abnormal point in the abnormal point set, retrieves the continuous coverage record set and coverage fluctuation change data set within the corresponding time period, restores the signal strength change trajectory and interference fluctuation trajectory of the abnormal point in chronological order, and generates abnormal point trajectory tracing data; The responsible section marking submodule extracts the boundary markers of the signal stable section and the signal fluctuating section based on the abnormal point trajectory tracing data, determines the specific location where the abnormal point trajectory crosses the section, marks the responsibility for the main impact section causing the abnormality, and generates the abnormal point responsible section marking result.

10. The distributed rooftop edge personnel approaching sensing alarm system based on wireless networking according to claim 9 is characterized in that: The system further includes a result output module, which includes: The multi-dimensional integration submodule obtains the edge personnel approach detection and alarm triggering results, the calibrated node deployment set of the parameter calibration module, and the abnormal point responsibility section marking results of the abnormality tracing module, extracts the detection abnormal point list, calibration coverage matching value sequence and responsibility section marking information, and integrates the data according to the three dimensions of detection results, calibration parameters, and tracing conclusions to generate a comprehensive output data set; Based on the comprehensive output data set, the structured output submodule classifies the list of detected abnormal points by location segment, arranges the calibration coverage matching value sequence by time period, sorts the responsible section marking information by the degree of influence, and outputs it in a format combining tables and text to generate the final comprehensive results of edge personnel approach detection and alarm.

Images