Digital relay telephone traffic real-time acquisition and intelligent distribution system for emergency dispatching

Abstract

The invention relates to the technical field of digital relays, and discloses a digital relay telephone traffic real-time acquisition and intelligent distribution system for emergency scheduling, which executes the following steps through five modules: receiving and normalizing UDP (User Datagram Protocol) data through a time domain buffer management module, completing space-time matching of telephone traffic and media streams through a cross-protocol association engine, and completing real-time transmission of the telephone traffic and the media streams. And unsuccessful data triggers a reverse tracing service for remedy, the intelligent scheduling controller performs grading processing according to the emergency degree and dynamically deploys resources, and finally the data is intelligently distributed to a message queue cluster through an elastic load balancing module. By constructing a multi-level heterogeneous data processing framework and dynamic redundancy transmission, intelligent space-time alignment and elastic resource scheduling mechanisms, accurate routing of first-aid instructions and zero-loss guarantee of key data are achieved, meanwhile, the self-adaptive fault-tolerant capability of the system to abnormal working conditions such as equipment clock skew and burst flow is enhanced, and the system reliability is improved. And high-availability technical support is provided for digital upgrading of emergency dispatching.

Description

Technical Field

[0001] The present invention relates to the field of digital relays, and more specifically, to a real-time collection and intelligent distribution system for digital relay traffic facing emergency dispatch. Background Art

[0002] With the rapid development of smart cities and emergency response systems, higher requirements are put forward for the real-time performance, reliability, and intelligent processing of digital relay traffic in emergency dispatch scenarios.

[0003] In the emergency dispatch network with mixed networking of multi-vendor devices and clock deviations, traditional systems have technical bottlenecks in aspects such as UDP broadcast packet loss, high-concurrency media stream distribution, multi-channel load balancing, and maintenance of traffic-media data correlation. Therefore, a real-time collection and intelligent distribution system for digital relay traffic is proposed to solve the above problems. Summary of the Invention

[0004] The present invention provides a real-time collection and intelligent distribution system for digital relay traffic facing emergency dispatch, which solves the technical problems of UDP broadcast packet loss, high-concurrency media stream distribution, multi-channel load balancing, and maintenance of traffic-media data correlation in the emergency dispatch network in related technologies.

[0005] The present invention provides a real-time collection and intelligent distribution system for digital relay traffic facing emergency dispatch, including: Time-domain buffer management module: By receiving and verifying the time deviation of UDP data packets, storing the advanced data in the future buffer, directly processing the immediate data, and performing historical compensation alignment on the lagged data, regularly clearing the expired cache and then merging and outputting the standardized data stream to the downstream processing module; Cross-protocol association engine module: By parsing and standardizing traffic and media data packets, constructing a spatio-temporal joint index, calculating the field similarity and establishing candidate matching pairs, dynamically verifying the time window deviation and then outputting the association result or triggering the traceback process; Reverse traceback service module: By receiving unassociated media packets and generating traceback query keys, performing multi-level retrieval in the real-time buffer and historical storage, after calculating the similarity and dynamically expanding the time window, returning the best matching result to the association module, or recording the characteristics of unmatched data; Intelligent scheduling controller module: By integrating the successfully associated and traced-back data streams, calculating the multi-dimensional emergency index and dividing into three-level processing queues, dynamically allocating resources and handling the risk of queue overflow, finally outputting the scheduling result to the elastic load balancing module and generating a system status report; Elastic Load Balancing Module: By collecting the status of cluster nodes in real time and quantifying the health, performing quantization-based resource allocation, detecting abnormal nodes to trigger fusing, dynamically migrating traffic and distributing data to MQ cluster nodes, and generating resource reports and feeding back the load status at the same time.

[0006] Further, the steps performed in the Time Domain Buffer Management Module are as follows: UDP Packet Reception and Timestamp Extraction: Receive the UDP broadcast packets sent by the voice collector, parse the timestamp field in the packet header, and obtain the gateway system time; Time Window Classification Decision: Classify the packet types according to the clock deviation and allocate them to the corresponding processing channels; Future Buffer Storage Management: Organize and store the early packets according to time slots, and maintain the circular buffer pointer; Historical Compensator Time Domain Alignment: Align the time scales of the late packets with the historical data, and use linear interpolation to compensate for the clock deviation; Regular Cleaning Service: Eliminate the expired data in each cleaning cycle to maintain the validity of the buffer space; Data Flow Transfer Control: Merge the processed data streams and output them to the Cross-Protocol Association Engine Module.

[0007] Further, the steps performed in the Cross-Protocol Association Engine Module are as follows: Dual-Protocol Data Input and Standardization: Receive the regularized data stream from the Time Domain Buffer Management Module, parse the traffic packets and media packets, and extract the standardized fields; Spatio-Temporal Index Construction: Build a joint index for the two types of data and generate a spatio-temporal hash key; Association Matrix Generation: Calculate the similarity of cross-protocol fields and establish a set of candidate matching pairs; Temporal Tolerance Verification: Verify whether the timestamp deviation is within the permitted range and dynamically adjust the time window; Association Decision and Output: Perform the final matching according to the similarity score, and output the association result or trigger a traceback.

[0008] Further, the steps performed in the Reverse Traceback Service Module are as follows: Trigger Condition and Input Preprocessing: Receive the unassociated media packets output by the Cross-Protocol Association Engine Module and extract the key traceback fields; Traceback Query Key Generation: Build a composite index key and define the traceable time range; Multi-Level Buffer Retrieval: First query the real-time buffer of the Time Domain Buffer Management Module, and extend the retrieval to the historical storage area; Cross-Protocol Similarity Calculation: Compare the field similarities of the candidate traffic packets and calculate the comprehensive matching score; Dynamic Window Expansion Decision: Judge whether the traceback limit is reached and automatically expand the spatio-temporal search range; Result feedback and recording: Return the best matching result to the cross-protocol association engine module, and record the characteristics of unmatched data.

[0009] Further, the steps executed in the intelligent scheduling controller module are as follows: Multi-source data integration input: Receive the successfully associated data stream and the traceability and remedial data stream, and merge the two into the input queue; Multi-dimensional urgency assessment: Calculate the composite urgency index of the data packet, and dynamically adjust the weight according to business rules; Three-level queue dynamic partitioning: Divide the processing queue according to the urgency index, and set the warning line for the queue capacity; Elastic resource allocation: Monitor the load status of each queue, and dynamically adjust the calculation resource allocation ratio; Exception handling and degradation: Detect the risk of queue overflow and trigger the resource borrowing mechanism; Output control and feedback: Push the scheduling result to the elastic load balancing module, and generate a system health status report.

[0010] Further, the steps executed in the elastic load balancing module are as follows: Multi-dimensional load status collection: Real-time collect the status indicators of the message queue cluster nodes, and receive the queue status feedback from the intelligent scheduling controller module; Node health quantification: Calculate the comprehensive load index of the node and divide the node health level; Quantized resource allocation: Abstract the node resources into quantum units and allocate processing quanta according to the health degree; Abnormal node fusing detection: Detect potential overloaded nodes and trigger the fusing protection mechanism; Dynamic traffic redistribution: Calculate the traffic migration volume between nodes and execute the smooth migration strategy; Data distribution and monitoring feedback: Distribute the data to the MQ cluster nodes according to the quantum allocation, generate a resource allocation report and feedback the status.

[0011] Further, in the steps of generating the association matrix, for any traffic packet i and media packet j, the calculation formula of the elements in the association matrix is as follows: ; Among them, the channel similarity: ; Among them, is the call identifier weight, is the channel number weight, is the exact matching function, is the channel similarity calculation function, is the call identifier of the i-th traffic packet, is the call identifier for the j-th media packet, represents the element in the i-th row and j-th column of the association matrix, indicating the similarity score between the i-th traffic packet and the j-th media packet, represents the timestamp of the i-th traffic packet, represents the timestamp of the j-th media packet, represents the channel number of the i-th traffic packet, represents the channel number of the j-th traffic packet, is the half-width of the instant window.

[0012] Furthermore, in the association decision and output, the decision rule formula is as follows: ; The output format is as follows: ; where, represents the element in the i-th row and j-th column of the association matrix, is the similarity threshold, is the standardized field set of the traffic packet, is the standardized field set of the media packet, is the output of the association result, is the empty set marker.

[0013] Furthermore, in the generation of the trace query key, the key generation formula is as follows: ; ; where, is the generated trace query key, is the number of times of time window extension, with an initial value of 0, is the maximum time deviation threshold, is the hash function, is the string concatenation operator, is the call identifier of the media packet, is the corrected timestamp of the media packet.

[0014] Furthermore, in the intelligent scheduling controller module, the scheduling result includes a priority marker, target node allocation, resource quota, timeliness constraint, and routing path.

[0015] The beneficial effects of the present invention are as follows: This technical solution constructs a multi-level heterogeneous data processing framework, significantly enhancing the real-time response ability and system reliability of digital trunk traffic in the first-aid dispatching scenario, effectively solving the core problems of traditional systems such as UDP data loss in complex network environments, congestion in high-concurrency media stream distribution, and cross-protocol timing misalignment. Through dynamic redundant transmission, intelligent spatio-temporal alignment, and elastic resource scheduling mechanisms, accurate routing of first-aid instructions and zero-loss guarantee of critical data are achieved. At the same time, the system's adaptive fault tolerance to abnormal working conditions such as device clock deviation and sudden traffic is enhanced, providing high-availability technical support for the digital upgrade of first-aid dispatching. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 FIG. is a structural block diagram of a real-time acquisition and intelligent distribution system for digital trunk traffic for first-aid dispatching proposed by the present invention; Figure 2 is of the present invention Figure 1 execution flowchart of Module 1; Figure 3 is of the present invention Figure 1 execution flowchart of Module 2; Figure 4 is of the present invention Figure 1 execution flowchart of Module 3; Figure 5 is of the present invention Figure 1 execution flowchart of Module 4; Figure 6 is of the present invention Figure 1 execution flowchart of Module 5. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0017] The subject matter described herein will now be discussed with reference to example embodiments. It should be understood that discussing these embodiments is only to enable those skilled in the art to better understand and thus implement the subject matter described herein. Without departing from the scope of protection of the content of this specification, changes can be made to the functions and arrangements of the elements discussed. Each example can omit, substitute, or add various processes or components as needed. Additionally, the features described relative to some examples can also be combined in other examples.

[0018] As Figures 1-6 shown, a real-time acquisition and intelligent distribution system for digital trunk traffic for first-aid dispatching includes: Module 1, Time Domain Buffer Management Module: By receiving and verifying the time deviation of UDP data packets, storing the early data in the future buffer, directly processing the immediate data, and performing historical compensation alignment on the late data. After regularly clearing the expired cache, it merges and outputs a standardized data stream to the downstream processing module; As Figure 2As shown in the figure, the following steps are performed by this module: 1.1, UDP Packet Reception and Timestamp Extraction: Receive the UDP broadcast message sent by the voice acquisition device, and parse the timestamp field in the message header ( ), and obtain the gateway system time ( ); Formula representation: ; Among them, is the device timestamp, is the gateway timestamp, is the timestamp difference, ; 1.2, Time Window Classification Decision: Classify the packet types according to the clock deviation and allocate them to the corresponding processing channels; Classification rules: ; Among them, is the leading determination threshold, , is the half-width of the instant window, , is the lag compensation threshold, ; 1.3, Future Buffer Storage Management: Organize and store the leading packets according to time slots, and maintain the circular buffer pointer; Storage strategy: ; Overwrite rule: ; Among them, is the storage slot index, is the time slot granularity, , is the maximum capacity of the circular buffer, , is the current pointer position, is the previous pointer position, is the leading determination threshold; 1.4, Historical Compensator Time Domain Alignment: Align the time scales of the lagging packets with the historical data, and use linear interpolation to compensate for the clock deviation; Compensation formula: ; Among them, α is the compensation coefficient, , is the corrected timestamp, is the device timestamp, is the system timestamp; 1.5, Regular cleaning service: Expired data is eliminated in each cleaning cycle to maintain the validity of the buffer space; Cleaning rules: ; Among them, represents the data validity period, is the buffer set, is the system timestamp, is the device timestamp of the cached data, is the cached data; 1.6, Data flow transfer control: The merged processed data stream is output to the next processing module; Output rules: ; Among them, is the final output data set, are various types of buffers, is the buffer type index, , = future, = real, = history, represents the future buffer, represents the real-time processing channel, represents the historical compensator; Module 2, Cross-protocol association engine module: By parsing and standardizing traffic and media data packets, constructing a spatio-temporal joint index, calculating field similarity and establishing candidate matching pairs, dynamically verifying the time window deviation and then outputting the association result or triggering the traceback process; As Figure 3 shown, the following steps are executed through this module: 2.1, Dual-protocol data input and standardization: Receive the regularized data stream from the time-domain buffer management module, parse the traffic packet ( ) and the media packet ( ), and extract the standardized fields; Traffic packet: ; Media packet: ; Among them, is the corrected timestamp, is the instant window threshold, is the traffic protocol data packet, is the media protocol data packet, is the traffic packet call identifier, is the traffic packet channel number, is the timestamp after traffic packet calibration, is the media packet call identifier, is the media packet channel number, is the timestamp after media packet calibration, is the original media packet data, is the media packet standardized field set, is the traffic packet standardized field set; 2.2, Spatiotemporal Index Construction: Construct a joint index for two types of data to generate a spatiotemporal hash key; Index Generation Formula: ; where, is the hash function, is the string concatenation operator, is the instant window threshold, is the call identifier, is the calibrated timestamp, is the generated hash key value; 2.3, Association Matrix Generation: Calculate the cross - protocol field similarity and establish a set of candidate matching pairs; For any traffic packet i and media packet j, the calculation formula for the elements in the association matrix: ; where the channel similarity: ; where, is the call identification weight, is the channel number weight, is the exact matching function, is the channel similarity calculation function, is the call identifier of the i - th traffic packet, is the call identifier of the j - th media packet, represents the element in the i - th row and j - th column of the association matrix, representing the similarity score between the i - th traffic packet and the j - th media packet, represents the timestamp of the i - th traffic packet, represents the timestamp of the j - th media packet, represents the channel number of the i - th traffic packet, represents the channel number of the j - th media packet; 2.4, Temporal Tolerance Verification: Verify whether the timestamp deviation is within the permitted range and dynamically adjust the time window; Verification Condition Formula: ; Window Adjustment Rule: ; Among them, is the maximum allowable time deviation, is the current processing queue depth, is the safety queue threshold, is the instant window threshold, is the timestamp after correction of the traffic packet, is the timestamp after correction of the media packet; 2.5, Association decision and output: Perform the final matching according to the similarity score, and output the association result or trigger the traceback; Decision rule formula: ; Output format: ; Among them, represents the element in the i-th row and j-th column of the association matrix, representing the similarity score between the i-th traffic packet and the j-th media packet, is the similarity threshold, , is the traffic packet standardized field set, is the media packet standardized field set, is the association result output, is the empty set marker; Module 3, Reverse Traceback Service Module: Receive the unassociated media packet and generate a traceback query key, perform multi-level retrieval in the real-time buffer and historical storage, and after similarity calculation and dynamic expansion of the time window, return the best matching result to the association module, or record the characteristics of the unmatched data; Execute the following steps through this module: 3.1, Trigger condition and input preprocessing: Receive the unassociated media packet output by the cross-protocol association engine module and extract the key traceback fields; Receive the unassociated media packet: ; Extract the key traceback fields: ; Among them, is the record of the unassociated successfully media packet, is the media packet standardized field set, is the empty set marker, is the set of key fields required for traceback, is the media packet call identifier, is the original data of the media packet, is the timestamp after correction of the media packet; 3.2, Traceback query key generation: Construct a composite index key and define the traceable time range; Key generation formula: ; ; wherein, is the generated trace query key, is the number of times the time window is extended, with an initial value of 0, is the maximum time deviation threshold, is the hash function, is the half-width of the instant window, is the string concatenation operator, is the media packet call identifier, is the timestamp of the media packet after correction; 3.3, Multi-level buffer retrieval: preferentially query the real-time buffer of the time domain buffer management module and expand the retrieval history storage area; Retrieval range definition: ; Priority rule: ; wherein, is the retrieval time range, is the timestamp of the media packet after correction, is the number of times the time window is extended, is the initial time deviation threshold, is the retrieval priority policy, is the time domain buffer, is the persistent history repository; 3.4, Cross-protocol similarity calculation: compare the field similarities of candidate traffic packets and calculate the comprehensive matching score; Scoring model: ; ; wherein, is the trace similarity score, is the time similarity weight, is the call identifier weight, is the timestamp difference, is the time decay coefficient, is the Levenshtein edit distance, is the media packet call identifier, is the traffic packet call identifier, is the maximum length of the call identifier, and its fixed value is 32; 3.5, Dynamic window extension decision: judge whether the trace limit is reached and automatically expand the spatio-temporal search range; Expansion strategy: ; Among them, is the updated expansion times, is the current expansion times, is the maximum expansion times, is the trace similarity score, is the trace matching threshold, is the current maximum similarity score; 3.6, Result feedback and recording: Return the best matching result to the cross - protocol association engine module, and record the characteristics of unmatched data; Feedback protocol: ; Among them, is the trace result feedback, is the media packet normalization field set, is the best - matched traffic packet normalization field set, is the trace similarity score, is the trace matching threshold, is the empty set marker; Module 4, Intelligent scheduling controller module: By integrating the successfully associated and traced - back remedial data streams, calculating the multi - dimensional emergency index and dividing into three - level processing queues, dynamically allocate resources, handle the risk of queue overflow, and finally output the scheduling result to the elastic load - balancing module and generate a system status report; Execute the following steps through this module: 4.1, Multi - source data integration input: Receive the successfully associated data stream and the traced - back remedial data stream, and merge them into the input queue; Receive the successfully associated data stream from the cross - protocol association engine module: ; Receive the traced - back remedial data stream from the reverse trace service module: ; Merge the input queue: ; Among them, is the similarity threshold, is the trace matching threshold, is the set of successfully associated data, is the set of traced - back remedial data, is the total merged data set, is the association result, is the trace result feedback, is the comprehensive similarity score, is the trace similarity score; 4.2, Multi-dimensional Urgency Assessment: Calculate the composite urgency index of the data packet and dynamically adjust the weight according to the business rules; Evaluation formula: ; ; where is the composite urgency index, is the timestamp difference, is the current system time, is the corrected timestamp, is the comprehensive similarity score, is the queue depth monitoring value, is the security queue depth threshold, is the timeliness weight, is the associated credibility weight, is the channel priority weight; 4.3, Three-level Queue Dynamic Partitioning: Partition the processing queue according to the urgency index and set the queue capacity warning line; Partitioning rule: ; where is the composite urgency index, is the urgent queue threshold, and the maximum capacity of the urgent queue is , is the lag queue threshold, represents the queue type; 4.4, Elastic Resource Allocation: Monitor the load status of each queue and dynamically adjust the calculation resource allocation ratio; Resource allocation formula: ; where is the resource allocation ratio of queue j, is the current backlog of queue j, is the current backlog of queue i, is the smoothing factor, is the average urgency index of queue j, is the average urgency index of queue i, is the queue index, where 1 = urgent level, 2 = normal level, 3 = lag level; 4.5, Abnormal Handling and Degradation: Detect the risk of queue overflow and trigger the resource borrowing mechanism; Overflow determination condition: ; Borrowing rule: ; wherein, is the current backlog of the emergency queue, is the current backlog of the normal queue, is the maximum capacity of the emergency queue, is the amount of resources that can be borrowed, is the lending ratio of the normal queue, is the maximum borrowing coefficient; 4.6, Output Control and Feedback: Push the scheduling result to the elastic load balancing module and generate a system health status report; Output Protocol: ; wherein, is the final scheduling output result, is the data set in queue j, is the resource allocation ratio of queue j, is the total computing resources of the cluster, is the minimum resource requirement for a single task, is the queue index; The scheduling result consists of the following parts: Priority Mark: Emergency / Normal / Delay three-level processing identifier; Target Node Allocation: The list of message queue (MQ) cluster nodes specified for each data packet; Resource Quota: The computing resource allocation plan such as the number of CPU cores and the size of memory; Timeliness Constraint: The maximum allowable processing delay threshold (e.g., ≤50ms); Routing Path: The transmission link planning for cross-protocol associated data; Module 5, Elastic Load Balancing Module: Dynamically migrate traffic and distribute data to the message queue (MQ) cluster nodes by collecting the status of cluster nodes in real time, quantifying the health degree, performing quantization resource allocation, detecting abnormal nodes to trigger fusing, and generating a resource report and feedback the load status at the same time; Execute the following steps through this module: 5.1, Multi-dimensional Load Status Collection: Collect the status metrics of message queue (MQ) cluster nodes in real time and receive the queue status feedback from the intelligent scheduling controller module; Collection Parameters: ; wherein, is the CPU usage rate of node i, is the CPU resources already used by node i, is the total CPU resources of node i, is the memory usage rate of node i, is the memory resource used by node i, is the total memory resource of node i, is the queue utilization rate of node i, is the real-time depth of the three-level queue, is the maximum capacity of the emergency queue, is the node index number; 5.2, Node health quantification: Calculate the comprehensive load index of the node and divide the node health level; Health calculation: ; Weight adjustment: ; Among them, is the health index of node i, is the CPU weight reference value, is the memory weight reference value, is the queue weight (dynamically adjusted), is the CPU utilization rate of node i, is the memory utilization rate of node i, is the queue utilization rate of node i, is the current backlog of the emergency queue, is the maximum capacity of the emergency queue; 5.3, Quantized resource allocation: Abstract the node resources into quantum units and allocate processing quanta according to the health level; Allocation calculation: ; ; Among them, is the number of quanta allocated to node i, is the health index of node i, is the health index of node j, is the total number of quanta to be allocated, is the total number of cluster nodes, is the node index for summation, is the merged total data set; 5.4, Detection of abnormal node fusing: Detect potential overloaded nodes and trigger the fusing protection mechanism; Fusing determination: ; Among them, is the node index number, is the CPU utilization rate of node i, is the memory utilization rate of node i, is the queue utilization rate of node i, is the CPU fusing threshold, is the memory fusing threshold, is the queue fusing threshold; 5.5, Dynamic traffic redistribution: Calculate the traffic migration volume between computing nodes and execute the smooth migration strategy; Migration calculation: ; ; Among them, is the quantum number migrated from node i to node j, is the current quantum number of node i, is the current quantum number of node j, is the ideal allocation amount, is the health index of node i, is the sum of the health of all nodes, is the total quantum number, is the migration damping coefficient; 5.6, Data distribution and monitoring feedback: Distribute data to MQ cluster nodes according to quantum allocation, generate a resource allocation report and feedback the status; Distribution strategy: ; Feedback format: ; Among them, is the feedback result set, is the node identifier, is the health index of node i, is the quantum number allocated to node i, is the total number of cluster nodes, is the node index number, is the data set distributed to node i, is the consistent hashing function, is the total number of cluster nodes.

[0019] Based on the above system and its architecture, the following is an example - the emergency response of the urban first aid dispatch system.

[0020] Scenario background: A multi-vehicle chain-reaction accident occurs on the road. Eyewitnesses at the scene and in-vehicle emergency communications use devices of different operators to call the 110 police emergency number and the 120 emergency medical number. The ambulance dispatch systems of surrounding hospitals receive multiple emergency requests simultaneously; System operation process: Data reception and buffering: Multiple voice collection devices connected to the base stations near the accident scene start recording emergency calls and continuously send traffic and on-site audio and video data through UDP broadcast; The time-domain buffer management module detects that due to a clock failure in a certain device, the media packet timestamp shows 300 milliseconds earlier than the gateway system time; The system automatically stores the "ahead" data in the future buffer and waits to be released when the real-time processing window arrives; Cross-protocol association processing: The incoming traffic packets of some calling numbers arrive late due to network jitter; The association engine discovers that the corresponding on-site video packets have arrived in advance and immediately starts the reverse tracing service; By comparing features such as the incoming number and the geographical information of the accident section, the delayed traffic packets are successfully associated with the earlier-arrived video stream; Intelligent priority determination: The system recognizes that multiple incoming calls describe the same accident and automatically merges them into the same emergency event; According to keywords such as "multiple injured" and "vehicle on fire" in the call content, the event is upgraded to the red emergency level; Bandwidth resources are preferentially allocated to transmit the on-site fire video to the fire command center; Dynamic load scheduling: Emergency dispatch terminals of multiple nearby hospitals simultaneously initiate resource requests, and the load balancing module detects that the CPU usage rate of the second hospital node soars; Some video streams are automatically switched to the standby cloud node for processing to ensure the uninterrupted transmission of emergency instructions; According to the number of idle ambulances in each hospital, the task of transporting the wounded is intelligently allocated; Closed-loop emergency response: A complete on-site assessment report (including voice recording, wounded location, and preliminary injury judgment) is generated within 2 minutes after the accident; The system automatically pushes the resource demand prediction to the Municipal Emergency Management Bureau, triggering the reserve support mechanism; After the emergency is over, all relevant data packets are automatically archived to form a complete emergency event disposal knowledge base.

[0021] The above describes the embodiments of the present invention, but the present invention is not limited to the above specific implementation manners. The above specific implementation manners are merely illustrative and not restrictive. Under the inspiration of the present invention, those of ordinary skill in the art can also make many forms, all of which fall within the protection scope of the present invention.

Claims

1. A real-time collection and intelligent distribution system for digital trunk traffic facing emergency dispatch, characterized in that Including: Time-domain buffer management module: By receiving and verifying the time deviation of UDP data packets, storing the early data in the future buffer, directly processing the immediate data, and performing historical compensation alignment on the late data, regularly cleaning the expired cache and then merging and outputting the standardized data stream to the downstream processing module; Cross-protocol association engine module: By parsing and standardizing the traffic and media data packets, constructing a spatio-temporal joint index, calculating the field similarity and establishing candidate matching pairs, dynamically verifying the time window deviation and then outputting the association result or triggering the traceback process; Reverse traceback service module: By receiving the unassociated media packets and generating traceback query keys, performing multi-level retrieval in the real-time buffer and historical storage, after calculating the similarity and dynamically expanding the time window, returning the best matching result to the association module, or recording the characteristics of the unmatched data; Intelligent scheduling controller module: By integrating the successfully associated and traceback-remedied data streams, calculating the multi-dimensional emergency index and dividing into three-level processing queues, dynamically allocating resources and handling the risk of queue overflow, finally outputting the scheduling result to the elastic load balancing module and generating a system status report; Elastic load balancing module: By real-time collecting the status of cluster nodes and quantifying the health degree, performing quantization resource allocation, detecting abnormal nodes and triggering fusing, dynamically migrating traffic and distributing data to the MQ cluster nodes, and at the same time generating a resource report and feedback the load status.

2. The real-time acquisition and intelligent distribution system of digital trunk traffic for first aid dispatching according to claim 1, wherein The steps executed in the time-domain buffer management module are as follows: UDP data packet reception and timestamp extraction: Receive the UDP broadcast message sent by the voice collector, parse the timestamp field in the message header, and obtain the gateway system time; Time window classification decision: Divide the data packet types according to the clock deviation and allocate them to the corresponding processing channels; Future buffer storage management: Organize and store the early packets according to time slots, and maintain the circular buffer pointer; Historical compensator time-domain alignment: Align the time scales of the late packets and historical data, and use linear interpolation to compensate for the clock deviation; Regular cleaning service: Eliminate the expired data every cleaning cycle to maintain the validity of the buffer space; Data flow transfer control: Merge the processed data streams and output them to the cross-protocol association engine module.

3. A real-time acquisition and intelligent distribution system for digital trunk traffic facing emergency dispatch according to claim 2, characterized in that, The steps executed in the cross-protocol association engine module are as follows: Dual-protocol data input and standardization: Receive the regularized data stream from the time-domain buffer management module, parse the traffic packets and media packets, and extract the standardized fields; Spatio-temporal index construction: Construct a joint index for the two types of data and generate a spatio-temporal hash key; Association matrix generation: Calculate the cross-protocol field similarity and establish a set of candidate matching pairs; Timing tolerance verification: Verify whether the timestamp deviation is within the permitted range and dynamically adjust the time window; Association decision and output: Perform the final matching according to the similarity score and output the association result or trigger the traceback.

4. A real-time acquisition and intelligent distribution system for digital trunk traffic facing emergency dispatch according to claim 3, wherein The steps executed in the reverse traceback service module are as follows: Trigger condition and input preprocessing: Receive the unassociated media packets output by the cross-protocol association engine module and extract the key traceback fields; Traceback query key generation: Construct a composite index key and define the traceable time range; Multi-level buffer retrieval: First query the real-time buffer of the time-domain buffer management module and expand the retrieval to the historical storage area; Cross - protocol similarity calculation: Compare the field similarities of candidate traffic packets and calculate the comprehensive matching score; Dynamic window expansion decision: Determine whether the tracing limit is reached and automatically expand the spatio - temporal search range; Result feedback and recording: Return the best matching result to the cross - protocol association engine module and record the characteristics of unmatched data.

5. The real-time acquisition and intelligent distribution system of digital trunk traffic for first-aid dispatching according to claim 4, characterized in that The steps executed in the intelligent scheduling controller module are as follows: Multi - source data integration input: Receive the successfully associated data stream and the tracing and remediation data stream, and merge the two into the input queue; Multi - dimensional urgency assessment: Calculate the composite urgency index of data packets and dynamically adjust the weights according to business rules; Three - level queue dynamic partitioning: Partition the processing queues according to the urgency index and set the warning line for queue capacity; Elastic resource allocation: Monitor the load status of each queue and dynamically adjust the calculation resource allocation ratio; Exception handling and degradation: Detect the risk of queue overflow and trigger the resource borrowing mechanism; Output control and feedback: Push the scheduling result to the elastic load balancing module and generate a system health status report.

6. The real-time acquisition and intelligent distribution system of digital trunk traffic for first aid dispatching according to claim 5, characterized in that, The steps executed in the elastic load balancing module are as follows: Multi - dimensional load status collection: Real - time collect the status indicators of message queue cluster nodes and receive the queue status feedback from the intelligent scheduling controller module; Node health quantification: Calculate the comprehensive load index of nodes and divide the node health levels; Quantized resource allocation: Abstract the node resources into quantum units and allocate processing quanta according to the health level; Abnormal node fusing detection: Detect potential overloaded nodes and trigger the fusing protection mechanism; Dynamic traffic redistribution: Calculate the traffic migration volume between nodes and execute the smooth migration strategy; Data distribution and monitoring feedback: Distribute data to MQ cluster nodes according to quantum allocation, generate a resource allocation report and feedback the status.

7. A real-time acquisition and intelligent distribution system for digital trunk traffic facing emergency dispatch according to claim 6, characterized in that In the steps of generating the association matrix, for any traffic packet i and media packet j, the calculation formula of the elements in the association matrix is as follows: ； Where Channel similarity: ； Among them, is the call identification weight, is the channel number weight, is the exact matching function, is the channel similarity calculation function, is the call identifier of the i-th traffic packet, is the call identifier of the j-th media packet, represents the element at the i-th row and j-th column of the association matrix, representing the similarity score between the i-th traffic packet and the j-th media packet, represents the timestamp of the i-th traffic packet, represents the timestamp of the j-th media packet, represents the channel number of the i-th traffic packet, represents the channel number of the j-th traffic packet, is the half-width of the instant window.

8. A real-time acquisition and intelligent distribution system for digital trunk traffic facing emergency dispatch according to claim 7, characterized in that In the association decision and output, the decision rule formula is as follows: ； The output format is as follows: ； Among them, represents the element in the i-th row and j-th column of the association matrix, is the similarity threshold, is the standardized field set of the traffic packet, is the standardized field set of the media packet, is the output of the association result, is the empty set marker.

9. The real-time acquisition and intelligent distribution system of digital trunk traffic for first-aid dispatch according to claim 8, characterized in that In the generation of the tracing query key, the key generation formula is as follows: ； ； Among them, is the generated trace query key, is the number of times the time window is extended, with an initial value of 0, is the maximum time deviation threshold, is the hash function, is the string concatenation operator, is the media packet call identifier, is the corrected timestamp of the media packet.

10. A real-time acquisition and intelligent distribution system for digital trunk traffic facing emergency dispatch according to claim 9, characterized in that, In the intelligent scheduling controller module, the scheduling result includes priority marking, target node allocation, resource quota, timeliness constraint and routing path.

Images