A method and system for efficient detection of intelligent SMS links

Abstract

This application provides an intelligent method and system for efficient detection of SMS links. The method involves deploying a swarm of drones to form a dynamic communication topology, used to monitor the latency mutation rate and packet loss rate of damaged SMS links in real time, generating a state fluctuation sequence. When a transmission latency gradient exceeds an emergency communication threshold, an anomaly marking mechanism is triggered, and these data are analyzed to identify anomaly clusters and their spatiotemporal coupling relationships. This information is then bound to timestamps, and a blockchain state change event is generated using multi-node parallel notarization technology. Based on the verification results, the drone distribution density is adjusted to generate a topology reconstruction factor, thereby updating the SMS transmission path and optimizing the communication network. The technical solution provided in this application can improve the stability and reliability of efficient SMS link detection.

Description

Technical Field

[0001] This application relates to the field of efficient SMS link detection technology, and in particular to an intelligent efficient SMS link detection method and system. Background Technology

[0002] During natural disasters or emergencies, communication infrastructure is often severely damaged, leading to interruptions or sharp performance degradation of traditional communication methods such as SMS services. In such situations, ensuring the reliability, timeliness, and stability of emergency communications becomes a critical technological requirement. Especially in rescue operations, real-time access to on-site information is crucial for command and dispatch. Therefore, a communication solution capable of rapid response and adaptive adjustment is needed, dynamically optimizing SMS transmission paths in complex environments to minimize latency and data loss rates, thus ensuring the effectiveness of emergency communications.

[0003] Currently, an advanced solution utilizes satellite communication networks as a backup communication method. When damage to terrestrial communication facilities is detected, data transmission is immediately switched to satellite links. This solution deploys a series of low-Earth orbit satellites globally, forming a wide-coverage communication network unaffected by ground-based disasters. This effectively provides stable SMS services, ensuring uninterrupted communication even under extreme conditions. Furthermore, satellite communication supports highly automated management and configuration, reducing the need for manual intervention and improving the overall system's response speed and efficiency.

[0004] While satellite communication systems offer an effective backup option for emergency response, they also have some significant limitations. First, satellite communication is relatively expensive, not only in terms of initial investment but also in daily operation and maintenance costs and usage fees, which can be a considerable burden for resource-constrained emergency response agencies. Second, because satellite communication relies on specific spatial locations and orbital parameters, it may not provide optimal service quality in certain geographical areas or during specific time windows, especially during periods of high-density user access, where bandwidth bottlenecks may occur, affecting communication efficiency. Finally, satellite communication systems lack flexibility and struggle to quickly adjust resource allocation according to real-time changing environmental conditions, which to some extent limits their ability to respond to emergencies. Summary of the Invention

[0005] This application provides an intelligent and efficient method and system for detecting SMS links, which solves the problems of poor stability and reliability in the prior art for detecting SMS links.

[0006] In a first aspect, embodiments of this application provide an intelligent and efficient method for detecting SMS links, including:

[0007] By deploying a drone swarm to form a dynamic communication topology, the latency mutation rate and packet loss rate data of the damaged SMS link are collected in real time, and an SMS link status fluctuation sequence is generated. When the SMS transmission latency gradient in the SMS link status fluctuation sequence exceeds the emergency communication threshold, an SMS link anomaly marker is triggered.

[0008] Based on the SMS link state fluctuation sequence, SMS link state migration detection is performed on the dynamic communication topology to identify abnormal association clusters of the transmission path and form a spatiotemporal coupling relationship with the SMS link anomaly marker.

[0009] The abnormal association cluster and SMS link state migration feature are bound to timestamps for multi-node parallel storage, generating a blockchain state change event containing the SMS link state fluctuation sequence.

[0010] The distribution density of drones is adjusted based on the verification results of the blockchain state change event, a text message link topology reconstruction factor is generated, and a closed-loop feedback is formed with the spatiotemporal coupling relationship to update the topology association detection path of text message transmission.

[0011] Optionally, the step of performing SMS link state transition detection on the dynamic communication topology based on the SMS link state fluctuation sequence, identifying abnormal association clusters of the transmission path, and forming a spatiotemporal coupling relationship with the SMS link anomaly marker includes:

[0012] The node aggregation path is generated based on the SMS link state fluctuation sequence. The aggregation weight is dynamically adjusted according to the delay gradient and packet loss rate of adjacent nodes, and a positive feedback association is formed with the triggering conditions of the SMS link anomaly marker.

[0013] SMS link state migration detection is performed on the dynamic communication topology, and abnormal association clusters of transmission paths are identified in the distribution characteristics of the node aggregation path and the SMS link state fluctuation sequence.

[0014] The spatiotemporal coupling relationship of the abnormal association clusters is quantified into spatiotemporal coupling coefficients, and the spatiotemporal coupling coefficients form a cross-validation relationship with the aggregation weights of the node aggregation paths;

[0015] Based on the cross-validation relationship, the spatial distribution boundary of the abnormal association cluster is corrected, abnormal association cluster correction parameters are generated, and the dynamic spatial association analysis process of the SMS link state fluctuation sequence is fed back through the node aggregation path.

[0016] Optionally, quantifying the spatiotemporal coupling relationship of the abnormal association cluster into a spatiotemporal coupling coefficient, wherein the spatiotemporal coupling coefficient forms a cross-validation relationship with the aggregation weight of the node aggregation path, includes:

[0017] Based on the three-dimensional coordinate distribution of UAV nodes in the emergency area, a coverage density index is generated by dividing the area into geographic grid units and calculating the ratio of the number of active nodes to the grid volume.

[0018] The trigger frequency distribution of SMS link anomaly markers is statistically analyzed according to a preset time window, an anomaly frequency histogram is generated, and the duration of the peak interval in the anomaly frequency histogram is extracted as the anomaly time clustering factor.

[0019] The coverage density index and the abnormal time clustering factor are weighted and fused to generate an initial spatiotemporal coupling coefficient. At the same time, the weight of the coverage density index is dynamically adjusted according to the priority of the emergency area.

[0020] Extract the weight distribution entropy value of the central region of the abnormal association cluster, and quantify the spatiotemporal coupling relationship into a spatiotemporal coupling coefficient based on the difference between the weight distribution entropy value and the initial spatiotemporal coupling coefficient.

[0021] An aggregated weight-spatiotemporal coupling cross-validation matrix is ​​constructed. A validation factor is generated by comparing the correlation between the gradient change rate of the aggregated weight and the update magnitude of the spatiotemporal coupling coefficient element by element. When the validation factor is lower than the confidence threshold, the spatial distribution boundary reconstruction of the abnormal association cluster is triggered.

[0022] Optionally, triggering the SMS link anomaly flag when the SMS transmission delay gradient in the SMS link state fluctuation sequence exceeds the emergency communication threshold includes:

[0023] Based on the difference between adjacent time windows in the SMS link state fluctuation sequence, the SMS transmission delay gradient is analyzed, and the normalized SMS transmission delay gradient is generated by normalizing the cumulative mean of the delay change rate within the sliding window.

[0024] Based on the historical distribution characteristics of the normalized SMS transmission delay gradient and the gradient change trend within the current window, the emergency communication threshold is dynamically adjusted.

[0025] When the normalized SMS transmission delay gradient exceeds the adjusted emergency communication threshold, an anomaly marker trigger condition is generated by combining the duration of the time exceeding the adjusted emergency communication threshold with the number of times it exceeds the threshold.

[0026] The parameters corresponding to the normalized SMS transmission delay gradient, the adjusted emergency communication threshold, and the anomaly marker triggering condition are used as associated parameters and passed to the subsequent SMS link anomaly analysis module to trigger the SMS link anomaly marker.

[0027] Optionally, the step of deploying a drone swarm to form a dynamic communication topology to collect real-time data on the latency mutation rate and packet loss rate of damaged SMS links, and generating an SMS link state fluctuation sequence, includes:

[0028] Based on the spatial distribution of the damaged SMS links and the current channel quality, the signal strength and interference level of each node in the UAV swarm are controlled to generate dynamic weight parameters, and a dynamic communication topology is formed through iterative updates of the UAV swarm position.

[0029] In the dynamic communication topology, the latency mutation rate and packet loss rate data of the damaged SMS link are collected at a preset period, and the packet loss rate data is supplemented with the packet loss rate variance within the collection time window as a fluctuation feature.

[0030] The latency mutation rate and packet loss rate data are timestamped and stitched together using a sliding window to generate a time-aligned collection dataset. The collection dataset is then weighted and fused based on the dynamic weight parameters to generate a SMS link status fluctuation sequence.

[0031] Optionally, the step of timestamping and concatenating the latency mutation rate and packet loss rate data using a sliding window to generate a time-aligned collection dataset includes:

[0032] A reference time axis is determined based on the clock synchronization protocol of each node in the drone cluster. The timestamps of the collected latency mutation rate and packet loss rate data are mapped to the reference time axis to generate a timestamp mapping table.

[0033] Based on the timestamp mapping table, the latency mutation rate and packet loss rate data with missing timestamps are filled by linear interpolation. The mean of adjacent valid data points is used as the interpolation reference value to generate the interpolated latency mutation rate and packet loss rate data.

[0034] The interpolated latency mutation rate and packet loss rate data are divided into multiple overlapping sliding windows. First and last timestamp calibration and boundary data smoothing are performed to generate latency mutation rate segments and packet loss rate segments aligned within the windows.

[0035] The latency mutation rate segment and the packet loss rate segment are concatenated in chronological order to form a unified latency and packet loss data block with a unified time dimension. The time coverage and data density are extracted as window metadata to generate a time-aligned collection dataset.

[0036] Optionally, the step of binding the abnormal association cluster and SMS link state transition features to timestamps for multi-node parallel notarization, and generating a blockchain state change event containing the SMS link state fluctuation sequence, includes:

[0037] The abnormal feature vector of the abnormal association cluster and the dynamic feature vector of the SMS link state transition feature are timestamped through a time window sliding alignment mechanism to generate an abnormal state joint feature vector group.

[0038] The abnormal state joint feature vector group is divided into multiple parallel evidence storage sub-units according to the timestamp label, and hash evidence storage is performed to generate an evidence storage hash chain, cross-node cross-verification is performed, and an evidence storage topology structure is constructed.

[0039] In the evidence storage topology, extract the state encoding value of the SMS link state transition feature corresponding to each timestamp tag, and perform fluctuation fitting on the state encoding value according to the state transition relationship in ascending order of timestamp to generate SMS link state fluctuation sequence.

[0040] The SMS link state fluctuation sequence is encapsulated and bound to the evidence storage hash chain to generate a blockchain state change event containing the SMS link state fluctuation sequence.

[0041] Secondly, embodiments of this application provide an intelligent SMS link high-efficiency detection system, including:

[0042] The data acquisition module forms a dynamic communication topology by deploying a cluster of drones to collect real-time data on the latency mutation rate and packet loss rate of damaged SMS links, and generates an SMS link status fluctuation sequence. When the SMS transmission latency gradient in the SMS link status fluctuation sequence exceeds the emergency communication threshold, an SMS link anomaly marker is triggered.

[0043] The identification module performs SMS link state migration detection on the dynamic communication topology based on the SMS link state fluctuation sequence, identifies abnormal association clusters of the transmission path, and forms a spatiotemporal coupling relationship with the SMS link anomaly marker.

[0044] The generation module binds the abnormal association cluster and SMS link state migration feature to timestamps for multi-node parallel storage, generating a blockchain state change event containing the SMS link state fluctuation sequence.

[0045] The update module adjusts the distribution density of drones based on the verification results of the blockchain state change event, generates a text message link topology reconstruction factor, and forms a closed-loop feedback with the spatiotemporal coupling relationship to update the topology association detection path of text message transmission.

[0046] Thirdly, embodiments of this application provide a computing device, including a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are invoked and executed by the processing component to realize an intelligent SMS link efficient detection method as described in the first aspect above.

[0047] Fourthly, embodiments of this application provide a computer storage medium storing a computer program, which, when executed by a computer, implements an intelligent SMS link efficient detection method as described in the first aspect.

[0048] In this embodiment, a dynamic communication topology is formed by deploying a drone swarm to collect real-time data on the latency mutation rate and packet loss rate of damaged SMS links, generating an SMS link state fluctuation sequence. When the SMS transmission latency gradient in the SMS link state fluctuation sequence exceeds the emergency communication threshold, an SMS link anomaly marker is triggered. Based on the SMS link state fluctuation sequence, SMS link state migration detection is performed on the dynamic communication topology to identify abnormal association clusters in the transmission path and form a spatiotemporal coupling relationship with the SMS link anomaly marker. The abnormal association clusters and SMS link state migration features are bound to timestamps for multi-node parallel storage, generating a blockchain state change event containing the SMS link state fluctuation sequence. The distribution density of the drones is adjusted according to the verification result of the blockchain state change event to generate an SMS link topology reconstruction factor, which forms a closed-loop feedback with the spatiotemporal coupling relationship to update the topology association detection path of SMS transmission.

[0049] The technical solution of this application has the following beneficial effects:

[0050] This application utilizes a drone swarm deployment to flexibly and quickly establish a temporary communication network, particularly suitable for emergency scenarios. This step enables real-time monitoring of SMS link status changes, allowing for timely detection of potential problems. The SMS link anomaly marking mechanism ensures the system can react rapidly and mark anomalies when SMS link performance deteriorates sharply, providing accurate data for subsequent processing. In-depth analysis of the dynamic communication topology accurately identifies key factors affecting communication quality and their interrelationships, enabling precise problem localization and description. Using blockchain technology to record anomaly information not only improves data security and immutability but also facilitates multi-party verification and traceability, enhancing system transparency and trustworthiness. Based on the verification results, drone deployment is optimized to further improve SMS link quality, achieving a highly efficient communication network that learns and adapts quickly.

[0051] Furthermore, node aggregation paths are generated based on SMS link state fluctuation sequences. Aggregation weights are dynamically adjusted and positive feedback associations are formed with anomaly markers. State transition detection is performed to identify anomalous association clusters and quantify their spatiotemporal coupling relationships. Cross-validation is used to correct the spatial distribution boundaries of these anomalous clusters, which are then fed back into the dynamic spatial association analysis process. This process enables fine-grained control and optimization of SMS link states, effectively improving the accuracy and response speed of anomaly detection, while also enhancing the stability and adaptability of the communication network.

[0052] These or other aspects of this application will become more apparent from the description of the following embodiments. Attached Figure Description

[0053] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0054] Figure 1 A flowchart of an intelligent SMS link high-efficiency detection method provided in this application is shown;

[0055] Figure 2 This paper presents a schematic diagram of the structure of an intelligent SMS link high-efficiency detection system provided in this application;

[0056] Figure 3 A schematic diagram of the structure of a computing device provided in this application is shown. Detailed Implementation

[0057] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.

[0058] In some of the processes described in the specification, claims, and accompanying drawings of this application, multiple operations appearing in a specific order are included. However, it should be clearly understood that these operations may not be executed in the order they appear herein, or may be executed in parallel. The operation numbers, such as 101, 102, etc., are merely used to distinguish different operations and do not themselves represent any execution order. Furthermore, these processes may include more or fewer operations, and these operations may be executed sequentially or in parallel. It should be noted that the descriptions such as "first," "second," etc., in this document are used to distinguish different messages, devices, modules, etc., and do not represent a chronological order, nor do they limit "first" and "second" to different types.

[0059] This application aims to construct a dynamic communication network using drone swarms to monitor the latency mutation rate and packet loss rate of SMS links in real time, generate state fluctuation sequences, and mark anomalies. Blockchain events are generated through multi-node parallel storage to ensure data security and traceability. Based on the verification results, the drone distribution density is adjusted to optimize the SMS link, forming a closed-loop feedback mechanism from detection to repair. This enables rapid response to environmental changes and self-optimization, improving the reliability and efficiency of SMS services in emergency communications and providing a flexible and reliable solution.

[0060] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0061] Figure 1 This application provides a flowchart of an intelligent SMS link high-efficiency detection method, as shown in the embodiments. Figure 1 As shown, the method includes:

[0062] 101. By deploying a drone swarm to form a dynamic communication topology, the latency mutation rate and packet loss rate data of the damaged SMS link are collected in real time to generate an SMS link status fluctuation sequence. When the SMS transmission latency gradient in the SMS link status fluctuation sequence exceeds the emergency communication threshold, an SMS link anomaly marker is triggered.

[0063] In this step, the dynamic communication topology is a temporary communication network structure formed by a swarm of drones to support emergency communication.

[0064] The latency mutation rate refers to the rate of latency change of a damaged SMS link within a specific time window, and is used to describe the latency changes during SMS transmission.

[0065] The packet loss rate data is supplemented with the variance of the packet loss rate within the collection time window as a fluctuation feature, which is used to describe the proportion of data packets lost during data transmission and their fluctuation.

[0066] The SMS link status fluctuation sequence is a time series generated based on latency mutation rate and packet loss rate data, used to reflect the status change trend of the SMS link.

[0067] The emergency communication threshold is a pre-set standard. When the SMS transmission delay gradient exceeds this standard, an alarm mechanism is triggered.

[0068] SMS link anomaly markers are generated when the SMS transmission delay gradient exceeds the emergency communication threshold, and are used to promptly detect and handle potential problems.

[0069] In this embodiment, firstly, a dynamic communication topology is formed by deploying a drone swarm to construct a temporary emergency communication network structure. Real-time data on the latency mutation rate and packet loss rate of damaged SMS links within a specific time window are collected, and the variance of the packet loss rate within the collection time window is added as a fluctuation characteristic. Secondly, an SMS link state fluctuation sequence is generated based on the collected data. This sequence reflects the trend of SMS link state changes, including latency changes and the proportion and fluctuation of lost data packets during data transmission. Finally, when the analysis of the SMS link state fluctuation sequence reveals that the SMS transmission latency gradient exceeds a pre-set emergency communication threshold, the system triggers an SMS link anomaly marker to promptly identify and address potential problems, ensuring the effectiveness and reliability of emergency communication. This process effectively utilizes the flexibility and real-time capabilities of the drone swarm, providing strong support for communication assurance in emergency situations.

[0070] In a rescue scenario following a sudden earthquake, drones were rapidly deployed over the disaster area, establishing a temporary communication network. The real-time data acquisition module began collecting various performance metrics of the SMS links in the disaster area. After a period of data accumulation, the system identified a significant increase in latency mutation rates in certain link segments, triggering anomaly markers and providing a clear direction for subsequent repair work.

[0071] 102. Based on the SMS link state fluctuation sequence, perform SMS link state migration detection on the dynamic communication topology, identify abnormal association clusters of the transmission path, and form a spatiotemporal coupling relationship with the SMS link anomaly marker;

[0072] In this step, SMS link state transition detection is an algorithmic tool used to monitor changes in the SMS link state.

[0073] Anomaly clusters refer to key factors affecting communication quality, which are identified by clustering analysis techniques from the distribution characteristics of node aggregation paths and SMS link state fluctuation sequences.

[0074] Spatiotemporal coupling is a numerical value that quantifies the temporal and spatial distribution characteristics of anomalous clusters and is used to assess their impact on the overall network.

[0075] By combining SMS link anomaly markers with spatiotemporal coupling, a closed-loop feedback mechanism is formed to achieve more accurate problem localization and optimization measures.

[0076] In this embodiment, firstly, based on the SMS link state fluctuation sequence, real-time monitoring is performed on each node in the dynamic communication topology to collect its state information. Next, time series analysis techniques are applied to identify the changing trends of the SMS link state, and clustering algorithms are used to identify abnormal association clusters. Then, the identified abnormal association clusters are compared and analyzed with existing SMS link anomaly markers to determine the spatiotemporal coupling relationship between the two. Finally, all information is integrated to generate a comprehensive report containing the spatiotemporal coupling relationship, which guides subsequent optimization measures.

[0077] Continuing with the earthquake rescue scenario described above, after identifying anomalous link segments, the system further analyzed the specific circumstances of these segments, identifying several key anomalous clusters. By quantifying their spatiotemporal coupling relationships, the system clarified the degree of impact of these clusters on overall communication efficiency, thus providing a basis for optimizing resource allocation.

[0078] 103. Bind the abnormal association cluster and SMS link state migration feature to timestamps for multi-node parallel storage and generate a blockchain state change event containing the SMS link state fluctuation sequence.

[0079] In this step, binding timestamps refers to the process of synchronizing the time information of abnormal association clusters and SMS link state migration characteristics.

[0080] Multi-node parallel storage is a technical means of storing data on multiple nodes simultaneously to ensure data reliability and tamper resistance.

[0081] A blockchain state change event is an immutable record containing all relevant data and verification information for long-term storage and verification. This event encapsulates the sequence of SMS link state fluctuations, providing a complete data chain.

[0082] In this embodiment, firstly, the abnormal association clusters and SMS link state transition features are bound to their corresponding timestamps to form a complete data packet. Next, hash-based evidence storage operations are performed in parallel on multiple nodes to ensure the authenticity and integrity of the data. Then, blockchain technology is used to encapsulate this data, creating a new blockchain state change event. Through a consensus mechanism, this event is recognized and recorded by all nodes in the network. Finally, this event is added to the blockchain to form a permanent record, ensuring the immutability of the data.

[0083] During earthquake relief efforts, all identified anomalies were simultaneously recorded by multiple drone nodes, creating a series of blockchain events. This not only provides robust data support for current rescue operations but also accumulates valuable experience for future disaster response.

[0084] 104. Adjust the distribution density of drones based on the verification results of the blockchain state change event, generate SMS link topology reconstruction factor, and form a closed-loop feedback with the spatiotemporal coupling relationship to update the topology association detection path of SMS transmission.

[0085] In this step, the verification result of the blockchain state change event is confirmed by comparing the data consistency across different nodes.

[0086] The SMS link topology reconstruction factor is a parameter generated by adjusting the distribution density of drones based on the verification results, and is used to optimize the topology of the SMS link.

[0087] The closed-loop feedback mechanism refers to combining the spatiotemporal coupling relationship with the SMS link topology reconstruction factor to form a continuous optimization process.

[0088] Updating the topology association detection path for SMS transmission is to adapt to new communication requirements and improve overall communication performance.

[0089] In this embodiment, firstly, the SMS link state fluctuation sequence recorded by each node is extracted from the blockchain network, and the number of times the SMS transmission delay gradient in the area where each drone node is located exceeds the emergency communication threshold is counted, which is recorded as the abnormal frequency. For each abnormal cluster, the number of drone nodes it covers and the fluctuation amplitude of the SMS link state migration characteristics within the cluster are calculated to generate the cluster abnormality intensity.

[0090] Secondly, for areas where the frequency of anomalies exceeds a set value, the distribution density of drones in that area is increased. Specifically, idle drones are relocated from low-anomaly-frequency areas to high-anomaly-frequency areas. If there are insufficient idle drones, new drone nodes are dynamically generated and added to the high-anomaly-frequency areas.

[0091] For areas where the frequency of anomalies is lower than the set value, the distribution density of drones in that area will be reduced. Specifically, some drones will be marked as standby or relocated to other areas.

[0092] Furthermore, based on the adjusted drone distribution density, the node coverage weight for each region is calculated. The weight is the ratio of the number of drones in that region to the number of drones in adjacent regions. Combined with the cluster anomaly strength, a text message link topology reconstruction factor is generated. This factor is a numerical value used to quantify the coverage capability of the current topology for anomalously associated clusters. The calculation formula is as follows:

[0093] Reconstruction factor = ∑(cluster anomaly intensity × node coverage weight)

[0094] Finally, the SMS link topology reconstruction factor is compared with the spatiotemporal coupling relationship: if the reconstruction factor is lower than the fluctuation amplitude of the abnormal association clusters recorded in the spatiotemporal coupling relationship, the drone distribution density in the high-abnormality frequency area is further increased. If the reconstruction factor is higher than the fluctuation amplitude of the abnormal association clusters recorded in the spatiotemporal coupling relationship, the current drone distribution density is maintained or the number of drones in the low-abnormality frequency area is slightly adjusted.

[0095] Based on the latest drone distribution density, the communication paths between nodes are recalculated, and the paths with higher node coverage weights are prioritized as the topology association detection paths for SMS transmission.

[0096] For example, in post-earthquake rescue operations, the system automatically adjusted the distribution density of drones and optimized the topology of SMS links based on previously recorded blockchain events. The new topology significantly improved communication efficiency within the disaster area, ensuring the smooth progress of rescue work.

[0097] In summary, steps 101 to 104 enabled precise monitoring and real-time feedback of the status of damaged SMS links. This method not only improves the efficiency of data collection and analysis but also enhances the system's adaptability and stability in emergency situations, providing strong technical support for responding to emergencies. The entire process, through precise data analysis, dynamic threshold adjustment, and an effective anomaly marking mechanism, constructs an efficient, flexible, and reliable emergency communication support system, ensuring the timely transmission of critical information and guaranteeing the smooth progress of rescue operations.

[0098] To further improve the monitoring and optimization of SMS link status in dynamic communication topologies, this solution details the process of identifying abnormal clusters of transmission paths based on SMS link status fluctuation sequences. This is achieved by aggregating paths through nodes and dynamically adjusting aggregation weights to establish spatiotemporal coupling relationships. Clustering analysis techniques, combined with a positive feedback mechanism, enhance the system's sensitivity and response speed to anomalies, ensuring accurate location of problem areas.

[0099] In some embodiments, step 102, which involves performing SMS link state transition detection on the dynamic communication topology based on the SMS link state fluctuation sequence, identifying abnormal association clusters of the transmission path, and forming a spatiotemporal coupling relationship with the SMS link anomaly marker, includes:

[0100] 201. Generate a node aggregation path based on the SMS link state fluctuation sequence, dynamically adjust the aggregation weight according to the delay gradient and packet loss rate of adjacent nodes, and form a positive feedback association with the triggering condition of the SMS link anomaly marker.

[0101] In step 201, the node aggregation path is a set of paths generated based on the SMS link status fluctuation sequence, containing latency gradient and packet loss rate information between adjacent nodes, used to dynamically adjust the aggregation weight. The aggregation weight is a value calculated based on the latency gradient and packet loss rate of adjacent nodes, used to reflect the connection strength between nodes. The SMS link anomaly marking trigger condition is a pre-set standard; when the SMS link status exceeds this standard, an alarm mechanism is triggered. Positive feedback correlation refers to the interaction between the aggregation weight adjustment and the anomaly marking trigger condition, to enhance the response speed to abnormal situations.

[0102] In this embodiment, a node aggregation path is first generated based on the SMS link state fluctuation sequence. Time series analysis algorithms are used to calculate the latency gradient and packet loss rate between adjacent nodes, and the aggregation weights are dynamically adjusted accordingly. Simultaneously, these adjustments are combined with the triggering conditions for SMS link anomaly marking, forming a positive feedback correlation. Ultimately, this process enhances the system's sensitivity to potential problems and its response efficiency, ensuring rapid detection and handling of anomalies.

[0103] In practical applications, for example, the latency gradient and packet loss rate data of each drone node are first extracted from the SMS link status fluctuation sequence, and the data is divided into blocks according to time windows (e.g., every 5 seconds).

[0104] For each data block, calculate the communication stability index between adjacent nodes:

[0105] Delay gradient = Current node delay - Neighboring node delay

[0106] Packet loss rate difference = current node packet loss rate - neighboring node packet loss rate

[0107] If the difference in latency gradient and packet loss rate between two nodes is lower than a set threshold, a connection is established, forming a node aggregation path (i.e., a set of stable communication paths).

[0108] Secondly, the initial aggregation weight = 1 / (delay gradient + packet loss rate difference + 1) (avoid division by zero).

[0109] If the delay gradient or packet loss rate of a certain path exceeds a dynamic threshold (such as twice the historical average), its aggregation weight is reduced (such as by multiplying by a decay coefficient of 0.8).

[0110] If the path's delay gradient and packet loss rate remain stable, gradually increase the aggregation weight (e.g., multiply by an enhancement factor of 1.2).

[0111] Finally, when the aggregate weight of a certain path drops to a critical value (such as 0.3), an SMS link anomaly flag is triggered.

[0112] Once the anomaly flag is triggered, the aggregation weight of the path is forcibly reduced to 0, and a backup drone node is prioritized to replace the path.

[0113] The aggregation weight of the new node is initialized to a high value (e.g., 0.8), forming a positive feedback loop: anomaly marking is triggered → weight adjustment → path optimization → reduction of subsequent anomalies.

[0114] 202. Perform SMS link state migration detection on the dynamic communication topology, and identify abnormal association clusters of transmission paths in the distribution characteristics of the node aggregation path and the SMS link state fluctuation sequence;

[0115] In step 202, the dynamic communication topology is a temporary communication network structure formed by a drone swarm, used to support emergency communication. SMS link state transition detection is an algorithmic tool used to monitor changes in SMS link state. Anomaly clusters refer to key factors affecting communication quality, identified through clustering analysis techniques from the distribution characteristics of node aggregation paths and SMS link state fluctuation sequences. This process helps to accurately pinpoint the source of the problem and guide subsequent optimization measures.

[0116] In this embodiment, firstly, all nodes in the dynamic communication topology are monitored in real time to collect their status information, and a text message link status fluctuation sequence is constructed using time series analysis technology. Next, a clustering algorithm is used to determine node aggregation paths based on node behavioral characteristics, aiming to discover groups of nodes with similar behavioral patterns. Then, through deep learning model analysis of the status fluctuation sequence, abnormal patterns are identified, representing abnormal association clusters. Finally, by combining the node aggregation paths with the identified abnormal patterns, specific abnormal transmission paths are further analyzed and located, and corresponding measures are taken for repair or optimization. This series of operations ensures the security and stability of the text message link.

[0117] In practical applications, the paths of each node are first aggregated, and their state transition characteristics (such as the number of latency spikes and the fluctuation range of packet loss rate) are calculated. A sliding window statistical method is used, and if the transition characteristics of a certain path exceed the 90th percentile of similar paths, it is marked as a suspicious path.

[0118] Secondly, the latency gradient and packet loss rate data of all suspicious paths are input into the clustering model and grouped according to spatial proximity (drone spacing < 100 meters) and temporal synchronization (latency change time difference < 1 second). The output shows abnormal association clusters: paths within the same group are considered to be affected by the same factor (such as regional signal interference).

[0119] 203. The spatiotemporal coupling relationship of the abnormal association cluster is quantified into a spatiotemporal coupling coefficient, and the spatiotemporal coupling coefficient forms a cross-validation relationship with the aggregation weight of the node aggregation path;

[0120] In step 203, the spatiotemporal coupling coefficient is a numerical value that quantifies the temporal and spatial distribution characteristics of anomalous clusters, used to assess their impact on the overall network. The aggregation weight is a parameter dynamically adjusted based on the inter-node delay gradient and packet loss rate. Cross-validation refers to the mutual verification between the spatiotemporal coupling coefficient and the aggregation weight to ensure the accuracy of the spatial distribution boundaries of anomalous clusters. This process improves the accuracy of anomalous cluster identification.

[0121] In this embodiment, firstly, based on the identified anomalous association clusters, the spatiotemporal distance between each cluster is calculated, including time intervals and geographical distances, thus deriving the spatiotemporal coupling coefficient. This step utilizes Geographic Information System (GIS) technology and time series analysis methods to quantify these parameters. Next, for each node's aggregation path, a graph theory algorithm is used to calculate its aggregation weight, which reflects the tightness of the node's connection in the network and its criticality in information transmission. Then, the spatiotemporal coupling coefficient and the aggregation weight are compared and analyzed to form a cross-validation relationship to confirm the authenticity and severity of the anomalous association clusters. Finally, the above results are integrated to generate a report containing all anomalous association clusters and their spatiotemporal coupling coefficients and aggregation weights, serving as a basis for optimizing the network structure and improving security.

[0122] In practical applications, the first step is to calculate the spatiotemporal coupling coefficient. This involves calculating the temporal and spatial coupling degrees for each anomalous association cluster and combining them to obtain the spatiotemporal coupling coefficient. For example, it can be calculated using the following formula:

[0123] Temporal coupling degree = temporal correlation of abrupt changes in path delay within a cluster (e.g., Pearson coefficient)

[0124] Spatial coupling degree = reciprocal of the average geographical distance between UAVs within the cluster (the closer the distance, the larger the value).

[0125] Spatiotemporal coupling coefficient = temporal coupling degree × spatial coupling degree

[0126] Secondly, cross-validation of spatiotemporal coupling coefficient and aggregation weight is performed: if a cluster has a high spatiotemporal coupling coefficient (>0.7) but a low mean aggregation weight (<0.4), then the cluster is confirmed as a true anomaly. If the spatiotemporal coupling coefficient is low but the aggregation weight is high, then it is determined to be a false detection and the cluster is removed.

[0127] 204. Based on the cross-validation relationship, correct the spatial distribution boundary of the abnormal association cluster, generate abnormal association cluster correction parameters, and feed them back to the dynamic spatial association analysis process of the SMS link state fluctuation sequence through the node aggregation path.

[0128] In step 204, the cross-validation relationship refers to the comparative analysis result between the spatiotemporal coupling coefficient and the aggregation weight of the node aggregation path, used to assess the authenticity and severity of the anomalous association cluster. The spatial distribution boundary of the anomalous association cluster refers to the boundary that defines the influence range of the anomalous association cluster in geographic space. The anomalous association cluster correction parameter is an adjustment parameter to the original spatial distribution boundary based on the cross-validation results, used to more accurately describe the actual impact area of ​​the anomaly. The dynamic spatial association analysis process is a process of continuously monitoring and analyzing the SMS link state fluctuation sequence, aiming to identify potential spatial association patterns in the network.

[0129] In this embodiment, firstly, based on the established cross-validation relationships, the spatial distribution boundary of each anomalous association cluster is analyzed to determine whether adjustment is needed. This step employs Geographic Information System (GIS) technology combined with machine learning algorithms to determine the correction direction and magnitude by comparing the spatiotemporal coupling coefficient and aggregation weights. Next, anomalous association cluster correction parameters are generated, which include the adjusted spatial distribution boundary information. Then, these correction parameters are fed back into the optimization process of node aggregation paths, using graph theory algorithms to recalculate the importance weights of each node and update the node aggregation paths. Finally, the corrected node aggregation path information is integrated into the dynamic spatial correlation analysis of the SMS link status fluctuation sequence, enabling real-time monitoring and optimization of the SMS link status. This entire process ensures a more accurate spatial distribution of anomalous association clusters, improving the system's response speed and accuracy to anomalies.

[0130] In practical applications, the spatial distribution boundary is first corrected, and for confirmed anomalous clusters, the range of influence is adjusted according to their spatiotemporal coupling coefficient.

[0131] If the spatiotemporal coupling coefficient is >0.8, expand the boundary to the adjacent node (e.g., increase the radius by 50 meters).

[0132] If the spatiotemporal coupling coefficient is less than 0.5, the boundary is narrowed down to the core node (e.g., only the path with the most significant time delay mutation is retained).

[0133] Generate correction parameters for abnormal association clusters (such as new boundary coordinates and influence intensity).

[0134] Secondly, the feedback is sent to dynamic spatial correlation analysis:

[0135] The corrected parameters are input into the node aggregation path generation module to forcibly reduce the aggregation weight of paths within abnormal clusters.

[0136] Update the analysis strategy for SMS link status fluctuation sequences, and prioritize monitoring nodes within the correction boundary during subsequent detection.

[0137] Here is a specific example:

[0138] In a rescue scenario following a sudden earthquake, drones were rapidly deployed over the disaster area, establishing a temporary communication network. A real-time data acquisition module began collecting performance metrics of the SMS link in the disaster area. After a period of data accumulation, the system generated node aggregation paths and dynamically adjusted aggregation weights. Subsequently, clustering analysis was used to identify several key anomalous clusters. Next, the spatiotemporal coupling relationships of these clusters were quantified into spatiotemporal coupling coefficients, which were cross-validated with the aggregation weights to correct the spatial distribution boundaries of the anomalous clusters. Finally, the system automatically adjusted the drone distribution density, optimized the SMS link topology, significantly improved communication efficiency within the disaster area, and ensured the smooth progress of rescue operations.

[0139] In summary, steps 201 to 204, from anomaly detection to optimization, form a complete closed-loop feedback mechanism. This method can dynamically optimize SMS transmission paths in complex environments, greatly improving the reliability and response speed of SMS services in emergency situations. Specifically, it enables rapid deployment of temporary communication facilities, real-time monitoring and marking of link anomalies, accurate identification of the root cause of problems, and ensures data accuracy through spatiotemporal coupling, ultimately achieving efficient and flexible emergency communication support. This system not only improves the adaptability and stability of the communication network but also provides strong support at critical moments.

[0140] In some embodiments, step 203, quantifying the spatiotemporal coupling relationship of the abnormal association cluster into a spatiotemporal coupling coefficient, wherein the spatiotemporal coupling coefficient forms a cross-validation relationship with the aggregation weight of the node aggregation path, includes:

[0141] 301. Based on the three-dimensional coordinate distribution of UAV nodes in the emergency area, a coverage density index is generated by dividing the area into geographic grid units and calculating the ratio of the number of active nodes to the grid volume.

[0142] In step 301, the coverage density index is an indicator generated based on the three-dimensional coordinate distribution of UAV nodes within the emergency area. It is calculated by dividing the area into geographic grid cells and statistically analyzing the ratio of the number of active nodes to the grid volume. This index is used to assess the density of communication resources within a given area, thereby guiding the effective allocation of resources. A geographic grid cell refers to dividing the emergency area into multiple small geographic units, which facilitates statistical analysis.

[0143] In this embodiment, firstly, the area is divided into uniform geographic grid units (e.g., a 100m × 100m × 50m cube) based on the three-dimensional coordinates (longitude, latitude, and altitude) of all UAV nodes within the emergency area. Secondly, the number of active UAV nodes in each grid unit is counted, and the ratio of this number to the grid volume is calculated to obtain a coverage density index (e.g., number of nodes / m³). Then, the index is smoothed to eliminate the noise influence of edge grids. Finally, the coverage density index of each grid unit is output to assess the density of regional communication resource distribution.

[0144] 302. Statistically analyze the trigger frequency distribution of SMS link anomaly markers according to a preset time window, generate an anomaly frequency histogram, and extract the duration of the peak interval in the anomaly frequency histogram as an anomaly time clustering factor.

[0145] In step 302, the anomaly frequency histogram is the result of statistically analyzing the frequency distribution of SMS link anomaly marker triggers within a preset time window, used to identify the frequency patterns of anomalies. The anomaly time clustering factor extracts the duration of peak intervals from the anomaly frequency histogram to quantify the temporal concentration of anomalies. This factor reflects the clustering characteristics of anomaly events within a specific time period.

[0146] In this embodiment, firstly, the number of times SMS link anomaly markers are triggered is counted according to a preset time window (e.g., every 10 minutes), generating an anomaly frequency histogram of time-frequency distribution. Secondly, the highest frequency interval in the histogram (e.g., frequency > 5 times in 3 consecutive windows) is identified using a peak detection algorithm, and the duration of this interval is calculated. Then, this duration is normalized into an anomaly time clustering factor (e.g., 0~1, where 1 indicates sustained high anomaly). Finally, the factor weights are dynamically adjusted based on historical data to ensure sensitivity to sudden anomalies.

[0147] 303. The coverage density index and the abnormal time clustering factor are weighted and fused to generate an initial spatiotemporal coupling coefficient, while the weight of the coverage density index is dynamically adjusted according to the priority of the emergency area.

[0148] In step 303, the initial spatiotemporal coupling coefficient is the result of a weighted fusion of the coverage density index and the anomalous temporal clustering factor, used to comprehensively assess the spatiotemporal coupling relationship. The weight of the coverage density index is dynamically adjusted according to the priority of the emergency area to adapt to changes in needs under different emergency situations. This process enables the spatiotemporal coupling coefficient to reflect the actual situation within the area, providing a more accurate assessment.

[0149] In this embodiment, firstly, the coverage density index of step 301 and the abnormal time clustering factor of step 302 are normalized (e.g., Z-score standardization). Secondly, weights are dynamically allocated according to the priority of emergency areas (e.g., coverage density weight accounts for 70% and time clustering factor accounts for 30% in high-priority areas). Then, an initial spatiotemporal coupling coefficient is generated using a weighted summation formula (e.g., coefficient = 0.7 × density index + 0.3 × time factor). Finally, the coefficients are thresholded (e.g., >0.6 indicates high-risk coupling areas), and grid cells requiring key monitoring are marked.

[0150] 304. Extract the weight distribution entropy value of the central region of the abnormal association cluster, and quantify the spatiotemporal coupling relationship into a spatiotemporal coupling coefficient based on the difference between the weight distribution entropy value and the initial spatiotemporal coupling coefficient;

[0151] In step 304, the weight distribution entropy value is an uncertainty measure of the importance of node aggregation paths within the central region of the anomalous association cluster, used to assess the uniformity of node distribution within that region. The initial spatiotemporal coupling coefficient is a value calculated based on the spatiotemporal distance at the time of initial identification of the anomalous association cluster, used to measure the strength of its spatial and temporal interactions. The spatiotemporal coupling coefficient is a quantitative indicator adjusted based on the difference between the weight distribution entropy value and the initial spatiotemporal coupling coefficient, used to more accurately describe the spatiotemporal characteristics of the anomalous association cluster.

[0152] In this embodiment, firstly, the weight distribution of node aggregation paths within the central region of the anomalous association cluster is calculated, and the entropy value of the weight distribution is calculated using the entropy formula from information theory to evaluate the uniformity and complexity of the node distribution. Next, this entropy value is compared with a previously determined initial spatiotemporal coupling coefficient to analyze the degree of difference between the two. Then, the spatiotemporal coupling coefficient is adjusted based on this difference, and an adaptive algorithm is used to incorporate new spatiotemporal features, thereby updating the spatiotemporal coupling coefficient. Finally, all data is integrated to obtain a spatiotemporal coupling coefficient that more accurately reflects the spatiotemporal characteristics of the anomalous association cluster for subsequent analysis.

[0153] Specifically, first, the aggregate weights of all nodes in the central region of the anomalous cluster (e.g., within a radius of 200m) are extracted, and their weight distribution entropy is calculated (a higher entropy value indicates a more chaotic weight distribution). Second, the difference between the entropy value and the initial spatiotemporal coupling coefficient is compared (e.g., difference = |entropy value - initial coefficient|). Then, if the difference exceeds the tolerance (e.g., >0.2), the spatiotemporal coupling coefficient is adjusted proportionally (e.g., new coefficient = initial coefficient × (1 + difference)). Finally, the corrected spatiotemporal coupling coefficient is output to characterize the spatiotemporal stability of the anomalous cluster.

[0154] 305. Construct an aggregated weight-spatiotemporal coupling cross-validation matrix. Generate a validation factor by comparing the correlation between the aggregated weight gradient change rate and the update magnitude of the spatiotemporal coupling coefficient element by element. When the validation factor is lower than the confidence threshold, trigger the spatial distribution boundary reconstruction of the abnormal association cluster.

[0155] In step 305, the aggregation weight-spatiotemporal coupling cross-validation matrix is ​​a two-dimensional matrix containing data on the rate of change of the aggregation weight gradient and the update magnitude of the spatiotemporal coupling coefficient. The rate of change of the aggregation weight gradient reflects the rate of change of the importance of the node aggregation path over time, while the update magnitude of the spatiotemporal coupling coefficient represents the degree of change of the spatiotemporal coupling coefficient. The validation factor is a comprehensive evaluation index generated based on the element-by-element comparison of the above two types of data, used to determine whether the spatial distribution boundary of the abnormal association cluster needs to be reconstructed.

[0156] In this embodiment, firstly, an aggregated weight-spatiotemporal coupling cross-validation matrix is ​​constructed, and data on the rate of change of aggregated weight gradients and the update magnitude of spatiotemporal coupling coefficients are collected. Next, statistical methods are applied to compare these two sets of data element by element to calculate the correlation between them. Then, a validation factor is generated based on the correlation results. If this factor is lower than a preset confidence threshold, it indicates that the current spatial distribution boundary may be inaccurate or has failed. Finally, when the validation factor is lower than the confidence threshold, the reconstruction process of the spatial distribution boundary of the abnormal association cluster is automatically triggered to ensure the high accuracy and reliability of the system.

[0157] Specifically, first, an aggregated weight-spatiotemporal coupling cross-validation matrix is ​​constructed. Rows represent time windows, columns represent spatial grids, and element values ​​are the ratio of the weight gradient change rate to the update magnitude of the spatiotemporal coupling coefficient. Second, the deviation of each ratio from the historical mean is calculated element-by-element to generate a validation factor (e.g., factor = 1 - |current ratio / mean - 1|). Then, if the factor is below a confidence threshold (e.g., <0.7), the spatiotemporal coupling is deemed to have failed. Finally, the spatial distribution boundary reconstruction of the abnormal association cluster is triggered, and steps 301-304 are re-executed to update the boundary parameters.

[0158] Here is a specific example:

[0159] In a rescue scenario following a sudden earthquake, drones were rapidly deployed over the disaster area to establish a temporary communication network. First, based on the 3D coordinate distribution of drone nodes, geographical grid cells were divided and a coverage density index was calculated. Next, the trigger frequency of SMS link anomaly markers was statistically analyzed, and anomaly time clustering factors were extracted. Then, the coverage density index and anomaly time clustering factors were weighted and fused, with the weights dynamically adjusted according to the priority of emergency areas. Afterward, the weighted distribution entropy value of the central region of the anomaly association clusters was calculated. Finally, an aggregated weight-spatiotemporal coupling cross-validation matrix was constructed, and a validation factor was generated. When the validation factor fell below a confidence threshold, the spatial distribution boundary of the anomaly association clusters was reconstructed, optimizing communication efficiency within the disaster area and ensuring the smooth progress of rescue work.

[0160] In summary, steps 301 to 305 achieve precise quantification and effective verification of the spatiotemporal coupling relationship of anomaly clusters. This method not only improves the accuracy of anomaly detection in dynamic communication topologies but also enhances the system's self-optimization capabilities in complex environments, greatly improving the reliability and response speed of emergency communication. This solution provides strong technical support for responding to emergencies, ensuring the timely transmission of critical information. The entire process, through precise data analysis, dynamic threshold adjustment, and an effective anomaly marking mechanism, constructs an efficient, flexible, and secure emergency communication support system.

[0161] In some embodiments, the step 101 of triggering a text message link anomaly flag when the text message transmission delay gradient in the text message link state fluctuation sequence exceeds the emergency communication threshold includes:

[0162] 401. Analyze the SMS transmission delay gradient based on the difference between adjacent time windows in the SMS link state fluctuation sequence, and normalize it by the cumulative mean of the delay change rate within the sliding window to generate a normalized SMS transmission delay gradient.

[0163] In step 401, the SMS link state fluctuation sequence records the transmission status of SMS messages at different points in time. The difference between adjacent time windows refers to the change in SMS transmission delay within two consecutive time periods. The SMS transmission delay gradient represents the speed and direction of these delay changes. The sliding window technique is used to observe the trend of data changes over a certain period of time. The cumulative mean is the process of averaging all delay change rates within the sliding window. Normalization is the process of converting the data into a standard range to facilitate comparison and analysis.

[0164] In this embodiment, firstly, the system extracts latency difference data for adjacent time windows (e.g., 5-second intervals) from the SMS link state fluctuation sequence. Secondly, it uses a sliding window technique (window length 30 seconds, step size 5 seconds) to calculate the cumulative mean of latency change rate within each window. Then, it converts the cumulative mean into a normalized SMS transmission latency gradient value within the range of 0-1 using a maximum-minimum normalization method. Finally, the system establishes a latency gradient change curve to provide basic data for subsequent threshold adjustment.

[0165] 402. Based on the historical distribution characteristics of the normalized SMS transmission delay gradient and the gradient change trend within the current window, dynamically adjust the emergency communication threshold;

[0166] In step 402, historical distribution characteristics refer to the data distribution of the normalized SMS transmission delay gradient over a past period. The gradient change trend within the current window reflects the development direction of the SMS transmission delay gradient in the latest time period. The emergency communication threshold is a standard used to trigger alarms or take action, dynamically adjusted based on real-time data to adapt to constantly changing environmental conditions.

[0167] In this embodiment, firstly, the historical distribution characteristics of the normalized SMS transmission delay gradient are analyzed to determine its normal fluctuation range. Next, the gradient change trend within the current window is evaluated to determine if any abnormal growth has occurred. Then, based on these two aspects of information, the emergency communication threshold is dynamically adjusted to ensure it can reflect the latest network conditions in a timely manner. Finally, the adjusted threshold is applied to the real-time monitoring system to quickly respond to any situations exceeding the threshold.

[0168] Specifically, first, the system analyzes the normalized delay gradient data from the past 24 hours, calculating its mean μ and standard deviation σ as historical distribution characteristics. Second, it monitors the gradient change trend within the current sliding window in real time. When a monotonically increasing trend is detected for three consecutive windows, a dynamic threshold adjustment mechanism is activated. Then, based on the current network load status (such as UAV node density and channel utilization) and the 3σ principle of historical data, the emergency communication threshold is set to a dynamic value between μ+2σ and μ+3σ. Finally, the threshold parameters are recalculated and updated every 5 minutes.

[0169] 403. When the normalized SMS transmission delay gradient exceeds the adjusted emergency communication threshold, an anomaly marker trigger condition is generated by combining the duration of the time exceeding the adjusted emergency communication threshold with the number of times it exceeds the threshold.

[0170] In step 403, the anomaly marking trigger condition is generated based on whether the normalized SMS transmission delay gradient exceeds the adjusted emergency communication threshold, the duration of the exceedance, and the number of times it occurs. This condition is used to determine when to trigger an SMS link anomaly marking, so as to detect potential problems in a timely manner. The duration of the exceedance refers to the duration of the exceedance of the threshold, and the number of exceedances refers to the frequency of exceeding the threshold. These parameters work together to ensure the accuracy and timeliness of anomaly marking.

[0171] In this embodiment, firstly, the system monitors whether the normalized SMS transmission delay gradient exceeds the adjusted emergency communication threshold. Next, it records the duration and cumulative number of times the threshold is exceeded. Then, based on pre-defined rules, if both the time and the number of occurrences meet preset conditions, an anomaly marker trigger condition is generated. Finally, these conditions are applied to the SMS link anomaly analysis module to initiate further inspection and response procedures.

[0172] Specifically, first, the system compares the normalized time delay gradient with a dynamic threshold in real time. When the gradient value exceeds the threshold, timing and counting begin. Second, two levels of triggering conditions are set: a primary condition is a single instance of exceeding the threshold for more than 10 seconds, and a secondary condition is a cumulative total of 3 instances of exceeding the threshold within 1 minute. Then, when either condition is met, a corresponding anomaly marker trigger signal is generated, with the secondary condition triggering a more severe anomaly level. Finally, the system records detailed parameters for each trigger (exceedance magnitude, duration, location, etc.) to form an anomaly event log.

[0173] 404. The parameters corresponding to the normalized SMS transmission delay gradient, the adjusted emergency communication threshold, and the anomaly marker triggering condition are used as associated parameters and passed to the subsequent SMS link anomaly analysis module to trigger the SMS link anomaly marker.

[0174] In step 404, the associated parameters include the normalized SMS transmission delay gradient, the adjusted emergency communication threshold, and parameters corresponding to the anomaly marking trigger conditions. These parameters are passed to the subsequent SMS link anomaly analysis module to trigger SMS link anomaly marking. This step achieves effective data transmission and integration, providing a foundation for subsequent analysis. The associated parameters not only contain current delay change information but also reflect historical trends and real-time conditions, helping to comprehensively assess the status of the SMS link.

[0175] In this embodiment, firstly, data such as the normalized SMS transmission delay gradient, adjusted emergency communication threshold, and anomaly marker triggering conditions are collected. Next, this data is packaged into correlation parameters and sent to the SMS link anomaly analysis module. Then, this module uses these parameters for in-depth analysis to identify potential anomalies. Finally, based on the analysis results, a decision is made on whether to trigger an SMS link anomaly marker so that appropriate corrective measures can be taken.

[0176] Specifically, first, the system packages the normalized delay gradient curve, dynamic threshold curve, and trigger condition parameters into a structured data packet. Second, the data packet is transmitted to the anomaly analysis module in real time via a dedicated message queue, ensuring a transmission latency of less than 100ms. Then, after parsing the data packet, the anomaly analysis module first performs rapid pattern matching (e.g., determining whether it matches known fault characteristics), and then performs in-depth analysis (e.g., machine learning model prediction). Finally, when the anomaly is confirmed to be valid, the system marks the anomaly on the topology map of the affected area and sends alarm commands to the relevant UAV nodes via the control channel.

[0177] Here is a specific example:

[0178] In a rescue scenario following a sudden earthquake, drones were rapidly deployed over the disaster area, establishing a temporary communication network. The real-time data acquisition module began collecting various performance indicators of the SMS transmission link in the disaster area. First, it calculated the latency difference between adjacent time windows, and then calculated and normalized the SMS transmission latency gradient using a sliding window technique. Next, based on historical data and the gradient change trend within the current window, it dynamically adjusted the emergency communication threshold. When the normalized SMS transmission latency gradient exceeded the adjusted emergency communication threshold, an anomaly marker was generated. Finally, the SMS link anomaly marker was triggered, ensuring smooth communication during the rescue operation.

[0179] In summary, steps 401 to 404 achieve accurate monitoring and rapid response to SMS link status. This method not only improves the accuracy of latency gradient analysis but also enhances the system's self-optimization capabilities in complex environments, significantly improving the reliability and response speed of emergency communications. This solution provides strong technical support for responding to emergencies, ensuring the timely transmission of critical information.

[0180] In some embodiments, step 101, which involves deploying a drone swarm to form a dynamic communication topology to collect real-time data on the latency mutation rate and packet loss rate of damaged SMS links and generate an SMS link state fluctuation sequence, includes:

[0181] 501. Based on the spatial distribution of the damaged SMS links and the current channel quality, control the signal strength and interference level of each node in the UAV swarm, generate dynamic weight parameters, and form a dynamic communication topology through iterative updates of the UAV swarm position.

[0182] In step 501, the dynamic weight parameters are generated based on the spatial distribution of the damaged SMS links and the signal strength and interference level of each node in the UAV swarm, controlling the current channel quality. Signal strength refers to the strength of the signal transmitted by the UAV nodes, used to ensure communication quality; interference level reflects the degree of interference from the surrounding environment on signal transmission. The dynamic communication topology is a temporary communication network structure formed by the UAV swarm through position iteration updates, used to support emergency communication. Position iteration update is an algorithm that optimizes the overall communication topology by adjusting the positions of the UAVs.

[0183] In this embodiment, the optimal signal strength and interference level of each UAV node are first assessed based on the spatial distribution of the damaged SMS links and the current channel quality, generating dynamic weight parameters. Next, an iterative position update algorithm for the UAV swarm is used to adjust the positions of each node to optimize the overall communication topology. This process employs wireless channel analysis techniques and adaptive optimization algorithms to ensure the stability and efficiency of the communication network. Ultimately, this process forms a dynamic communication topology capable of rapidly responding to changes.

[0184] Specifically, firstly, the system acquires a spatial distribution heatmap of damaged SMS links and real-time channel quality indicators (such as SNR and RSSI) through GPS positioning and channel sounding. Secondly, based on reinforcement learning algorithms, it calculates the optimal signal transmission power (signal strength) and frequency band selection strategy (interference level) for each UAV node, generating dynamic weight parameters that include location weight, channel weight, and timeliness weight. Then, a distributed consensus algorithm is used to coordinate cluster nodes, gradually approximating the optimal topology through three-dimensional position iteration (updated every 30 seconds). Finally, a self-healing dynamic communication topology network is formed, where key nodes are deployed with N+1 redundancy to ensure reliability.

[0185] 502. In the dynamic communication topology, the latency mutation rate and packet loss rate data of the damaged SMS link are collected at a preset period, and the packet loss rate data is supplemented with the packet loss rate variance within the collection time window as a fluctuation feature.

[0186] In step 502, the latency mutation rate refers to the rate of latency change of the damaged SMS link within a specific time window, used to describe the latency changes during SMS transmission; the packet loss rate data includes the variance of the packet loss rate within the collection time window as a fluctuation feature, used to describe the proportion of data packets lost during data transmission and their fluctuations. These data are collected at a preset period to reflect the real-time status of the SMS link. The preset period refers to a set time interval within which data is collected to ensure the timeliness and accuracy of the data.

[0187] In this embodiment, firstly, a damaged SMS link is selected in the dynamic communication topology, and data is collected at a preset period (e.g., every 5 minutes). Next, the latency mutation rate of the selected link is calculated, which involves differential processing of latency data at consecutive time points to determine the rate of latency change. Then, packet loss rate data within the same time period is collected, and the variance of the packet loss rate within that time window is calculated to quantify the fluctuation of network performance. Finally, the latency mutation rate, packet loss rate, and their variance are integrated into a complete dataset as the basis for evaluating the health status of the SMS link.

[0188] Specifically, first, the system identifies all damaged SMS links in the dynamic topology and initiates data collection according to an adaptive collection cycle (initially 5 minutes, dynamically adjusted based on link stability). Second, it uses latency probe technology to measure end-to-end latency and calculates the latency abrupt change rate (unit: ms / s) using second-order difference. Then, it employs deep packet inspection technology to statistically analyze the packet loss rate and simultaneously calculates the variance of the packet loss rate within a 30-second time window as a fluctuation characteristic. Finally, it appends a timestamp accurate to milliseconds and node location information to each data sample, forming a structured monitoring data packet.

[0189] 503. The latency mutation rate and packet loss rate data are timestamped and spliced ​​with a sliding window to generate a time-aligned collection dataset. The collection dataset is then weighted and fused based on the dynamic weight parameters to generate a text message link status fluctuation sequence.

[0190] In step 503, the time-aligned acquisition dataset is a collection of data generated by timestamping and concatenating the latency mutation rate and packet loss rate data using a sliding window. The acquisition dataset is then weighted and fused based on dynamic weight parameters to generate a SMS link status fluctuation sequence. Timestamp alignment refers to aligning data from different sources in chronological order to ensure data consistency; sliding window concatenation is a technique that stitches data from consecutive time periods into a continuous time series. Dynamic weight parameters are used to weight and fuse data from different time periods to generate the final status fluctuation sequence. This sequence contains information on link latency changes and data loss, used to assess link quality.

[0191] In this embodiment, the latency mutation rate and packet loss rate data are first timestamped, and then a sliding window technique is used to concatenate data from different time periods into a continuous time series. Next, these data are weighted and fused using dynamic weight parameters to generate a SMS link state fluctuation sequence. Time series analysis and data fusion techniques are applied in this process to ensure data consistency and reliability. The final generated state fluctuation sequence provides a solid foundation for subsequent anomaly detection and optimization.

[0192] Specifically, firstly, the system uses a time-series alignment algorithm to synchronize the collected data from different nodes at the microsecond level according to GPS clocks. Secondly, a continuous time-domain dataset is constructed using a sliding window stitching technique (window length 1 minute, overlap rate 50%). Then, a three-level weighted fusion is performed based on dynamic weight parameters: spatial weight (node ​​importance), temporal weight (data freshness), and quality weight (signal-to-noise ratio). Finally, a SMS link state fluctuation sequence containing latency mutation rate, packet loss rate, and variance is generated, encapsulated in JSON format, and digitally signed to ensure data integrity.

[0193] Here is a specific example:

[0194] In a rescue scenario following a sudden earthquake, drones were rapidly deployed over the disaster area to establish a temporary communication network. First, based on the spatial distribution of damaged SMS links and the current channel quality, the optimal signal strength and interference level of each drone node were assessed, generating dynamic weight parameters. Next, an iterative position update algorithm for the drone swarm was used to adjust the positions of each node, resulting in a dynamic communication topology. Then, data on the latency mutation rate and packet loss rate of the damaged SMS links were collected at preset intervals (e.g., every minute). Finally, a sliding window technique was used to stitch the collected data into a time-aligned dataset, generating a sequence of SMS link status fluctuations. This process not only improved the accuracy of data collection but also enhanced the system's optimization capabilities in complex environments, significantly improving the reliability and response speed of emergency communications.

[0195] In summary, steps 501 to 503 achieved accurate monitoring and real-time feedback of the status of damaged SMS links. This method not only improves the efficiency of data collection and analysis but also enhances the system's adaptability and stability in emergency situations, providing strong technical support for responding to emergencies. This solution ensures the timely transmission of critical information and guarantees the smooth progress of rescue operations. The entire process, through dynamically adjusting the communication topology, accurately collecting link performance data, and generating status fluctuation sequences, constructs an efficient and flexible emergency communication support system.

[0196] In some embodiments, step 503, which involves timestamp alignment and sliding window stitching of the latency mutation rate and packet loss rate data to generate a time-aligned acquisition dataset, includes:

[0197] 601. Based on the clock synchronization protocol of each node in the UAV cluster, a reference time axis is determined, and the timestamps of the collected latency mutation rate and packet loss rate data are mapped to the reference time axis to generate a timestamp mapping table.

[0198] In step 601, the reference timeline is a standard timeline determined based on the clock synchronization protocol of each node in the UAV swarm, used to unify the time reference of each node. Clock synchronization protocols, such as Precision Time Protocol (PTP) or Network Time Protocol (NAT), ensure that all nodes use the same time reference. The timestamp mapping table maps the timestamps of the collected latency mutation rate and packet loss rate data to the reference timeline, ensuring that all data points are processed within the same time frame. This eliminates data discrepancies caused by time asynchrony.

[0199] In this embodiment, firstly, a clock synchronization protocol is applied in the drone swarm to ensure that the time of all nodes remains consistent, thereby establishing a unified reference timeline. Next, latency mutation rate and packet loss rate data uploaded by each node are collected, and their original timestamps are recorded. Then, these original timestamps are transformed according to the reference timeline to generate a timestamp mapping table, ensuring that each data point accurately corresponds to a unified time coordinate. Finally, all data and corresponding timestamp mapping information are integrated to form a complete timestamp mapping table for subsequent data analysis.

[0200] Specifically, firstly, the drone swarm uses PTP (Precise Time Protocol) to achieve microsecond-level clock synchronization, establishing a UTC timeline based on the master control node. Secondly, the latency mutation rate and packet loss rate data collected by each node are accompanied by local timestamps, and a time conversion algorithm is used to uniformly map each node's timestamps to the reference timeline. Then, a three-column table containing the original timestamp, the reference timestamp, and the data content is established as a timestamp mapping table. Finally, the mapping results are validated for consistency to ensure that the time error does not exceed 1 millisecond.

[0201] 602. Based on the timestamp mapping table, perform linear interpolation to fill the latency mutation rate and packet loss rate data with missing timestamps, using the mean of adjacent valid data points as the interpolation reference value, and generate the interpolated latency mutation rate and packet loss rate data.

[0202] In step 602, linear interpolation imputation is a method used to fill in missing timestamps in data points. The mean of adjacent valid data points is used as the interpolation reference value to ensure the accuracy of the interpolation result. The interpolated time delay mutation rate and packet loss rate data refer to the time series data after interpolation processing, ensuring the continuity and integrity of the data. This method can effectively reduce the impact of data loss and improve data availability.

[0203] In this embodiment, firstly, the timestamp mapping table is checked to identify data points with missing timestamps in the latency mutation rate and packet loss rate. Next, for each missing data point, the nearest valid data point is found, and the average of these two points is calculated as the interpolation reference value. Then, a linear interpolation method is used to fill in the missing data points based on the interpolation reference value. Finally, all original data and interpolation results are integrated to generate a complete interpolated latency mutation rate and packet loss rate dataset, ensuring data continuity and uninterrupted processing.

[0204] Specifically, first, the system scans the timestamp mapping table to identify data breakpoints where the time interval exceeds twice the sampling period. Second, for each missing point, the system locates the nearest valid data points (within a time window ± 3 sampling periods) and calculates their arithmetic mean as the interpolation benchmark. Then, a linear weighted algorithm is used for interpolation, assigning higher weights (0.7:0.3) to data points that are closer together. Finally, a complete dataset containing the original data and the blue interpolated data is generated, and the source of the interpolation is marked on the data flags.

[0205] 603. Divide the interpolated delay mutation rate and packet loss rate data into multiple overlapping sliding windows, perform first and last timestamp calibration and boundary data smoothing, and generate delay mutation rate segments and packet loss rate segments aligned within the windows.

[0206] In step 603, a sliding window refers to dividing the data into multiple overlapping time periods, each containing a certain number of data points. First and last timestamp alignment adjusts the start and end timestamps of each window to ensure time alignment. Boundary data smoothing smooths the data at the window edges to reduce noise interference. The time delay abrupt change rate and packet loss rate segments aligned within the window refer to the time series data segments after the above processing. These processing measures help improve data quality and consistency.

[0207] In this embodiment, the interpolated latency mutation rate and packet loss rate data are divided into multiple overlapping sliding windows. First and last timestamp calibration is performed on each window to ensure time alignment. Next, the data at the window edges is smoothed to reduce noise interference. This process employs sliding window technology and data smoothing algorithms to ensure data time alignment and quality. Finally, aligned latency mutation rate and packet loss rate segments within the windows are generated.

[0208] Specifically, first, the continuous data stream is divided into sliding windows of 60 seconds each with a 30% overlap. Second, boundary calibration is performed on each window: the front boundary is aligned to the nearest whole second, and the back boundary ensures that the entire data period is included. Then, a Savitzky-Golay filter is used for boundary smoothing, and a 21-point quadratic polynomial is used to fit the edge data. Finally, the calibrated and smoothed delay abruptness rate and packet loss rate segments are output, each accompanied by a window ID and boundary marker.

[0209] 604. The latency mutation rate segment and the packet loss rate segment are concatenated in chronological order to form a unified latency and packet loss joint data block with a unified time dimension, and the time coverage and data density are extracted as window metadata to generate a time-aligned acquisition dataset.

[0210] In step 604, the joint latency and packet loss data block is a unified time-dimensional data set formed by concatenating latency mutation rate segments and packet loss rate segments in chronological order. Window metadata includes information such as time coverage and data density, used to describe the data characteristics of each window. The time-aligned acquisition dataset is a collection of metadata from multiple windows, used for subsequent analysis. This metadata helps to better understand the overall data situation and supports more refined data analysis and optimization.

[0211] In this embodiment, firstly, the latency mutation rate and packet loss rate data processed in step 602 are arranged in chronological order and merged into a joint latency and packet loss data block with a unified time dimension. Next, the time coverage and data density of this data block are calculated, and this information is extracted as window metadata. Then, based on the window metadata, the organization of the dataset is further optimized to ensure that all data are under the same time reference. Finally, all processing results are integrated to generate the final time-aligned acquisition dataset, providing a prepared data foundation for subsequent analysis.

[0212] Specifically, first, the processed data fragments are concatenated into continuous data blocks according to time sequence, establishing a two-dimensional latency-packet loss rate matrix. Second, the key metadata for each data block is calculated: time coverage range (start and end time difference) and data density (effective sampling points / theoretical sampling points). Then, a B+ tree index structure is used to organize the data blocks, with the time range as the primary key to build a fast query index. Finally, a complete collection dataset containing raw data, processed data, and metadata is generated and stored in a columnar compressed format.

[0213] Here is a specific example:

[0214] In a rescue scenario following a sudden earthquake, to ensure synchronized operation of all devices in a drone swarm, the system employed a clock synchronization protocol to create a baseline timeline. Subsequently, latency mutation rate and packet loss rate data collected from damaged SMS links were assigned raw timestamps and converted to their positions on the baseline timeline using a timestamp mapping table. When missing latency mutation rate and packet loss rate data were found within certain time periods, the system used linear interpolation to fill in the gaps. By calculating the mean of adjacent valid data points as the interpolation baseline, the system successfully filled these gaps, making the entire dataset more complete. The system then concatenated all processed latency mutation rate and packet loss rate data fragments in chronological order, forming a unified latency and packet loss data block with a consistent time dimension. By analyzing the temporal coverage and data density of this data, the system extracted key window metadata. This series of operations helped the rescue team better understand the trends in network status changes, take timely measures to address potential problems, and improve rescue efficiency and success rate.

[0215] In summary, steps 601 to 604 achieved accurate monitoring and real-time feedback of the status of damaged SMS links. This method not only improves the efficiency of data collection and analysis but also enhances the system's adaptability and stability in emergency situations, providing strong technical support for responding to emergencies. The entire process, through techniques such as time alignment, interpolation padding, and sliding window splicing, constructs an efficient and flexible emergency communication support system. This solution ensures the timely transmission of critical information and guarantees the smooth progress of rescue operations. Ultimately, these steps work together to form a system capable of rapidly responding to changes and providing high-quality data support, greatly improving the reliability and efficiency of emergency communications.

[0216] In some embodiments, step 103, which involves binding the abnormal association cluster and the SMS link state transition feature to timestamps for multi-node parallel notarization and generating a blockchain state change event containing the SMS link state fluctuation sequence, includes:

[0217] 701. The abnormal feature vector of the abnormal association cluster and the dynamic feature vector of the SMS link state transition feature are timestamped through a time window sliding alignment mechanism to generate an abnormal state joint feature vector group.

[0218] In step 701, the anomaly feature vector is a dataset describing the characteristics of anomaly-related clusters, including metrics such as latency mutation rate and packet loss rate; the dynamic feature vector is a dataset reflecting the characteristics of SMS link state transitions, such as transmission path changes and node aggregation weights. The time window sliding alignment mechanism is a technique that adjusts the time window to ensure consistency in timestamps from different sources, generating a joint feature vector group for anomaly states. This vector group contains time-synchronized data for all relevant features, used for subsequent analysis and processing.

[0219] In this embodiment, the abnormal feature vectors of the abnormal association clusters and the dynamic feature vectors of the SMS link state transition features are first timestamped using a time window sliding alignment mechanism. Sliding window technology and time series analysis algorithms are employed to ensure that data from different sources are processed within the same timeframe. The resulting joint feature vector group of abnormal states contains time-synchronized data for all relevant features, providing a foundation for subsequent data processing.

[0220] Specifically, firstly, the system extracts multi-dimensional anomaly feature vectors (including metrics such as latency mutation rate, packet loss rate variance, and spatial distribution density) from anomaly association clusters, and simultaneously extracts dynamic feature vectors (including parameters such as path jump count and weight change gradient) from state transition features. Secondly, the system uses a Dynamic Time Warping (DTW) algorithm to align the time dimensions of the two feature sequences, achieving millisecond-level timestamp synchronization through a sliding window mechanism (window length 10 seconds, step size 2 seconds). Then, the aligned feature vectors are standardized (Z-score normalization), and finally, they are concatenated in time order to generate a unified anomaly state joint feature vector group, with each vector accompanied by a timestamp label accurate to milliseconds.

[0221] 702. Divide the joint feature vector group of the abnormal state into multiple parallel evidence storage sub-units according to the timestamp label, perform hash evidence storage, generate evidence storage hash chain, perform cross-node cross-verification, and construct evidence storage topology structure.

[0222] In step 702, the joint feature vector group of abnormal states refers to a set of feature values ​​extracted from the SMS link that represent different abnormal states. The timestamp label is the time identifier of each feature vector, used to determine its position in the time series. The parallel evidence storage subunit is a small data block formed by dividing the feature vectors according to their timestamps, facilitating distributed storage and verification. Hash evidence storage uses a hash algorithm to encrypt the data, ensuring data integrity and immutability. The evidence storage hash chain is a chain composed of a series of interconnected hash values, each hash value calculated based on the previous hash value. The evidence storage topology is a network structure formed through cross-node cross-verification, used to enhance data security and reliability.

[0223] In this embodiment, firstly, the joint feature vector group of abnormal states is divided into multiple parallel evidence storage sub-units based on timestamp labels. Next, a hash algorithm is applied to each sub-unit to generate a corresponding hash value, and this hash value is used to construct an evidence storage hash chain. Then, cross-node cross-verification is performed to ensure that the data on each node is consistent and has not been tampered with. Finally, based on the verification results, a distributed evidence storage topology is constructed. This structure not only enhances data security but also improves the system's fault tolerance.

[0224] Specifically, firstly, the system divides the joint feature vector group into fixed-size evidence storage sub-units (each unit containing 30 seconds of data) based on timestamps, and organizes the data blocks using a Merkle tree structure. Secondly, a double hash operation (SHA-256+SM3) is performed on each sub-unit to generate an evidence storage hash chain containing forward hash pointers. Then, cross-validation is performed among at least 5 nodes using a Byzantine fault tolerance mechanism, requiring consensus among at least 3 / 5 of the nodes to confirm the data's validity. Finally, a multi-level evidence storage topology is constructed based on the verification results, where the main chain nodes store the complete hash chain, and the edge nodes store local verification results, forming a hierarchical verification network.

[0225] 703. Extract the state code value of the SMS link state transition feature corresponding to each timestamp tag in the evidence storage topology, and perform fluctuation fitting on the state code value according to the state transition relationship in ascending order of timestamp to generate SMS link state fluctuation sequence.

[0226] In step 703, the state code value is a quantitative representation of the SMS link state transition characteristics, facilitating computer processing. The state transition relationship describes the pattern of SMS link state changes over time. Fluctuation fitting is a statistical method used to identify and predict data change trends. The SMS link state fluctuation sequence is a set of state code values ​​arranged in chronological order, reflecting the dynamic change process of the SMS link state.

[0227] In this embodiment, firstly, the state encoding value of the SMS link state transition feature corresponding to each timestamp tag is extracted from the evidence storage topology. Next, based on the increasing timestamp order, the state transition relationship is analyzed, and fluctuation fitting technology is used to process the state encoding value. Then, through fluctuation fitting, the trend of the state encoding value changing over time is identified, generating an SMS link state fluctuation sequence. Finally, all information is integrated to form a complete SMS link state fluctuation sequence, providing basic data for subsequent analysis.

[0228] Specifically, first, the state code values ​​(8-bit binary codes, each representing a specific state flag) corresponding to each timestamp are extracted from the evidence storage topology. Second, a hidden Markov model is used to analyze the state transition probabilities and establish a time dependency matrix. Then, missing codes are repaired using cubic spline interpolation, and observation noise is eliminated using a Kalman filter. Finally, the processed code values ​​are arranged in time sequence, and sliding window statistics (mean, variance, range) are added to generate a standardized SMS link state fluctuation sequence with data intervals accurate to 100 milliseconds.

[0229] 704. Encapsulate and bind the SMS link state fluctuation sequence with the evidence storage hash chain to generate a blockchain state change event containing the SMS link state fluctuation sequence.

[0230] In step 704, event encapsulation and binding is the process of combining the SMS link state fluctuation sequence with the evidence storage hash chain to form a complete blockchain state change event. This event contains all relevant data and verification information, ensuring data integrity and traceability. The blockchain state change event is an immutable record used for long-term storage and verification.

[0231] In this embodiment, firstly, the generated SMS link state fluctuation sequence is matched and associated with the evidence storage hash chain. Next, blockchain technology is used to encapsulate these two data parts into an event, creating a new blockchain state change event. Then, a consensus mechanism ensures that this event is recognized and recorded by all nodes in the network. Finally, this event is added to the blockchain, forming a permanent record containing the SMS link state fluctuation sequence, ensuring data integrity and immutability.

[0232] Specifically, firstly, the state fluctuation sequence is logically bound to the evidence storage hash chain via a smart contract, constructing an event structure containing the following elements: ① Event header (block height, timestamp) ② Data body (compressed fluctuation sequence) ③ Verification information (Merkle root, digital signature) ④ Association pointer (previous event hash). Secondly, an improved DPoS consensus mechanism (delegated proof-of-stake) is used for network-wide broadcasting, requiring confirmation from at least 51% of the accounting nodes. Then, the confirmed events are written to the blockchain in chronological order, and sidechain technology is used to store a large volume of fluctuation sequence data. Finally, a receipt containing a unique event identifier is generated for subsequent query and verification.

[0233] Here is a specific example:

[0234] In a rescue scenario following a sudden earthquake, drones were rapidly deployed over the disaster area, establishing a temporary communication network. First, the system segmented the joint feature vector group of abnormal states into multiple parallel evidence storage sub-units based on timestamp labels. Then, these sub-units were hashed for evidence storage, generating an evidence storage hash chain. To ensure data authenticity and integrity, the system performed cross-node cross-validation to confirm data consistency across all nodes. Next, the system extracted the state code values ​​corresponding to each timestamp label from the evidence storage topology. Following the ascending order of timestamps, the system analyzed state transition relationships and processed the data using fluctuation fitting technology, generating a text message link state fluctuation sequence. This step helped the rescue team better understand the changing trends of the text message link state, enabling them to more effectively address communication bottlenecks and ensure the rapid transmission of emergency information. Finally, during earthquake rescue operations, the system encapsulated and bound the text message link state fluctuation sequence to the evidence storage hash chain, generating blockchain state change events. This process not only ensured data integrity and authenticity but also enabled all relevant parties to track and verify the data in real time. This approach greatly improves the transparency and efficiency of emergency response, ensures the reliable transmission of critical information, and lays a solid foundation for the successful implementation of rescue operations.

[0235] In summary, steps 701 to 704 achieve accurate monitoring and real-time feedback of abnormal association clusters and SMS link state transition characteristics. This method not only improves the efficiency of data collection and analysis but also enhances the system's adaptability and stability in emergency situations, providing strong technical support for responding to emergencies. The entire process, through techniques such as time alignment, hash notarization, and state fitting, constructs an efficient, flexible, and secure emergency communication support system. This solution ensures the timely transmission of critical information and guarantees the smooth progress of rescue operations. Ultimately, these steps work together to form a system capable of rapidly responding to changes and providing high-quality data support, greatly improving the reliability and efficiency of emergency communication.

[0236] Figure 2 This application provides a schematic diagram of the structure of an intelligent SMS link high-efficiency detection method system, as shown in the embodiment. Figure 2 As shown, the device includes:

[0237] The acquisition module 21 forms a dynamic communication topology by deploying a cluster of drones to collect real-time data on the latency mutation rate and packet loss rate of damaged SMS links, and generates an SMS link status fluctuation sequence. When the SMS transmission latency gradient in the SMS link status fluctuation sequence exceeds the emergency communication threshold, an SMS link anomaly marker is triggered.

[0238] The identification module 22 performs SMS link state migration detection on the dynamic communication topology based on the SMS link state fluctuation sequence, identifies abnormal association clusters of the transmission path, and forms a spatiotemporal coupling relationship with the SMS link anomaly marker.

[0239] The generation module 23 binds the abnormal association cluster and SMS link state migration feature to timestamps for multi-node parallel storage, and generates a blockchain state change event containing the SMS link state fluctuation sequence.

[0240] The update module 24 adjusts the distribution density of drones based on the verification results of the blockchain state change event, generates a text message link topology reconstruction factor, and forms a closed-loop feedback with the spatiotemporal coupling relationship to update the topology association detection path of text message transmission.

[0241] Figure 2 The intelligent SMS link high-efficiency detection method system described above can execute Figure 1 The implementation principle and technical effects of the intelligent SMS link high-efficiency detection method described in the illustrated embodiment will not be repeated here. The specific methods by which each module and unit of the intelligent SMS link high-efficiency detection system in the above embodiments perform operations have been described in detail in the embodiments related to this method, and will not be elaborated upon here.

[0242] In one possible design, Figure 2 The intelligent SMS link high-efficiency detection system of the embodiment shown can be implemented as a computing device, such as... Figure 3 As shown, the computing device may include a storage component 31 and a processing component 32;

[0243] The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are invoked and executed by the processing component 32.

[0244] The processing component 32 is used for the above Figure 1 The embodiment describes an intelligent and efficient method for detecting SMS links.

[0245] The processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above-described method. Alternatively, the processing component may be implemented as one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above-described method.

[0246] Storage component 31 is configured to store various types of data to support operations at the terminal. The storage component can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0247] Of course, computing devices may also include other components, such as input / output interfaces, display components, communication components, etc.

[0248] Input / output interfaces provide interfaces between processing components and peripheral interface modules, which can be output devices, input devices, etc.

[0249] The communication components are configured to facilitate wired or wireless communication between computing devices and other devices.

[0250] The computing device can be a physical device or an elastic computing host provided by a cloud computing platform. In this case, the computing device can refer to a cloud server, and the aforementioned processing components, storage components, etc., can be basic server resources rented or purchased from the cloud computing platform.

[0251] This application also provides a computer storage medium storing a computer program, which, when executed by a computer, can perform the above-described functions. Figure 1 The embodiment shown is an intelligent and efficient method for detecting SMS links.

[0252] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0253] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0254] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0255] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for intelligent and efficient detection of SMS links, characterized in that, include: By deploying a drone swarm to form a dynamic communication topology, the latency mutation rate and packet loss rate data of the damaged SMS link are collected in real time, and an SMS link status fluctuation sequence is generated. When the SMS transmission latency gradient in the SMS link status fluctuation sequence exceeds the emergency communication threshold, an SMS link anomaly marker is triggered. Based on the SMS link state fluctuation sequence, SMS link state migration detection is performed on the dynamic communication topology to identify abnormal association clusters of the transmission path and form a spatiotemporal coupling relationship with the SMS link anomaly marker. The abnormal association cluster and SMS link state migration feature are bound to timestamps for multi-node parallel storage, generating a blockchain state change event containing the SMS link state fluctuation sequence. The distribution density of drones is adjusted based on the verification results of the blockchain state change event, a text message link topology reconstruction factor is generated, and a closed-loop feedback is formed with the spatiotemporal coupling relationship to update the topology association detection path of text message transmission. The step of performing SMS link state transition detection on the dynamic communication topology based on the SMS link state fluctuation sequence, identifying abnormal association clusters in the transmission path, and forming a spatiotemporal coupling relationship with the SMS link anomaly marker includes: The node aggregation path is generated based on the SMS link state fluctuation sequence. The aggregation weight is dynamically adjusted according to the delay gradient and packet loss rate of adjacent nodes, and a positive feedback association is formed with the triggering conditions of the SMS link anomaly marker. SMS link state migration detection is performed on the dynamic communication topology, and abnormal association clusters of transmission paths are identified in the distribution characteristics of the node aggregation path and the SMS link state fluctuation sequence. The spatiotemporal coupling relationship of the abnormal association clusters is quantified into spatiotemporal coupling coefficients, and the spatiotemporal coupling coefficients form a cross-validation relationship with the aggregation weights of the node aggregation paths; Based on the cross-validation relationship, the spatial distribution boundary of the abnormal association cluster is corrected, abnormal association cluster correction parameters are generated, and the dynamic spatial association analysis process of the SMS link state fluctuation sequence is fed back through the node aggregation path.

2. The method according to claim 1, characterized in that, The step of quantifying the spatiotemporal coupling relationship of the abnormal association cluster into a spatiotemporal coupling coefficient, wherein the spatiotemporal coupling coefficient forms a cross-validation relationship with the aggregation weight of the node aggregation path, includes: Based on the three-dimensional coordinate distribution of UAV nodes in the emergency area, a coverage density index is generated by dividing the area into geographic grid units and calculating the ratio of the number of active nodes to the grid volume. The trigger frequency distribution of SMS link anomaly markers is statistically analyzed according to a preset time window, an anomaly frequency histogram is generated, and the duration of the peak interval in the anomaly frequency histogram is extracted as the anomaly time clustering factor. The coverage density index and the abnormal time clustering factor are weighted and fused to generate an initial spatiotemporal coupling coefficient. At the same time, the weight of the coverage density index is dynamically adjusted according to the priority of the emergency area. Extract the weight distribution entropy value of the central region of the abnormal association cluster, and quantify the spatiotemporal coupling relationship into a spatiotemporal coupling coefficient based on the difference between the weight distribution entropy value and the initial spatiotemporal coupling coefficient. An aggregated weight-spatiotemporal coupling cross-validation matrix is ​​constructed. A validation factor is generated by comparing the correlation between the gradient change rate of the aggregated weight and the update magnitude of the spatiotemporal coupling coefficient element by element. When the validation factor is lower than the confidence threshold, the spatial distribution boundary reconstruction of the abnormal association cluster is triggered.

3. The method according to claim 1, characterized in that, The step of triggering an SMS link anomaly flag when the SMS transmission delay gradient in the SMS link state fluctuation sequence exceeds the emergency communication threshold includes: Based on the difference between adjacent time windows in the SMS link state fluctuation sequence, the SMS transmission delay gradient is analyzed, and the normalized SMS transmission delay gradient is generated by normalizing the cumulative mean of the delay change rate within the sliding window. Based on the historical distribution characteristics of the normalized SMS transmission delay gradient and the gradient change trend within the current window, the emergency communication threshold is dynamically adjusted. When the normalized SMS transmission delay gradient exceeds the adjusted emergency communication threshold, an anomaly marker trigger condition is generated by combining the duration of the time exceeding the adjusted emergency communication threshold with the number of times it exceeds the threshold. The parameters corresponding to the normalized SMS transmission delay gradient, the adjusted emergency communication threshold, and the anomaly marker triggering condition are passed as associated parameters to the SMS link anomaly analysis module to trigger the SMS link anomaly marker.

4. The method according to claim 1, characterized in that, The method involves deploying a drone swarm to form a dynamic communication topology, thereby collecting real-time data on the latency mutation rate and packet loss rate of damaged SMS links and generating an SMS link state fluctuation sequence, including: Based on the spatial distribution of the damaged SMS links and the current channel quality, the signal strength and interference level of each node in the UAV swarm are controlled to generate dynamic weight parameters, and a dynamic communication topology is formed through iterative updates of the UAV swarm position. In the dynamic communication topology, the latency mutation rate and packet loss rate data of the damaged SMS link are collected at a preset period, and the packet loss rate data is supplemented with the packet loss rate variance within the collection time window as a fluctuation feature. The latency mutation rate and packet loss rate data are timestamped and stitched together using a sliding window to generate a time-aligned collection dataset. The collection dataset is then weighted and fused based on the dynamic weight parameters to generate a SMS link status fluctuation sequence.

5. The method according to claim 4, characterized in that, The process of timestamping and concatenating the latency mutation rate and packet loss rate data using a sliding window to generate a time-aligned collection dataset includes: A reference time axis is determined based on the clock synchronization protocol of each node in the drone cluster. The timestamps of the collected latency mutation rate and packet loss rate data are mapped to the reference time axis to generate a timestamp mapping table. Based on the timestamp mapping table, the latency mutation rate and packet loss rate data with missing timestamps are filled by linear interpolation. The mean of adjacent valid data points is used as the interpolation reference value to generate the interpolated latency mutation rate and packet loss rate data. The interpolated latency mutation rate and packet loss rate data are divided into multiple overlapping sliding windows. First and last timestamp calibration and boundary data smoothing are performed to generate latency mutation rate segments and packet loss rate segments aligned within the windows. The latency mutation rate segment and the packet loss rate segment are concatenated in chronological order to form a unified latency and packet loss data block with a unified time dimension. The time coverage and data density are extracted as window metadata to generate a time-aligned collection dataset.

6. The method according to claim 1, characterized in that, The step of binding the abnormal association cluster and SMS link state transition features to timestamps for multi-node parallel notarization, generating a blockchain state change event containing the SMS link state fluctuation sequence, includes: The abnormal feature vector of the abnormal association cluster and the dynamic feature vector of the SMS link state transition feature are timestamped through a time window sliding alignment mechanism to generate an abnormal state joint feature vector group. The abnormal state joint feature vector group is divided into multiple parallel evidence storage sub-units according to the timestamp label, and hash evidence storage is performed to generate an evidence storage hash chain, cross-node cross-verification is performed, and an evidence storage topology structure is constructed. In the evidence storage topology, extract the state encoding value of the SMS link state transition feature corresponding to each timestamp tag, and perform fluctuation fitting on the state encoding value according to the state transition relationship in ascending order of timestamp to generate SMS link state fluctuation sequence. The SMS link state fluctuation sequence is encapsulated and bound to the evidence storage hash chain to generate a blockchain state change event containing the SMS link state fluctuation sequence.

7. An intelligent SMS link high-efficiency detection system, characterized in that, include: The data acquisition module forms a dynamic communication topology by deploying a cluster of drones to collect real-time data on the latency mutation rate and packet loss rate of damaged SMS links, and generates an SMS link status fluctuation sequence. When the SMS transmission latency gradient in the SMS link status fluctuation sequence exceeds the emergency communication threshold, an SMS link anomaly marker is triggered. The identification module performs SMS link state migration detection on the dynamic communication topology based on the SMS link state fluctuation sequence, identifies abnormal association clusters of the transmission path, and forms a spatiotemporal coupling relationship with the SMS link anomaly marker. The generation module binds the abnormal association cluster and SMS link state migration feature to timestamps for multi-node parallel storage, generating a blockchain state change event containing the SMS link state fluctuation sequence. The update module adjusts the distribution density of drones based on the verification results of the blockchain state change event, generates a text message link topology reconstruction factor, and forms a closed-loop feedback with the spatiotemporal coupling relationship to update the topology association detection path of text message transmission. The step of performing SMS link state transition detection on the dynamic communication topology based on the SMS link state fluctuation sequence, identifying abnormal association clusters in the transmission path, and forming a spatiotemporal coupling relationship with the SMS link anomaly marker includes: The node aggregation path is generated based on the SMS link state fluctuation sequence. The aggregation weight is dynamically adjusted according to the delay gradient and packet loss rate of adjacent nodes, and a positive feedback association is formed with the triggering conditions of the SMS link anomaly marker. SMS link state migration detection is performed on the dynamic communication topology, and abnormal association clusters of transmission paths are identified in the distribution characteristics of the node aggregation path and the SMS link state fluctuation sequence. The spatiotemporal coupling relationship of the abnormal association clusters is quantified into spatiotemporal coupling coefficients, and the spatiotemporal coupling coefficients form a cross-validation relationship with the aggregation weights of the node aggregation paths; Based on the cross-validation relationship, the spatial distribution boundary of the abnormal association cluster is corrected, abnormal association cluster correction parameters are generated, and the dynamic spatial association analysis process of the SMS link state fluctuation sequence is fed back through the node aggregation path.

8. A computing device, characterized in that, It includes a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are invoked and executed by the processing component to implement the intelligent SMS link high-efficiency detection method as described in any one of claims 1 to 6.

9. A computer storage medium, characterized in that, The device contains a computer program, which, when executed by a computer, implements an intelligent SMS link high-efficiency detection method as described in any one of claims 1 to 6.

Images