Abnormal risk intelligent emergency response method and system based on multi-source communication big data

Abstract

The invention discloses an abnormal risk intelligent emergency response method and system based on multi-source communication big data, and belongs to the technical field of communication big data and emergency command. The method comprises the steps that S1, multi-source heterogeneous data collection and gridding modeling are carried out; s2, constructing a risk thermodynamic diagram based on a dynamic weight factor; s3, weak signal passive positioning based on signaling characteristics; s4, communication guarantee type A path planning optimization; acquiring real-time accumulated water depth information of each road section of the road network, eliminating impassable road sections, and reconstructing a cost function of an algorithm A; according to the method, the problems of positioning failure and easy interruption of a command link in a disaster environment are solved, and the communication reliability and risk coverage precision of emergency scheduling are improved.

Description

Technical Field

[0001] This invention belongs to the field of communication big data and emergency command technology, specifically relating to an intelligent emergency response method and system for abnormal risks based on multi-source communication big data. Background Technology

[0002] In current emergency responses to sudden natural disasters (such as rainstorms and earthquakes), rescue and dispatch technologies based on Geographic Information Systems (GIS) and General Navigation Satellite Systems (GNSS) have been widely applied. However, in actual engineering practice and emergency command processes, the following bottlenecks still urgently need to be addressed:

[0003] 1) Low utilization and poor robustness of signaling data: Most existing emergency platforms rely on GPS coordinates actively reported by users' mobile phones. However, under extreme disaster conditions, power outages or base station damage often weaken or disable satellite positioning signals. Although the operator's core network stores a large amount of base station signaling data (such as TA time advance and RSRP signal strength), due to the high discreteness and volatility of this data, there is a lack of effective compensation algorithms to convert it into high-precision geospatial coordinates, resulting in the inability to passively detect target personnel in "isolated islands".

[0004] 2) Risk assessment is limited in scope and lacks spatial and temporal precision: Traditional disaster assessments are mostly based on macro-level statistics of administrative divisions, making it difficult to achieve dynamic risk monitoring at the second or grid level. When dealing with abnormal population movement in densely populated areas, the lack of quantitative calculation of the dynamic evolution of user density at the base station side (such as the density surge rate per unit time) makes it impossible for the command center to accurately predict high-risk points for secondary disasters.

[0005] 3) Disconnection between rescue route planning and communication continuity: Existing route optimization algorithms (such as traditional A) The algorithm (or Dijkstra's algorithm) primarily uses "shortest path" or "fastest time" as its objective function. In extreme scenarios where base stations are widely damaged, if rescue teams enter communication blind spots, it will lead to the interruption of the command and control link between the front and rear. Current systems generally lack a path planning model that can use "communication coverage quality" as a hard weight constraint, and cannot ensure real-time communication support for rescue routes in complex geographical environments. Summary of the Invention

[0006] In response to the problems mentioned in the background technology, this invention proposes an intelligent emergency response method and system for abnormal risks based on multi-source communication big data. Combining the developed "intelligent emergency command platform", it starts from the underlying data architecture and aims to solve the problems of accurate positioning of trapped personnel in weak signal environments and optimal path finding under communication guarantee constraints by deeply mining multi-source communication big data (including TA, RSRP and gridded population density flow).

[0007] Technical Solution: To solve the above-mentioned technical problems, the present invention adopts the following technical solution:

[0008] An intelligent emergency response method for abnormal risks based on multi-source communication big data includes the following steps:

[0009] S1: Multi-source heterogeneous data acquisition and gridded modeling;

[0010] The system utilizes a data access unit to acquire polygonal data for meteorological warnings, road network GIS data, and communication signaling data. Based on the spatial grid setting module, the target area is divided into geographic grid units with a side length of L, and the temporary user identifiers in the communication signaling data are mapped to the corresponding grid units.

[0011] S2: Construction of a risk heatmap based on dynamic weighting factors;

[0012] Real-time calculation of each grid cell in the sliding window The user density change rate and movement speed deviation within the grid are calculated; the risk factor configuration interface is called to obtain the preset density change weight, speed deviation weight and population attribute weight, and to calculate the grid comprehensive risk value;

[0013] S3: Passive localization of weak signals based on signaling characteristics;

[0014] The status of base stations is monitored. When it is identified that the base station in the target area may be damaged or the RSRP signal strength is lower than the threshold, the time advance value of the surviving base station is extracted. Based on the mapping relationship between the time advance value and the distance, a multi-circle intersection positioning model is constructed.

[0015] S4: Communication guarantee type A Path planning optimization;

[0016] Obtain real-time water depth information for each road segment of the road network, eliminate impassable road segments, and reconstruct A. The cost function of the algorithm.

[0017] As a preferred option, the specific implementation details in S1 are as follows:

[0018] Weather warning polygon data: including warning type, level, geographic polygon boundary, and effective time window;

[0019] Road network GIS data: including road network topology, road attributes, and real-time water depth obtained from IoT sensors;

[0020] Communication signaling data: Real-time signaling streams of 4G / 5G base stations obtained from the operator's core network, including timing advance, reference signal received power, serving cell ID, and de-identified temporary user identifiers.

[0021] As a preferred option, the specific calculation process in S2 is as follows:

[0022] S21. Data Mapping: Map the polygonal range of the weather warning and the coverage range of the base station sector to the geographic grid respectively, and calculate the weather risk coverage rate of each grid.

[0023] S22. Density and Velocity Field Calculation: Aggregate the number of active users within the grid using a second-level sliding window to calculate the real-time user density of the grid. The average mobile speed within the grid is calculated based on the handover time difference between adjacent base stations. ;

[0024] S23. Risk Value Quantification: Based on the risk factor configuration module, calculate the comprehensive risk value for each grid cell.

[0025] As a preferred option, the specific implementation details in S23 are as follows:

[0026] For each grid cell Calculate the overall risk value The calculation logic formula is as follows:

[0027] ;

[0028] in, Indicates the time window of density change; Indicates the preset maximum movement speed; α represents the weight of population attributes; β represents the weight of density change; β represents the weight of velocity change, set to 0.20; γ represents the weight of population attributes; D t D represents the user density within the grid at time t; t-1 This represents the user density within the grid at time t-1.

[0029] As a preferred option, the specific implementation process in S3 is as follows:

[0030] S31. Abnormal Status Identification: The system scans the base station status in real time. When it detects a sudden change in the time advance value exceeding the threshold or a continuous attenuation of the reference signal received power, it is determined to be a damaged base station or an abnormal signal area.

[0031] S32. Compensation positioning calculation: Extract the list of surviving base stations around the anomaly point, convert the physical distance through the time advance value, construct a multi-circle intersection positioning model, and then introduce the RSRP attenuation model to perform weighted correction on the coordinate cluster information;

[0032] S33. Behavior Deviation Verification: Call the historical behavior analysis module to calculate the ratio of the target user's current dwell time to the historical average dwell time; when the deviation exceeds the set threshold, mark the coordinate cluster as a high-confidence trapped point.

[0033] As a preferred option, the specific calculation process in S4 is as follows:

[0034] S41. Constraint definition: including environmental constraints and communication guarantee constraints;

[0035] Environmental constraints are introduced into the path search to remove road sections with real-time water depth greater than a threshold and to remove areas predicted for secondary disasters.

[0036] Communication assurance constraints: If a path segment is covered by a surviving base station, delete risky paths that are not covered by surviving base stations;

[0037] S42. Cost Function Reconstruction: Reconstructing A The cost function of the algorithm;

[0038] S43. Objective function weight configuration: Set the path travel time weight and communication coverage intensity weight;

[0039] S44. Output the optimal rescue route with communication support, including the number of road segments, estimated travel time, communication coverage, and avoidance of uncovered road segments.

[0040] As a preferred option, in S42, A is reconstructed. Algorithm cost function for:

[0041] ;

[0042] Actual cost The newly added communication cost item is as follows:

[0043] ;

[0044] Where h(n) represents the estimated cost from the current node n to the target point; This represents the communication cost weighting coefficient; This represents the cumulative actual cost from the starting point to the parent node; The cost indicating whether the current road segment is within the communication coverage area; This represents the travel time cost of the current node n on the road segment; This represents the actual cost from the starting point to the current node n.

[0045] An intelligent emergency response system for abnormal risks based on multi-source communication big data is proposed, implementing the intelligent emergency response method for abnormal risks of multi-source communication big data as described in any of the above items. The system adopts a B / S architecture and includes:

[0046] Data layer: Used to store base station topology data, RSRP / TA measured data, and road network constraint data;

[0047] Core layer: includes risk factor configuration module, passive positioning calculation module, and path planning optimization module;

[0048] Presentation layer: used for visual rendering of dynamic risk heat maps, distribution maps of damaged base stations, and generated communication-supported rescue routes.

[0049] Beneficial effects: Compared with the prior art, the present invention has the following advantages:

[0050] (1) This invention proposes a passive positioning method for weak signals based on signaling features. By extracting the TA value of surviving base stations, a multi-circle intersection positioning model is constructed. The RSRP attenuation model and the deviation of user historical behavior are introduced for verification, thereby achieving high-confidence passive positioning of trapped personnel in the environment of base station damage. At the same time, by reconstructing A The algorithm's cost function takes communication coverage as a hard constraint to ensure reliable communication throughout the rescue route, thus systematically solving the problems of positioning failure and command link interruption under extreme disasters.

[0051] (2) This invention integrates meteorological early warning polygon data with gridded modeling communication signaling data to calculate the user density change rate and movement speed deviation in each grid cell in real time. Combined with configurable dynamic weighting factors, it constructs a dynamic risk heat map that is updated in seconds, thereby achieving accurate identification and spatial visualization of high-risk lingering areas and abnormal population flows. Attached Figure Description

[0052] Figure 1 This is a diagram of the intelligent emergency response system architecture for abnormal risks based on multi-source communication big data of the present invention.

[0053] Figure 2 This is a diagram of the damaged base station and abnormal signal identification interface of the intelligent emergency command platform of the present invention;

[0054] Figure 3 This is a diagram of the meteorological early warning data management interface of the intelligent emergency command platform of the present invention;

[0055] Figure 4 This is a diagram of the road network GIS data management interface of the intelligent emergency command platform of the present invention;

[0056] Figure 5 This is a diagram of the spatial grid setting interface of the intelligent emergency command platform of the present invention;

[0057] Figure 6 This is a diagram of the weak signal passive positioning instrument panel interface of the intelligent emergency command platform of the present invention;

[0058] Figure 7 This is a diagram of the communication support path planning interface of the intelligent emergency command platform of the present invention;

[0059] Figure 8 This invention relates to the intelligent emergency command platform A. Algorithm parameter configuration interface screenshot;

[0060] Figure 9 This is a flowchart of the intelligent emergency response method for abnormal risks of the present invention;

[0061] Figure 10 This is a schematic diagram of the construction of a grid-based dynamic risk heat map according to the present invention;

[0062] Figure 11 This is a schematic diagram of the passive weak signal localization process based on signaling features according to the present invention;

[0063] Figure 12 The communication protection mechanism of this invention is A. Path planning diagram. Detailed Implementation

[0064] The present invention will be further illustrated below with reference to specific embodiments. These embodiments are implemented based on the technical solutions of the present invention, and it should be understood that these embodiments are only used to illustrate the present invention and are not intended to limit the scope of the present invention.

[0065] Example 1

[0066] This embodiment provides an intelligent emergency response method for abnormal risks based on multi-source communication big data. This method is an emergency response method for abnormal risks based on actual system test data. It is illustrated with actual operation data of the "Intelligent Emergency Command Platform". The platform front-end is developed using the Vue.js framework, and the back-end integrates a stream processing engine.

[0067] This method is executed based on a collaborative architecture of a data access layer, a core engine layer, and a decision output layer, such as... Figure 9 As shown, the specific steps include:

[0068] S1. Multi-source heterogeneous data acquisition and gridded modeling;

[0069] The following three types of basic data streams are obtained through the data access interface:

[0070] The system uses a data access unit to acquire meteorological warning polygon data, road network GIS data, and communication signaling data. Based on the spatial grid setting module, the target area is divided into geographic grid units with a side length of L (set to 500 meters in this embodiment), and the user temporary identifier in the communication signaling data is mapped to the corresponding grid unit.

[0071] Weather warning polygon data: including warning type, level, geographic polygon boundary and effective time window.

[0072] like Figure 3As shown, the system accesses real-time data through the base station signaling data management module. Imported through the meteorological early warning data management module, the currently active warning is "Heavy Rain (Red)," covering the city's core area and surrounding low-lying areas (grid coverage number xxx).

[0073] Communication signaling data: Real-time signaling streams of 4G / 5G base stations are obtained from the signaling acquisition probes on the operator's core network side, including timing advance (TA), reference received signal power (RSRP), serving cell ID (ECGI), and de-identified temporary user identifier (TMSI).

[0074] like Figure 2 As shown, when accessing 4G / 5G base station signaling, the interface displays the total number of monitored base stations as xxx and the number of alarms as x. For "xx base station" (ID: xx, location: xxx), the system identifies its status as "damaged", the fault time as xx, and the abnormal signal characteristic as "connection lost".

[0075] Road network GIS data (optional): Includes road network topology, road attributes, and real-time water depth data obtained from IoT sensors.

[0076] like Figure 4 As shown, the road network base map is loaded through the "Road Network GIS Data Management" module (RoadNetworkMap.vue), which contains xx road segments and covers an area of ​​xx square kilometers.

[0077] S2. Construction of risk heatmap based on dynamic weighting factors;

[0078] Use the Spatial Grid Settings module to divide the target area into standard geographic grid cells with a side length of L (recommended value 500 meters), such as... Figure 10 As shown, perform the following calculations:

[0079] S21. Data Mapping: Map the polygonal range of the weather warning and the coverage range of the base station sector to the geographic grid respectively, and calculate the weather risk coverage rate of each grid.

[0080] In the "Risk Factor Configuration" interface, the system administrator configures the calculation parameters according to the current rainstorm disaster type:

[0081] like Figure 5 As shown, in the "Spatial Grid Settings" module, the target area (minimum longitude xx, maximum longitude xx) is divided into a standard grid with a side length of xx meters, the total number of grids is xx, and the data aggregation time window is set to xx seconds.

[0082] S22. Density and Velocity Field Calculation: Aggregate the number of active users within the grid using a sliding window on the order of seconds (e.g., 10 seconds) to calculate the real-time user density of the grid. The average mobile speed within the grid is calculated based on the handover time difference between adjacent base stations. ;

[0083] User density D t The calculation formula is: D t =N users / Area; where N users This indicates the number of unique users in the grid within the current second-level time window; Area represents the grid area.

[0084] Density change rate The calculation formula is:

[0085]

[0086] in, D represents the time window for density changes. t D represents the user density within the grid at time t; t-1 This represents the user density within the grid at time t-1.

[0087] S23. Risk Value Quantification: Based on the Risk Factor Configuration module, for each grid cell... Calculate the overall risk value The calculation logic formula is as follows:

[0088]

[0089] in, The density change time window (default 10 seconds); The preset maximum movement speed; For population attribute weights; The weighting coefficients for each factor, i.e., the measured parameters, are configured as follows: the weight for density change α is set to 0.50, the weight for velocity change β is set to 0.20, and the weight for population attributes γ is set to 0.30. t D represents the user density within the grid at time t; t-1 This represents the user density within the grid at time t-1.

[0090] Finally, the results are output as a dynamic risk heatmap. The system generates a "Dynamic Risk Heatmap Dashboard," which displays the current total risk score (xx), the number of high-risk grids (xx), and the areas mainly concentrated in regions where the user density change rate reaches xx%.

[0091] The map grid is based on the risk value calculated in real time. The map is colored, with red representing high-risk areas. This view is the primary basis for the command center to assess regional risk situations, and it can visually present potential risk points such as abnormal population gatherings or stagnant movement.

[0092] S3. Passive localization of weak signals based on signaling characteristics;

[0093] The status of base stations is monitored. When it is identified that the base station in the target area may be damaged or the RSRP signal strength is lower than the threshold, the time advance value of the surviving base station is extracted. Based on the mapping relationship between the time advance value and the distance, a multi-circle intersection positioning model is constructed.

[0094] A historical behavior analysis module is introduced to calculate the deviation of the current stay time by retrieving the average historical stay time of the target user over the past 30 days. When the deviation exceeds a preset threshold, the location result is marked as a high-confidence trapped point.

[0095] In this embodiment, the historical behavior deviation is calculated based on a stay duration of 30 minutes. The deviation threshold is set to 3 times based on experience, that is, when the stay duration exceeds 90 minutes and the area is in an abnormal signal zone, an alarm is triggered.

[0096] In weak signal environments where satellite positioning fails, compensatory positioning can be attempted using base station physical layer parameters. When base station xx is detected to be damaged, the system automatically triggers the "weak signal passive positioning" process, such as... Figure 11 As shown, the process is as follows:

[0097] S31. Abnormal Status Identification: The system scans the base station status in real time. When it detects a sudden change in TA value exceeding the threshold (e.g., 3 time slots) or continuous attenuation of RSRP, it is determined to be a damaged base station or an abnormal signal area.

[0098] Validation of algorithm parameters: In the "Real-time Algorithm Test" interface, input test data TA value = 100 (corresponding to a distance of approximately xxkm) and RSRP value = -100dBm to verify the effectiveness of the triangulation logic.

[0099] S32, Compensation positioning solution;

[0100] Input: Extract a list of surviving base stations (e.g., BS_xx, BS_xx) around the anomaly point, including TA and RSRP data.

[0101] TA distance conversion: Convert the physical distance d using the TA value to construct a multi-circle intersection positioning model.

[0102] The formula for calculating the physical straight-line distance d between the user equipment and the base station is: (rice).

[0103] Weighted correction: The initial coordinates are solved using the least squares method, and the RSRP decay model is introduced to perform weighted correction on the coordinate cluster information.

[0104] S33. Behavioral Deviation Verification: Call the historical behavior analysis module to calculate the ratio of the target user's current dwell time to the historical average dwell time (e.g., 30-day average). When the deviation exceeds the set threshold (the system default is 90 minutes), the coordinate cluster is marked as a high-confidence trapped point.

[0105] Parameter settings: The historical data window is set to the last 30 days, the average stay time is set to 30 minutes (1800s), and the deviation threshold is set to 3 times (i.e., >90 minutes is considered abnormal).

[0106] Fusion Computation: Set the fusion weight to xx%.

[0107] S34. Output Results: Output the coordinate cluster of the trapped point with high confidence; in the "Coordinate Cluster and Details Output of Trapped Personnel" interface, the system identifies the target personnel TP001 (location: xx), whose dwell time varies by xx minutes, and whose status is marked as "abnormal," with a location confidence level of xx%. Figure 6 As shown.

[0108] S4, communication guarantee type A Path planning;

[0109] For rescue scenarios, the traditional shortest path search algorithm is improved by incorporating communication coverage as a hard constraint, such as... Figure 12 As shown, it specifically includes:

[0110] S41. Constraint definition: including environmental constraints and communication guarantee constraints;

[0111] Environmental constraints are introduced into the path search to remove road sections with real-time water depth greater than a threshold (e.g., 50cm, information dependent on synchronization with other departments) and to remove areas predicted for secondary disasters.

[0112] Communication assurance constraints: The path segment must be covered by surviving base stations; risky paths without surviving base station coverage are deleted.

[0113] S42. Cost Function Reconstruction: Reconstructing A Algorithm cost function The actual cost New communication cost item added:

[0114]

[0115] in, This is the communication cost weighting coefficient (its value range is empirically defined as 1.2 to 1.8, and the system default configuration is 1.50). The value is defined as follows: if the road segment is covered by a surviving base station, it is 0; otherwise, it is the physical length of the road segment. This forces the algorithm to prioritize searching for paths with good communication coverage, and only traverse short-distance blind spots when there is no solution. This represents the cumulative actual cost from the starting point to the parent node; The cost indicating whether the current road segment is within the communication coverage area; This represents the travel time cost of the current node n. h(n) represents the estimated cost from the current node n to the target point. This represents the actual cost from the starting point to the current node n.

[0116] Communication assurance constraint: The path segment must be covered by at least one surviving base station.

[0117] S43. Objective function weight configuration: Improved A The algorithm parameters are configured as follows: maximum possible vehicle speed Communication cost weighting coefficient .

[0118] S44. Output Communication-Guaranteed Optimal Rescue Route: The system-generated visual route display shows that the planned optimal rescue route includes x road segments, with an estimated travel time of xx minutes and a communication coverage rate of xx%, effectively avoiding uncovered road segments (number of uncovered road segments: x). For example... Figure 7 As shown.

[0119] Example 2

[0120] like Figure 1 As shown, this embodiment also proposes an intelligent emergency response system for abnormal risks based on multi-source communication big data. This system is a web-based intelligent emergency response system, adopting a B / S architecture. The front-end is built based on the Astro and Vue 3 frameworks, and its main functional modules include:

[0121] Data layer: Used to store base station topology data (base station ID, location, coverage area, ECG cell identity), RSRP / TA measured data, and road network constraint data (road network topology, road attributes, real-time water depth data).

[0122] The TaRsrpDataInput.vue component provides a file import interface, supporting the uploading of TA / RSRP data and base station coordinate data in CSV / Excel format. Drag and drop upload is supported, with a file size limit of 50MB.

[0123] Weather warning data can be managed through WeatherWarningContent.vue, which supports viewing the warning type, level (such as red, orange) and effective time.

[0124] The core layer includes the RiskFactorConfigForm module, the Passive Positioning Calculation module, and the PathPlanningDashboard module.

[0125] Specifically:

[0126] The heatmap engine, DynamicHeatmapDashboard.vue, is responsible for rendering the real-time risk map, and RiskFactorConfigForm.vue is integrated for real-time adjustments. Weight parameters.

[0127] The location analysis engine, DamagedBaseStationContent.vue, is responsible for identifying damaged base stations (such as those with severe RSRP attenuation), while TrappedPersonnelDashboard.vue is responsible for fusion calculations and outputting a list of trapped personnel.

[0128] Algorithm configuration engine: AlgorithmOptimizationContent.vue provides support for A... Algorithm core parameters (such as) ) has graphical configuration capabilities.

[0129] Presentation layer: used for visual rendering of dynamic risk heat maps, distribution maps of damaged base stations, and generated communication-supported rescue routes.

[0130] Visualization output: GeneratedPathVisualization.vue overlays the calculated path onto the GIS base map and uses different colors to distinguish the path risk level (e.g., red warning for high-risk paths).

[0131] Metric Monitoring: The "Algorithm Optimization Settings" interface displays real-time algorithm performance metrics (search efficiency, path optimality, communication coverage), with the current system performance score displayed as xx points (performance xx). For example... Figure 8 As shown.

[0132] Rescue Execution Layer: The emergency command center uses a unified navigation menu built with AppSidebarLayout.vue to dispatch various functional modules in real time, and the rescue vehicle terminals receive navigation instructions containing communication support information generated by PathPlanningDashboard.

[0133] 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 principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. An intelligent emergency response method for abnormal risks based on multi-source communication big data, characterized in that, Includes the following steps: S1: Multi-source heterogeneous data acquisition and gridded modeling; The data access unit is used to acquire meteorological early warning polygon data, road network GIS data, and communication signaling data. The spatial grid setting module divides the target area into geographic grid cells with a side length of L, and maps the temporary user identifier in the communication signaling data to the corresponding grid cell; S2: Construction of a risk heatmap based on dynamic weighting factors; Real-time calculation of each grid cell in the sliding window The rate of change in user density and deviation in movement speed within the area; The interface is configured by calling risk factors to obtain preset density change weights, velocity deviation weights, and population attribute weights, and to calculate the grid comprehensive risk value. S3: Passive localization of weak signals based on signaling characteristics; The status of base stations is monitored. When it is identified that the base station in the target area may be damaged or the RSRP signal strength is lower than the threshold, the time advance value of the surviving base station is extracted. A multi-circle intersection positioning model is constructed based on the mapping relationship between time advance value and distance; S4: Communication guarantee type A Path planning optimization; Obtain real-time water depth information for each road segment of the road network, eliminate impassable road segments, and reconstruct A. The cost function of the algorithm.

2. The intelligent emergency response method for abnormal risks based on multi-source communication big data according to claim 1, characterized in that: In S1, the specific implementation details are as follows: Weather warning polygon data: including warning type, level, geographic polygon boundary, and effective time window; Road network GIS data: including road network topology, road attributes, and real-time water depth obtained from IoT sensors; Communication signaling data: Real-time signaling streams of 4G / 5G base stations obtained from the operator's core network, including timing advance, reference signal received power, serving cell ID, and de-identified temporary user identifiers.

3. The intelligent emergency response method for abnormal risks based on multi-source communication big data according to claim 1, characterized in that: In S2, the specific calculation process is as follows: S21. Data Mapping: Map the polygonal range of the weather warning and the coverage range of the base station sector to the geographic grid respectively, and calculate the weather risk coverage rate of each grid. S22. Density and Velocity Field Calculation: Aggregate the number of active users within the grid using a second-level sliding window to calculate the real-time user density of the grid. The average mobile speed within the grid is calculated based on the handover time difference between adjacent base stations. ; S23. Risk Value Quantification: Based on the risk factor configuration module, calculate the comprehensive risk value for each grid cell.

4. The intelligent emergency response method for abnormal risks based on multi-source communication big data according to claim 3, characterized in that: In S23, the specific implementation details are as follows: For each grid cell Calculate the overall risk value The calculation logic formula is as follows: ； in, Indicates the time window of density change; Indicates the preset maximum movement speed; α represents the weight of population attributes; β represents the weight of density change; β represents the weight of velocity change, set to 0.20; γ represents the weight of population attributes; D t D represents the user density within the grid at time t; t-1 This represents the user density within the grid at time t-1.

5. The intelligent emergency response method for abnormal risks based on multi-source communication big data according to claim 1, characterized in that: In S3, the specific implementation process is as follows: S31. Abnormal Status Identification: The system scans the base station status in real time. When it detects a sudden change in the time advance value exceeding the threshold or a continuous attenuation of the reference signal received power, it is determined to be a damaged base station or an abnormal signal area. S32. Compensation positioning calculation: Extract the list of surviving base stations around the anomaly point, convert the physical distance through the time advance value, construct a multi-circle intersection positioning model, and then introduce the RSRP attenuation model to perform weighted correction on the coordinate cluster information; S33. Behavior Deviation Verification: Call the historical behavior analysis module to calculate the ratio of the target user's current dwell time to the historical average dwell time; when the deviation exceeds the set threshold, mark the coordinate cluster as a high-confidence trapped point.

6. The intelligent emergency response method for abnormal risks based on multi-source communication big data according to claim 1, characterized in that: In S4, the specific calculation process is as follows: S41. Constraint definition: including environmental constraints and communication guarantee constraints; Environmental constraints are introduced into the path search to remove road sections with real-time water depth greater than a threshold and to remove areas predicted for secondary disasters. Communication assurance constraints: If a path segment is covered by a surviving base station, delete risky paths that are not covered by surviving base stations; S42. Cost Function Reconstruction: Reconstructing A The cost function of the algorithm; S43. Objective function weight configuration: Set the path travel time weight and communication coverage intensity weight; S44. Output the optimal rescue route with communication support, including the number of road segments, estimated travel time, communication coverage, and avoidance of uncovered road segments.

7. The intelligent emergency response method for abnormal risks based on multi-source communication big data according to claim 6, characterized in that: In S42, reconstruct A Algorithm cost function for: ； Actual cost The newly added communication cost item is as follows: ； Where h(n) represents the estimated cost from the current node n to the target point; This represents the communication cost weighting coefficient; This represents the cumulative actual cost from the starting point to the parent node; The cost indicating whether the current road segment is within the communication coverage area; This represents the travel time cost of the current node n on the road segment; This represents the actual cost from the starting point to the current node n.

8. An intelligent emergency response system for abnormal risks based on multi-source communication big data, implementing the intelligent emergency response method for abnormal risks of multi-source communication big data as described in any one of claims 1 to 7, characterized in that: The system adopts a B / S architecture and includes: Data layer: Used to store base station topology data, RSRP / TA measured data, and road network constraint data; Core layer: includes risk factor configuration module, passive positioning calculation module, and path planning optimization module; Presentation layer: used for visual rendering of dynamic risk heat maps, distribution maps of damaged base stations, and generated communication-supported rescue routes.

Images