A complex electromagnetic environment dynamic construction and deduction system based on digital twinning

Abstract

The application provides a complex electromagnetic environment dynamic construction and deduction system based on digital twinning, relates to the technical field of data processing, and comprises a digital twinning bottom model construction module, a state time sequence modeling module, a trigger interval determination module, an influence subgraph generation module, a local evolution update module and an overall fusion generation module; is used for establishing a digital twinning electromagnetic environment model, receiving equipment state data flow and forming a state change record, determining a target trigger interval, constructing an influence subgraph, locally updating a signal coupling relationship and a frequency spectrum occupation state of an affected equipment node, and fusing with an unchanged area result to generate an overall electromagnetic environment deduction result at a current time, thereby improving the accuracy of the system.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a system for dynamically constructing and extrapolating complex electromagnetic environments based on digital twins. Background Technology

[0002] In existing technologies, the construction of complex electromagnetic environments is usually achieved by using a "rule base, simulation platform and offline data update" approach: First, a scenario model is established in the electromagnetic simulation system based on geographic information, equipment parameters, spectrum configuration and propagation model; then, communication base stations, wireless terminals and various transmission sources are configured through preset scripts to complete the electromagnetic situation analysis and inference within a specific area and time range, and generate results such as coverage strength, signal interference level and link quality.

[0003] However, in scenarios such as intelligent transportation in large cities, when roadside communication devices, vehicle-mounted terminals, and temporary signal sources (such as mobile hotspots or emergency communication devices) frequently connect or their locations dynamically change, existing technologies typically require manual updates of device parameters and rerunning of the simulation process, making it difficult to reflect environmental changes in a timely manner. In practical applications, if newly added mobile hotspots occupy part of the frequency band during peak hours, and the system still extrapolates based on the spectrum data before the update, the platform may determine that the quality of the vehicle-to-everything (V2X) communication link at a certain intersection is normal, but the actual terminal experiences data packet loss or increased latency, thus affecting the accuracy of traffic signal coordinated control. Summary of the Invention

[0004] The purpose of this invention is to provide a system for dynamically constructing and extrapolating complex electromagnetic environments based on digital twins, aiming to solve the problems mentioned in the background art.

[0005] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows: A system for dynamically constructing and extrapolating complex electromagnetic environments based on digital twins, the system comprising: The digital twin base model construction module is used to associate and organize geographic information, equipment node information and spectrum resource information in the physical space, establish a digital twin electromagnetic environment model that maps to the physical space, and configure a unique identifier and corresponding historical operating status sequence for each equipment node in the digital twin electromagnetic environment model. The state timing modeling module is used to continuously receive state data streams reported by communication devices based on unique identifiers, accumulate historical operating state sequences by time stamping, and generate state change records based on the state change amplitude between adjacent time slices. The trigger interval determination module is used to divide the environmental change trigger interval according to the correspondence between the state change record and the preset change determination condition, and generate the target trigger interval that meets the preset change determination condition. The influence subgraph generation module is used to extract the device nodes whose state changes from the digital twin electromagnetic environment model according to the target triggering interval, and combine the signal propagation correlation and spectrum occupancy conflict relationship between the device nodes to construct an influence subgraph consisting of the set of affected device nodes and their associated spectrum resource subsets. The local evolution update module is used to update the signal coupling relationship and spectrum occupancy status between the affected device node sets within the scope of the influence subgraph, generate local update results, and write the local update results back to the digital twin electromagnetic environment model. The overall fusion generation module is used to retain the previous time-instance results of the unaffected subgraphs in the digital twin electromagnetic environment model, and to stitch and fuse the local update results with the previous time-instance results to generate the overall electromagnetic environment simulation results for the current time-instance.

[0006] The above-described solution of the present invention has at least the following beneficial effects: First, by constructing a digital twin electromagnetic environment model that maps to the physical space and continuously receiving equipment status data to form a status change record, the construction process of the electromagnetic environment is transformed from relying on preset rules and offline data updates to a dynamic expression driven by real-time data. This allows the system to reflect the real-time changes in equipment operating status, spatial location, and spectrum occupancy. On this basis, by triggering interval determination, processing is only performed on time intervals where changes occur, giving the identification and processing of environmental changes a clear temporal basis and avoiding the repeated execution of the overall simulation process for unchanged periods.

[0007] Furthermore, by extracting the changed device nodes based on the target trigger interval and constructing an influence subgraph, the update scope of the electromagnetic environment is limited to the nodes and spectrum resources related to the current change, thereby reducing the repeated calculation of the unchanged area and maintaining the continuity of the results in the unchanged area. On this basis, by performing local evolution updates with the influence subgraph as the processing scope, the signal coupling relationship and spectrum occupancy status can be updated around the changed nodes, so that the inference results can reflect the impact of device access or frequency band changes in a timely manner.

[0008] Furthermore, by fusing the local update results with the previous time-to-time results of the unaffected subgraph coverage area, the latest changes can be introduced while preserving the continuity of historical states. This generates the overall electromagnetic environment projection results corresponding to the current time, ensuring that the electromagnetic environment construction process is continuously expressed in the time dimension and fully covered in the spatial dimension. Based on this, in scenarios where devices frequently connect or spectrum occupancy changes, the link status judgment bias caused by using historical spectrum data can be avoided, ensuring that the projection results are consistent with actual environmental changes. This supports the continuous perception and analysis of electromagnetic environment changes by relevant applications. Attached Figure Description

[0009] Figure 1 This is an architecture diagram of a system for dynamically constructing and extrapolating complex electromagnetic environments based on digital twins, provided by an embodiment of the present invention. Detailed Implementation

[0010] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0011] like Figure 1 As shown, embodiments of the present invention propose a system for dynamically constructing and extrapolating complex electromagnetic environments based on digital twins. The system includes: The digital twin base model construction module is used to associate and organize geographic information, equipment node information and spectrum resource information in the physical space, establish a digital twin electromagnetic environment model that maps to the physical space, and configure a unique identifier and corresponding historical operating status sequence for each equipment node in the digital twin electromagnetic environment model. The state timing modeling module is used to continuously receive state data streams reported by communication devices based on unique identifiers, accumulate historical operating state sequences by time stamping, and generate state change records based on the state change amplitude between adjacent time slices. The trigger interval determination module is used to divide the environmental change trigger interval according to the correspondence between the state change record and the preset change determination condition, and generate the target trigger interval that meets the preset change determination condition. The influence subgraph generation module is used to extract the device nodes whose state changes from the digital twin electromagnetic environment model according to the target triggering interval, and combine the signal propagation correlation and spectrum occupancy conflict relationship between the device nodes to construct an influence subgraph consisting of the set of affected device nodes and their associated spectrum resource subsets. The local evolution update module is used to update the signal coupling relationship and spectrum occupancy status between the affected device node sets within the scope of the influence subgraph, generate local update results, and write the local update results back to the digital twin electromagnetic environment model. The overall fusion generation module is used to retain the previous time-instance results of the unaffected subgraphs in the digital twin electromagnetic environment model, and to stitch and fuse the local update results with the previous time-instance results to generate the overall electromagnetic environment simulation results for the current time-instance.

[0012] In this embodiment of the invention, by constructing a digital twin electromagnetic environment model that maps to physical space, and by associating and organizing device node information, spectrum resource information, and historical operating state sequences, the model can continuously carry dynamic changes in device status, spectrum occupancy, and spatial location within a unified framework. This provides a continuous and consistent data foundation for subsequent simulations of complex electromagnetic environments. By receiving state data streams based on unique identifiers and forming state change records, discrete reported data can be transformed into time-evolving environmental change data. Furthermore, by combining the determination results of environmental change trigger intervals, the periods of change can be specifically identified.

[0013] By extracting the device nodes whose states change based on the target trigger interval, and constructing an influence subgraph by combining the signal propagation correlation and spectrum occupancy conflict relationships between the device nodes, dynamic changes can be confined to the set of nodes and spectrum resources related to the current change, allowing subsequent inference processing to revolve around the affected parts. Compared to updating the entire digital twin electromagnetic environment model as a whole, this approach reduces the repeated participation of unchanged areas in the state update process and maintains the continuity of results in unchanged areas.

[0014] By processing the signal coupling relationships and spectrum occupancy status among the affected device nodes using the affected subgraph as the processing scope, and writing the local update results back to the digital twin electromagnetic environment model, the local changes can be reflected in the overall model in a timely manner. Combined with the overall fusion generation process, the previous time-instance results are retained for areas not covered by the affected subgraph, and the local update results are spliced ​​and fused with the previous time-instance results. This allows for the generation of the overall electromagnetic environment projection results corresponding to the current time-instance after environmental changes occur, thus balancing the response to local changes with the continuous representation of the overall environment.

[0015] The following explanation is based on specific scenarios: Within large transportation hubs, the operating status, spatial location, and spectrum occupancy of fixed communication equipment, mobile terminal equipment, and temporary access equipment change over time. When some temporary access equipment enters the station area and occupies existing frequency bands, the system first generates a status change record based on the status data stream, then determines the corresponding target trigger interval. Subsequently, an influence subgraph is constructed around the changed equipment nodes, and the signal coupling relationship and spectrum occupancy status between relevant equipment nodes are locally updated. The updated results are then fused with the results of the previous moment for the unchanged areas to obtain the overall electromagnetic environment projection results for the current moment. This facilitates the continuous construction and projection of the electromagnetic environment change process within the area.

[0016] In a preferred embodiment of the present invention, the preset change determination condition is used to determine whether the changes in the state change record constitute valid environmental changes, including: Corresponding change judgment rules are established for equipment operating status, spectrum occupancy status and spatial location status respectively. Among them, the equipment operating status change judgment rule is used to identify changes in equipment start / stop status, working mode switching or service type changes, and the judgment is made by counting the number of status items that have changed and the type of change. The spectrum occupancy status change judgment rule is used to identify changes in the start position of the frequency band, changes in the end position of the frequency band, and changes in the number of occupied intervals. It determines whether the set judgment criteria are met by comparing the amount of change in the frequency band range between the current time slice and the previous time slice. The spatial position change determination rule is used to identify the positional offset of a device node between two time slices, and the determination is made by calculating the change in distance and direction between the current position and the previous position. Furthermore, the above three types of change judgment results are combined according to the device node dimension. When any dimension change meets the set judgment criteria, or multiple dimension changes occur simultaneously and meet the combined judgment requirements, the corresponding time slice is marked as a valid change time slice, thus forming the basis for subsequent trigger interval judgment.

[0017] In a preferred embodiment of the present invention, the digital twin base model construction module includes: Acquire geographic information of the target area, including at least road distribution information, building distribution information, area boundary information, and equipment installation location coordinates. Align geographic information from different sources according to a unified spatial coordinate benchmark to form a base spatial map of the target area. Obtain device node information deployed in the target area. The device node information includes at least the device category, device installation location, device transmission power, device receiving sensitivity, device operating frequency band, device communication direction, and device access relationship. Then, load each device node into the basic spatial base map according to the device category and spatial location to form a device node distribution layer. Acquire available spectrum resource information within the target area. The spectrum resource information includes at least the frequency band range, frequency band purpose, occupancy status, corresponding device node and time stamp information. Then, attach each spectrum resource information to the device node distribution layer according to the device node-frequency band correspondence to form a spectrum resource association layer. Based on the correspondence between the basic spatial base map, the device node distribution layer, and the spectrum resource association layer, geographic objects, device nodes, and spectrum resources in the same spatial location are uniformly organized to establish a digital twin base model that includes spatial location relationships, node connection relationships, and spectrum occupancy relationships. Each device node in the digital twin is assigned a unique identifier, which includes at least a node category identifier, a region identifier, and a sequence identifier, to ensure that the status data of the same device node in different time slices can be continuously tracked. Based on the initial operating data of each device node, a corresponding historical operating status sequence is established. The historical operating status sequence includes at least the device operating status, spectrum occupancy status, spatial location status, and time stamp information. A one-to-one correspondence is established between the historical operating status sequence and a unique identifier to obtain a digital twin electromagnetic environment model that can be used for subsequent dynamic updates. A consistency check is performed on the established digital twin electromagnetic environment model. The consistency check includes at least whether there are missing spatial locations, whether there are unrelated device nodes and spectrum resources, and whether there are duplicate unique identifiers. After the check is completed, the model is written to the storage medium for subsequent state timing modeling module to call.

[0018] In a preferred embodiment of the present invention, the state timing modeling module includes: The status data stream reported by each device node is continuously received according to the preset sampling period. The status data stream includes at least device operating status data, spectrum occupancy data, spatial location data and corresponding acquisition time information. The device identification information in the state data stream is matched with the unique identifier in the digital twin electromagnetic environment model. Each newly arrived state data is written to the end of the historical operating state sequence of the corresponding device node and the current time slice mark is attached to form a time sequence state set arranged in chronological order. The historical operating status sequence is standardized, with equipment operating status recorded using a unified status category, spectrum occupancy status recorded using a unified frequency band division, and spatial location status recorded using a unified coordinate expression method, to ensure the comparability of data from different time slices. Extract the current time slice data and the previous time slice data from the historical operating status sequence of the same device node, and compare whether the device operating status category has changed, whether the spectrum occupancy range has changed, and whether the spatial location has migrated, to obtain the state change content of the device node between adjacent time slices. The magnitude of the state change is determined based on the state change content. The magnitude of the change in the device operating state is determined by comparing the number of state category changes and the differences between the categories. The magnitude of the change in the spectrum occupancy state is determined by comparing the change in the start and end range of the frequency band and the change in the occupied interval. The magnitude of the change in the spatial position is determined by comparing the position offset distance between two time slices. The changes in device operating status, spectrum occupancy status, and spatial location of each device node between adjacent time slices are merged and associated with the corresponding time slice markers to form a status change record. The status change records of all device nodes are summarized in chronological order to form a global status change record set that can be called by the trigger interval determination module.

[0019] In a preferred embodiment of the present invention, the trigger interval determination module includes: The change sequence generation unit is used to collect the change content of each device node in each time slice based on the differences in device operation status, spectrum occupancy status and spatial location information of each device node in the state change record between adjacent time slices, and generate the corresponding multidimensional change quantity sequence. The composite identifier determination unit is used to match the changes in device operating status, spectrum occupancy status, and spatial location in each time slice with the corresponding preset determination conditions based on the multidimensional change sequence, and to combine and associate the matching results to generate composite change identifiers for each time slice. The trigger interval construction unit is used to perform continuous connection processing on each time slice according to the composite change identifier and in chronological order, and to combine the time slices with continuous composite change identifiers to construct an environmental change trigger interval. The target trigger interval generation unit is used to filter the constructed environmental change trigger intervals based on the duration and intensity of the change corresponding to the environmental change trigger interval, and generate the target trigger interval.

[0020] In this embodiment of the invention, by aggregating the device operating status, spectrum occupancy status, and spatial location information of device nodes between adjacent time slices, a multidimensional change sequence is formed. Furthermore, by using composite identifiers to determine and combine different types of changes, the original state change information can be transformed into change identifiers with a unified expression. Based on this composite change identifier, continuous connection processing can identify intervals with continuous change characteristics in the time dimension, thereby avoiding repeated triggering of isolated changes or short-term fluctuations. Further, by filtering the duration and intensity of environmental change trigger intervals, target trigger intervals for subsequent processing can be obtained, ensuring that the subsequent construction of the influence subgraph and the local evolution update process are based on a clear change period, thus guaranteeing consistency between the triggering conditions and the environmental change process.

[0021] In a preferred embodiment of the present invention, the change sequence generation unit includes: Read the adjacent time slice change data corresponding to each device node in the status change record, and extract the change content according to three dimensions: device operating status difference, spectrum occupancy status difference, and spatial location information difference. For differences in equipment operating status, count the number of operating status items that have changed in the current time slice relative to the previous time slice, record the correspondence between the state before the change and the state after the change, and generate the change amount of equipment operating status. For the differences in spectrum occupancy status, extract the changes in the start position, end position, and number of occupied intervals between the current time slice and the previous time slice, and organize the changes according to the unified frequency band scale to generate spectrum occupancy status change data; For differences in spatial location information, extract the coordinate position data of the current time slice and the previous time slice, calculate the offset distance and offset direction between the current position and the previous position, and generate the spatial location change. The changes in the device's operating status, spectrum occupancy status, and spatial location corresponding to the same device node in the same time slice are collected in the order of the time slices to form a single-node multidimensional change record for that device node. The single-node multidimensional change records of all device nodes are further summarized according to time slices, so that each time slice corresponds to a time slice change set containing the change content of multiple device nodes. The changes in each time slice are arranged in chronological order, and each time slice change set is assigned a time slice number and a unique device node identifier. Finally, a multidimensional change sequence is generated that can be called by the composite identifier determination unit.

[0022] In a preferred embodiment of the present invention, the composite identifier determination unit includes: Read the multidimensional change sequence output by the change sequence generation unit, and extract the device operation status change, spectrum occupancy status change, and spatial location change of each device node in each time slice according to the time slice order; Establish separate rules for determining changes in equipment operating status, spectrum occupancy status, and spatial location. The rules for determining changes in equipment operating status are used to determine whether the equipment has switched operating categories, started / stopped states, or switched working modes. The rules for determining changes in spectrum occupancy status are used to determine whether changes in the start position of the frequency band, the end position of the frequency band, and the number of occupied intervals have reached preset judgment conditions. The rules for determining changes in spatial location are used to determine whether changes in position offset distance and offset direction have reached preset judgment conditions. The changes in the device operating status of each device node within each time slice are compared item by item with the rules for determining changes in device operating status, and the matching results of device operating status changes are generated based on the comparison results. The spectrum occupancy status change of each device node in each time slice is compared with the spectrum occupancy status change judgment rules item by item, and the spectrum occupancy status change matching result is generated based on the comparison results. The spatial position change of each device node within each time slice is compared with the spatial position change judgment rules item by item, and the spatial position change matching result is generated based on the comparison results. Based on the correspondence of the same device node and the same time slice, the matching results of device operation status change, spectrum occupancy status change, and spatial location change are merged, and the merged results are combined and associated to identify single-dimensional change, two-dimensional change, and three-dimensional change. Based on the combined association results, a composite change identifier is generated for each device node in the corresponding time slice. The composite change identifier includes at least the time slice number, the unique identifier of the device node, the change dimension category, and the change dimension combination type. The composite change identifiers corresponding to all device nodes within the same time slice are summarized to form a set of composite change identifiers for that time slice, and then output for the trigger interval construction unit to call.

[0023] In a preferred embodiment of the present invention, the trigger interval construction unit includes: Read the set of composite change identifiers for each time slice output by the composite identifier determination unit, and sort them from front to back according to the time slice number; Starting from the initial time slice, check in turn whether there are composite change identifiers that connect with the current time slice in subsequent time slices. The connection relationship includes at least the continuous changes of the same device nodes between adjacent time slices, the continuous occurrence of the same change dimension combination type, or the existence of related changes that are passed from the change result of the previous time slice to the next time slice. When adjacent time slices satisfy the connection relationship, the corresponding time slices are connected continuously, and the result of the continuous connection is recorded as the same candidate trigger interval; When the connection relationship between adjacent time slices is not met, the extension of the current candidate trigger interval is terminated, and the subsequent time slice that meets the conditions is used as the new starting time slice to continue the continuous connection processing. For each candidate trigger interval, record its start time slice, end time slice, number of device nodes covered, and number of composite change identifier types to form candidate trigger interval description information; All candidate trigger intervals are checked, and intervals that contain only a single isolated time slice and do not meet the continuity determination requirements are removed, while candidate trigger intervals that meet the continuity connection conditions are retained. The retained candidate trigger intervals are defined as environmental change trigger intervals and output for the target trigger interval generation unit to call.

[0024] In a preferred embodiment of the present invention, the target triggering interval generation unit includes: Read the environmental change trigger intervals output by the trigger interval construction unit, and extract the start time slice, end time slice, number of device nodes, number of composite change identifiers, and combination of change dimensions corresponding to each environmental change trigger interval; The duration of each environmental change trigger interval is determined based on the number of time slices covered between the start and end time slices. The intensity of change in each environmental change trigger interval is determined by merging the cumulative number of matching results for changes in device operating status, spectrum occupancy status, and spatial location within each environmental change trigger interval. The intensity of change is obtained by statistically analyzing the number of occurrences of various types of changes, the number of device nodes involved, and the number of combinations of composite change dimensions within the interval. The duration of each environmental change trigger interval is compared with the preset duration condition, and candidate intervals that meet the duration condition are selected. The change intensity of candidate intervals that meet the duration condition is compared with the preset intensity condition, and the intervals that meet the change intensity condition are selected for retention. The retained intervals are numbered in chronological order, and the corresponding device node range, change dimension combination and time span information are recorded to generate the target trigger interval. The generated target trigger interval is output to the influence subgraph generation module for subsequent initial node extraction unit to call.

[0025] In a preferred embodiment of the present invention, the influence subgraph generation module includes: The initial node extraction unit is used to extract the device nodes whose states have changed from the digital twin electromagnetic environment model according to the target triggering interval, and generate an initial node set. The propagation range extension unit is used to extend the propagation range node by node according to the initial node set, along the signal propagation path relationship between device nodes, and to constrain the propagation path depth and propagation influence range during the extension process, thereby generating an extended node set; The conflict chain generation unit is used to identify pairs of device nodes with overlapping spectrum occupancy based on the spectrum occupancy status of each device node in the extended node set, and extend along the conflict association path between the device node pairs to generate a set of conflict node chains. The influence subgraph construction unit is used to organize associated device nodes, spectrum resource subsets, and connection relationships between nodes based on the initial node set, the extended node set, and the conflict node chain set, and generate an influence subgraph with propagation path constraints and spectrum conflict chain structure.

[0026] In this embodiment of the invention, an initial node set is formed by extracting device nodes whose states change based on the target trigger interval. This set is then expanded node by node along the signal propagation path between the device nodes, allowing the change to propagate from the initial node to associated nodes, thus forming an extended node set related to the change. By combining the spectrum occupancy status of the device nodes to identify node pairs with overlapping spectrum occupancy relationships and extending along the conflict association path, a conflict node chain set reflecting interference relationships can be further constructed. By unifying the initial node set, the extended node set, and the conflict node chain set, and establishing node connection relationships and spectrum association relationships, an influence subgraph with structured connection characteristics can be formed. This allows the propagation relationships and conflict relationships between device nodes to be expressed in the same structure, thereby providing a clear processing scope and relationship constraints for subsequent local evolution updates.

[0027] In a preferred embodiment of the present invention, the influence subgraph construction unit includes: Read the initial node set output by the initial node extraction unit, the extended node set output by the propagation range extension unit, and the conflict node chain set output by the conflict chain generation unit, and summarize the three types of node data according to the unique identifier of the device node. Duplicate device nodes in the initial node set, extended node set, and conflict node chain set are merged, and the spatial location, device type, operating frequency band, connection relationship, and source of change of the corresponding device node are retained to form the total set of affected nodes; Based on the spectrum occupancy information of each device node in the total set of affected nodes, extract the spectrum resource records that have occupancy, conflict or allocation relationships with each device node, and form a spectrum resource subset according to the correspondence between device nodes and spectrum resources; Based on the nodes in the initial node set as source nodes, the nodes in the extended node set as propagation-related nodes, and the nodes in the conflict node chain set as conflict-related nodes, source node markers, propagation node markers, and conflict node markers are established respectively, and the corresponding markers are written into the total set of affected nodes. Based on the existing signal propagation path relationships between device nodes, connect each node in the total set of affected nodes to form propagation connection edges. According to the node order in the conflict node chain set, nodes with overlapping spectrum occupancy or interference transmission relationship are sequentially connected to form conflict connection edges. The propagation connection edges and conflict connection edges are organized in a unified manner, and each connection edge is given connection type information, associated frequency band information and node order information to form a set of node connection relationships. A graph structure data is constructed based on the total set of affected nodes, the subset of spectrum resources, and the set of node connection relationships. The graph structure data includes at least a node table, a spectrum resource table, a propagation connection relationship table, and a conflict connection relationship table. The completed graph structure data is subjected to consistency verification. The consistency verification includes at least whether all initial nodes are included in the graph structure, whether the extended nodes meet the propagation path constraints, whether the conflict node chain maintains the chain order, and whether the spectrum resource subset has a corresponding relationship with the nodes. After the verification is completed, an influence subgraph is generated and output to the local evolution update module for invocation.

[0028] In a preferred embodiment of the present invention, the local evolution update module includes: The propagation hierarchy generation unit is used to generate multiple propagation hierarchy with sequential dependencies by dividing the device nodes into layers according to the connection order and path depth of the propagation path, based on the signal propagation path relationship between each device node in the influence subgraph. The hierarchical recursive correction unit is used to perform state updates on each level of device nodes in the order from the starting level to the ending level according to the propagation level, and to perform transmission correction on the signal coupling relationship and spectrum occupancy conflict state of the current level of device nodes according to the update results of the previous level of device nodes, thereby generating a hierarchical update result sequence that is recursively applied layer by layer. The path result unification unit is used to perform consistency processing on the update results of the same device node under different propagation paths according to the hierarchical update result sequence, and generate a unified state result for each device node. The local result write-back unit is used to integrate the device node status after each level of update based on the unified status result, generate local update results, and write the local update results into the corresponding device node positions in the digital twin electromagnetic environment model.

[0029] In this embodiment of the invention, by dividing the data hierarchically according to the signal propagation path relationships between device nodes in the influence subgraph, a propagation hierarchy structure with sequential dependencies can be formed, allowing the update process of device nodes to unfold layer by layer according to the propagation path. By updating the state according to the propagation hierarchy and correcting the signal coupling relationship and spectrum occupancy conflict state of the current level device nodes based on the update results of the previous level, the signal influence can be gradually transmitted between nodes according to the propagation path, thereby ensuring that the update process is consistent with the actual propagation relationship. Furthermore, by uniformly processing the update results of the same device node under different propagation paths, the inconsistency problem caused by multi-path propagation can be avoided, and a unified state result can be formed. Finally, by writing the unified state result back to the digital twin electromagnetic environment model, a consistent expression of local update results in the model can be achieved, thereby supporting the subsequent overall fusion generation process.

[0030] In a preferred embodiment of the present invention, the path result unification unit includes: Read the hierarchical update result sequence output by the hierarchical recursive correction unit, and collect the corresponding node update results under different propagation levels and different propagation paths according to the unique identifier of the device node; Multiple update results corresponding to the same device node are classified and organized. The classification content includes at least the propagation path identifier, propagation level identifier, signal coupling relationship update content, spectrum occupancy conflict status update content, and result generation time order. Update the content based on the signal coupling relationship of the same device node under different propagation paths, identify update results with consistent content, and merge and record update results with consistent content. For update results with inconsistent content, their corresponding conflict intensity, propagation path sequence, and propagation level position are compared respectively. The conflict intensity is determined by comparing the number of frequency bands involved in the conflict, the number of conflict-related nodes, and the number of conflict-continuous records. The propagation path sequence is determined by comparing the generation order of the path in which the update result is located. The propagation level position is determined by comparing the distance between the level of the node corresponding to the update result and the initial node set. The inconsistent update results of the same device node are sorted in the following order: first, by comparing the intensity of the conflict; then, by comparing the order of the propagation paths; and finally, by comparing the position of the propagation level. The update result of the priority position after sorting is determined as the target update result, and its corresponding signal coupling relationship state and spectrum occupancy conflict state are taken as the unified state of the device node. The unified status result set is formed by summarizing all device nodes that have completed the unified processing. The unified status result set includes at least the unique identifier of the device node, the unified signal coupling relationship status, the unified spectrum occupancy conflict status, and the corresponding time slice mark. A completeness check is performed on the unified state result set. The completeness check includes at least whether all device nodes in the influence subgraph form a unified state result, whether each device node corresponds to only one unified state result, and whether the unified state result retains the source path information. After the check is completed, the unified state result corresponding to each device node is output for the time continuity constraint unit or the local result write-back unit to call.

[0031] In a preferred embodiment of the present invention, the local result write-back unit includes: Read the unified state result output by the path result unification unit, or when the local evolution update module contains a time continuity constraint unit, read the corrected unified state result transmitted by the time continuity constraint unit, and establish a list of nodes to be written back according to the unique identifier of the device node. Based on the list of nodes to be written back, locate the corresponding device node record in the digital twin electromagnetic environment model, and extract the original state data of the corresponding device node. The original state data includes at least the original signal coupling relationship state, the original spectrum occupancy state, the spatial location state, and the previous time slice marking information. The corrected unified state result or the signal coupling relationship state and spectrum occupancy conflict state in the unified state result are written to the corresponding device node record, and the unchanged device attribute information and basic spatial location information are retained. Append the current time slice flag to the already written device node record, and write the updated node status to the end of the historical running status sequence corresponding to the device node in chronological order to form the latest status record of the device node in the current time slice; After writing to all device nodes within the coverage area of ​​the affected subgraph, the latest state records of each device node in the current time slice are merged and organized to generate a local update result set. The local update result set includes at least the unique identifier of the device node, the updated signal coupling relationship status, the updated spectrum occupancy status, and the current time slice marker. The local update result set is updated synchronously with the node index relationship, spectrum resource relationship and spatial location relationship in the digital twin electromagnetic environment model to ensure that the node status, spectrum resource occupancy status and node connection relationship within the model remain consistent. After the write-back is completed, a write-back verification is performed on the digital twin electromagnetic environment model. The write-back verification includes at least whether all nodes to be written back have been completed, whether the current time slice mark is complete, whether the historical running state sequence is continuous in chronological order, and whether the set of local update results is consistent with the node state in the model. After the verification is completed, the local update results are output for the overall fusion generation module to call.

[0032] In a preferred embodiment of the present invention, the propagation range extension unit includes: The path depth control subunit is used to mark the node level on each propagation path level by level when expanding nodes along the signal propagation path, and terminate the expansion of the corresponding propagation path when the node level reaches the preset path depth, thereby generating a candidate node set with path depth limited. The influence accumulation judgment subunit is used to accumulate and record the propagation influence of each node along the propagation path based on the signal influence transmission of each node in the candidate node set, and to filter the nodes whose cumulative influence meets the preset influence judgment conditions to generate an influence constraint node set. The spectrum association filtering subunit is used to filter nodes with spectrum associations based on the correspondence between the spectrum occupancy information of each node in the set of influence constraint nodes and the spectrum occupancy information of the nodes in the initial node set, and generate a set of spectrum association nodes. The direction consistency constraint subunit is used to retain nodes with continuous and consistent propagation directions based on the signal propagation direction relationship between nodes in the spectrum-associated node set, and generate an extended node set.

[0033] In this embodiment of the invention, by marking the node levels of each propagation path level by level during the propagation path expansion process, and terminating the expansion of the corresponding path when a preset path depth is reached, the propagation range can be limited to a boundary, preventing the node expansion process from extending indefinitely. By accumulating and recording the signal influence received by each node in the propagation path path and filtering based on the accumulated influence, the expansion nodes can be constrained not only by topological connections but also by the influence transmission during the propagation process. Furthermore, by filtering based on the spectrum occupancy relationship between nodes, the expansion nodes can be associated with the initial nodes in the spectrum dimension, thereby preventing nodes in irrelevant frequency bands from entering the expansion range. Combined with the consistency constraint of the propagation direction, the node expansion can be made continuous along the signal propagation direction, thus forming an expansion node set under multiple constraints of path depth, influence accumulation, spectrum association, and propagation direction, giving the construction range of the influence subgraph clear boundary conditions.

[0034] In a preferred embodiment of the present invention, the preset path depth, used to limit the maximum expansion level of a node along the propagation path during the propagation range expansion process, includes: Starting with the initial set of nodes, its level is defined as the initial level, and the nodes are numbered level by level during the propagation path expansion process. Each time the number of nodes is expanded, the level increases by one unit. The maximum number of allowed layers can be preset according to the application scenario, for example, the maximum propagation layer can be determined based on the size of the area, device density or propagation attenuation characteristics; During the expansion process, when the level number of a certain node reaches the preset maximum level, further expansion starting from that node stops, thereby limiting the maximum length of the propagation path. This method can control the size of the expansion node set, ensuring that the propagation range is consistent with the actual signal influence range.

[0035] In a preferred embodiment of the present invention, the preset influence determination condition is used to screen whether the cumulative influence of nodes in the propagation path reaches an effective influence, including: The signal impact received by the node along the propagation path is recorded and accumulated level by level. The signal impact includes at least the degree of signal strength change, the number of impact duration levels, and the number of nodes involved in the impact. The signal strength variation is classified into different levels, such as low, medium and high, the number of sustained levels is counted and the number of involved nodes is statistically analyzed. By comprehensively evaluating the above-mentioned influencing factors, when the signal strength change reaches a set level and the number of sustained levels reaches a set number, or the number of nodes involved exceeds a set range, the node is determined to meet the influence judgment conditions; in this way, nodes with actual influence significance in the propagation process are screened out.

[0036] In a preferred embodiment of the present invention, the influence accumulation determination subunit includes: Read the candidate node set output by the path depth control subunit and extract the propagation path information and hierarchical information corresponding to each node; Along each propagation path, starting from the starting node, the process is traversed downwards level by level. The impact of the signal received at each node along the path is recorded. The impact of the signal includes at least the changes in signal strength, the propagation attenuation, and the impact of the connection relationship between nodes. The signal impact of each node on the path is accumulated level by level. The impact received by each node is combined with the cumulative impact of the previous node to form the cumulative impact record of that node. Based on preset impact judgment conditions, the cumulative impact records of each node are compared. The preset impact judgment conditions include at least the degree of impact reaching a set threshold, the duration of impact reaching a set range, and the number of nodes involved in the impact reaching a set number. Nodes whose cumulative impact records meet the preset impact judgment conditions are selected, and their corresponding path information and hierarchical information are retained to form an impact constraint node set; The set of influence constraint nodes is organized so that each node contains a unique identifier, propagation path, hierarchical information and cumulative influence record, and output for the spectrum association filtering subunit to call.

[0037] In a preferred embodiment of the present invention, the spectrum correlation screening subunit includes: Read the set of influence constraint nodes of the influence accumulation determination subunit output, and extract the spectrum occupancy information corresponding to each node; Read the spectrum occupancy information of each node in the initial node set and establish a correspondence table between the initial nodes and spectrum resources; Each node in the set of influencing constraint nodes is matched one by one with the nodes in the initial node set to compare whether their spectrum occupancy ranges overlap, are adjacent, or belong to the same frequency band category; for nodes with overlapping spectrum occupancy, they are directly determined as spectrum-related nodes. For nodes with frequency band adjacency, the determination is made based on whether the interval between frequency bands is less than a preset interval condition. Nodes that meet the condition are determined to be spectrum-related nodes. For nodes belonging to the same frequency band category, the determination is made based on the consistency of frequency band usage or service type, and nodes that meet the conditions are determined to be spectrum-related nodes; All nodes that meet the spectrum association conditions are aggregated, and their corresponding path information, hierarchical information, and cumulative impact records are retained to form a spectrum association node set; The set of spectrum-related nodes is deduplicated to ensure that each node retains only one record in the set, and the result is output for the direction consistency constraint subunit to call.

[0038] In a preferred embodiment of the present invention, the direction consistency constraint subunit includes: Read the set of spectrum-related nodes output by the spectrum association filtering subunit, and extract the propagation path information and propagation direction information between each node, wherein the propagation direction information includes at least the direction identifier of the signal from the source node to the target node; For each node in the propagation path, traverse the path in order and determine whether the propagation direction between adjacent nodes is consistent, that is, determine whether the signal transmission direction is continuously advancing in the same direction. For nodes whose propagation direction changes in the opposite direction or deviates from the main direction of the path, they are marked as nodes with inconsistent directions and removed from the current path; For nodes that maintain a consistent propagation direction, they are retained in the current path and continue to participate in the direction consistency judgment of subsequent nodes; After performing directional consistency filtering on all propagation paths, the nodes retained in each path are aggregated to form a set of nodes with consistent direction. The set of nodes with consistent orientation is integrated with the corresponding path information, hierarchical information and spectrum association information to generate the final extended node set, which is then output for the influence subgraph construction unit to call.

[0039] In a preferred embodiment of the present invention, the conflict chain generation unit includes: The conflict node pair identification subunit is used to identify device node pairs that occupy the same frequency band and have signal coverage overlap in space based on the spectrum occupancy information of each device node in the extended node set, and generate an initial conflict node pair set. The frequency band continuous extension subunit is used to extend the device nodes along the continuous frequency band coverage direction according to the spectrum occupancy interval relationship of each node in the initial conflict node set, and to include the device nodes with continuous frequency band occupancy relationship into the frequency band extension node set. The interference path construction subunit is used to organize the interference transmission order between nodes according to the signal interference transmission relationship between each device node in the frequency band extension node set, and generate an interference path set. The conflict chain generation subunit is used to chain device nodes according to the interference transmission order based on the set of interference paths, and to constrain the number of nodes in the chain to generate a set of conflict node chains with path length constraints.

[0040] In this embodiment of the invention, by identifying node pairs with overlapping signal coverage based on the spectrum occupancy information and spatial location relationships of device nodes, the basic node combination causing interference can be determined. Further analysis of the spectrum interval relationships between node pairs, and expansion along the direction of continuous frequency band coverage, allows nodes with continuous spectrum occupancy relationships to be included in the same expansion range, thus reflecting the correlation at the frequency band level. By organizing interference paths according to the signal interference propagation relationships between nodes, the interference propagation order between nodes can be expressed in a structured manner. Finally, by chaining nodes according to the interference propagation order and constraining the number of nodes in the chain, a set of conflict node chains with path length limitations can be formed. This ensures that the conflict relationship not only reflects the direct association between nodes but also the path structure during interference propagation, thus providing a clear structural basis for expressing the conflict relationship in the influence subgraph.

[0041] In a preferred embodiment of the present invention, the conflict node pair identification subunit includes: Read the set of extended nodes output by the propagation range extension unit, and extract the spectrum occupancy information and spatial location information corresponding to each device node; The device nodes in the extended node set are grouped according to their spectrum occupancy information, and device nodes that occupy the same frequency band or have overlapping frequency band ranges are grouped into the same candidate group. Within each candidate group, spatial relationship determination is performed on any two device nodes. By comparing the spatial coordinates of the two nodes with their signal coverage range, it is determined whether their signal coverage areas overlap. For a combination of device nodes that satisfy the conditions of overlapping spectrum occupancy and overlapping signal coverage areas, it is recorded as a conflict node pair, and each conflict node pair is given corresponding frequency band information, spatial location relationship information and unique node identifier. The conflicting node pairs in all candidate groups are summarized, and duplicate node pairs are deduplicated to form an initial set of conflicting node pairs for use by subsequent frequency band continuous expansion sub-units.

[0042] In a preferred embodiment of the present invention, the frequency band continuous extension subunit includes: Read the initial set of conflict node pairs output by the conflict node pair identification subunit, and extract the spectrum occupancy information of each node, wherein the spectrum occupancy information includes at least the frequency band start position and the frequency band end position; Starting with each pair of conflicting nodes, obtain their corresponding spectrum occupancy range, and search for other device nodes in the extended node set that are adjacent to or connected to the frequency band range. For determining the adjacent or connecting relationship, the interval range between the end position of two frequency band intervals and the start position of another interval is compared. When the interval range is less than the preset frequency band interval condition, the corresponding node is determined to be a frequency band continuous node. Nodes that meet the frequency band continuity condition are gradually added to the expansion set, and the next round of frequency band continuity expansion is carried out based on the newly added nodes until there are no new nodes that meet the condition. Record the node relationships formed during the expansion process, and save the source node and expansion order information of each node; All the expanded nodes are aggregated to form a frequency band expansion node set, which is then output for the interference path construction subunit to call.

[0043] In a preferred embodiment of the present invention, the interference path construction subunit includes: Read the set of frequency band extension nodes output by the frequency band continuous extension sub-unit, and extract the connection relationship information, spectrum occupancy information and spatial position relationship between each node; The direction of signal interference propagation is determined based on the spatial relationship and spectrum occupancy between nodes. The direction of interference propagation is determined by a comprehensive assessment of the distance between nodes, the difference in transmission power, and the degree of spectrum overlap. Starting with the node in the conflict node pair, the subsequent nodes are searched level by level along the direction of interference propagation, and the connection order between the nodes is recorded as the path order. During the path construction process, when a node has already appeared in the current path, the extension of the path is terminated to avoid forming a loop path; For each completed path, record its starting node, path node sequence, and path length information, and save the frequency band information and interference source information involved in the path; All constructed paths are summarized to form a set of interference paths, which are then called by the conflict chain generation subunit.

[0044] In a preferred embodiment of the present invention, the conflict chain generation subunit includes: Read the interference paths to construct the interference path set output by the sub-unit, and parse each interference path according to the node order; Based on the node order of each interference path, the device nodes in the path are connected sequentially according to the interference transmission order to form a node chain structure, and each node in the chain is assigned an identifier indicating its sequential position in the chain. Based on the preset path length constraint, the number of nodes in each node chain is checked. When the number of nodes exceeds the preset length, the node chain is truncated, and only the node sequence that meets the length constraint is retained. Perform an integrity check on the truncated node chain to ensure that the interference transmission relationship between the nodes in the chain is still maintained; All node chains that satisfy the path length constraint are aggregated, and duplicate node chains are deduplicated to form the final set of conflicting node chains. For each conflicting node chain, add corresponding path information, frequency band information, and node order information, and output them for the influence subgraph construction unit to call.

[0045] In a preferred embodiment of the present invention, the hierarchical recursive correction unit includes: The recursive order determination subunit is used to determine the update order of device nodes according to the propagation level generated by the propagation level generation unit, the propagation path depth, and the node connection relationship, and generate a recursive update sequence. The path weight transfer subunit is used to transfer the signal influence status of the previous level device node according to the influence weight corresponding to the propagation path in the recursive update sequence, and form the corresponding path influence record. The coupling relationship correction subunit is used to perform path-by-path correction processing on the signal coupling relationship of the current level device node based on the path influence record, and generate coupling update results. The conflict state coordination adjustment subunit is used to synchronously adjust the spectrum occupancy conflict state of the current level device node according to the conflict correlation while generating the coupled update result, and generate the hierarchical recursive update result.

[0046] In this embodiment of the invention, by determining the update order of device nodes according to the propagation hierarchy and node connection relationships, state updates can be organized according to the sequence of signal propagation paths, thereby avoiding result deviations caused by inconsistent update orders. By transmitting the signal influence state of the previous-level device nodes according to the propagation path and forming a path influence record, the transmission process of signal influence on different paths can be expressed. Furthermore, by correcting the signal coupling relationship of the current-level device nodes path by path based on the path influence record, the influence of each propagation path on the node state can be reflected in the update process. Combined with the synchronous adjustment of spectrum occupancy conflict states, signal coupling relationships and spectrum conflict relationships can be considered simultaneously during the update process, thereby ensuring that the hierarchical recursive update results remain consistent under the constraints of propagation paths and conflict relationships, providing path-dependent update results for local evolution updates.

[0047] In a preferred embodiment of the present invention, the recursive order determination of the sub-unit includes: Read the propagation hierarchy structure output by the propagation hierarchy generation unit, and extract the unique identifier of the device node in each hierarchy and its hierarchy number; Based on the propagation path relationship, establish a node connection mapping relationship for the device nodes in each level, so that each node can be associated with its upstream and downstream nodes. Starting from the level where the initial node set is located, process the nodes at each level in ascending order of their level numbers. Within the same level, the device nodes are sorted according to the connection relationship between nodes and the order of propagation paths, with priority given to nodes that have a clear connection relationship with upstream nodes and shorter paths. The nodes at each level are concatenated according to the hierarchical order and the sorting results within the level to form a global recursive update sequence, and each node in the sequence is appended with its hierarchical information and path source information; The generated recursive update sequence is checked to ensure that there are no cross-level reverse updates or duplicate updates. After confirming that there are no errors, the sequence is output for the path weight transfer subunit to call.

[0048] In a preferred embodiment of the present invention, the path weight transfer subunit includes: The recursive order is read to determine the recursive update sequence output by the sub-unit, and the upstream node set and corresponding propagation path information of each device node in the sequence are obtained in sequence. For each device node, extract the corresponding signal impact status from its upstream node set, where the signal impact status includes at least the signal strength change, the impact of spectrum occupancy conflict, and the propagation path identifier. The influence weight is determined based on the propagation path information. The distance relationship, path level difference and path continuity are evaluated sequentially according to the preset priority rules, and the influence weight is determined based on the evaluation results. For example, the closer the node is, the smaller the level difference is, and the more continuous the path is, the higher its corresponding influence weight. The signal influence status of upstream nodes is transmitted and processed according to the corresponding influence weights. The influence of multiple upstream nodes is accumulated one by one to form the path influence record of the current node. During the accumulation process, the path identifier, weight, and impact content of the impact from different sources are recorded to ensure that subsequent processing can distinguish the impact sources from different paths. The generated path impact record is appended to the current node and output for use by the coupling relationship correction subunit.

[0049] In a preferred embodiment of the present invention, the coupling relationship correction subunit includes: Read the path impact record output by the path weight transfer subunit and extract the signal impact content of the current device node under each propagation path; Based on the signal strength changes in the path impact record, the coupling relationship between the current device node and its associated nodes is adjusted. The coupling relationship adjustment includes at least signal strength level adjustment, connection strengthening or weakening, and communication link status changes. For the influence from different paths, the influence is weighted according to the corresponding influence weight in the path influence record, so that the path with higher weight has a greater impact on the correction of the coupling relationship; When multiple paths produce different adjustment results for the same coupling relationship, they are processed one by one according to the weight and path order in the path influence record to form a comprehensive coupling relationship correction result. The corrected coupling relationship result is compared with the original coupling relationship of the current node, and the changes and the source path information of the changes are recorded to generate the coupling update result; the coupling update result is output for the conflict state collaborative adjustment subunit to call.

[0050] In a preferred embodiment of the present invention, the conflict state collaborative adjustment subunit includes: Read the coupling update result output by the coupling relationship correction subunit, and extract the spectrum occupancy status corresponding to the current device node and its associated conflict node information; Based on the relationship between conflicting nodes, check the spectrum occupancy conflict status of the current device node to determine whether there is frequency band overlap, continuous frequency band occupancy, or impact of conflict chain transmission. For conflicting frequency bands, the spectrum occupancy status of the current device node is adjusted according to the impact intensity of the conflicting nodes and the conflict propagation path. The adjustment methods include changing the frequency band occupancy range, adjusting the occupancy priority, and updating the conflict marker. During the adjustment process, the signal strength changes reflected in the coupling update results are linked with the spectrum conflict situation, so that the signal enhancement or weakening has a corresponding impact on the spectrum occupancy conflict state. The adjusted spectrum occupancy status and coupling relationship correction results are integrated to form the complete updated status of the current device node at this level. The update status of each device node at the current level is summarized according to the recursive update sequence to generate a hierarchical recursive update result, which is then output for the unified path result unit to call.

[0051] In a preferred embodiment of the present invention, the path result unification unit includes: The path result collection subunit is used to summarize the update results of the same device node under different propagation paths, classify and identify them according to the propagation path, and generate a path result set. The conflict intensity determination subunit is used to determine the conflict intensity of each update result based on the spectrum occupancy conflict and signal impact degree corresponding to each update result in the path result set, and generate a conflict intensity sequence. The priority sorting subunit is used to sort the update results in the path result set according to the conflict intensity sequence and the order of propagation paths, and generate a priority sequence. The result adjudication generation subunit is used to select the update result in the path result set according to the priority sequence, and determine the update result with the highest priority as the unified state result for generation.

[0052] In this embodiment of the invention, by summarizing the update results of the same device node under different propagation paths and classifying them according to the propagation path, multiple state results generated by multi-path propagation can be centrally managed. By determining the conflict intensity based on the spectrum occupancy conflict and signal impact of each update result, the degree of influence of different paths on the device node can be distinguished. Furthermore, by sorting the update results according to the conflict intensity and propagation path order, a sequence of update results with priority can be formed. Selecting the corresponding update result as the unified state result based on this priority sequence can avoid the state inconsistency problem caused by multi-path propagation, enabling the device node to form a unique state expression under the influence of multiple paths, thereby ensuring the consistency of data in subsequent local updates and overall fusion processes.

[0053] In a preferred embodiment of the present invention, the conflict intensity determination subunit includes: Read the path result set output by the path result collection subunit, and extract the spectrum occupancy information, conflict node information and signal impact record corresponding to each update result; For each update result, count the number of frequency bands involved in the corresponding spectrum occupancy conflict, and record the overlap between frequency bands; The number of conflicting nodes involved in the update results is counted, and the scope of conflict propagation is determined based on the association paths between the conflicting nodes; Based on the signal impact record, the degree of signal strength change in the updated results is evaluated, and the degree of signal strength change is divided according to the range of signal enhancement or weakening. The number of frequency bands, the number of conflicting nodes, and the degree of signal impact are comprehensively merged to generate a corresponding conflict intensity value for each update result. The conflict intensity value is obtained by classifying and evaluating the above three types of factors and combining them according to a unified rule. The unified rule is to classify the frequency bands, the number of conflicting nodes, and the degree of signal impact into levels respectively, and combine them according to a preset weight order. Based on the unique identifier of the device node, the conflict intensity values ​​corresponding to each path are sorted to form a conflict intensity sequence, which is then output for the priority sorting subunit to call.

[0054] In a preferred embodiment of the present invention, the local evolution update module further includes: The time continuity constraint unit is used to match and align the unified state result corresponding to the current time slice with the historical state of the device node corresponding to the previous time slice based on the unique identifier of the device node after the unified state result is generated by the path result unification unit, and generate a state alignment result. The state deviation identification unit is used to compare the state changes of each device node between the current time slice and the previous time slice according to the state alignment result, identify device nodes that exceed the preset continuous change range, and generate a set of deviation nodes. The continuity correction unit is used to correct the unified state result of the corresponding device node according to the deviation node set, so that the change between the corrected device node state and the historical state of the corresponding device node in the previous time slice is kept within a preset continuous change range, and the corrected unified state result is generated. The result transmission unit is used to transmit the corrected unified state result to the local result write-back unit, so that the local result write-back unit can generate a local update result.

[0055] In this embodiment of the invention, after the unified state result is generated by the path result unification unit, the unified state result corresponding to the current time slice is matched and aligned with the historical state of the device node corresponding to the previous time slice, thereby establishing a state correspondence in the time dimension. By comparing the state changes of each device node between the current time slice and the previous time slice node by node, device nodes whose state changes exceed a preset continuous change range can be identified, thereby locating the set of nodes that need adjustment. Furthermore, by correcting the unified state result of such nodes, the corrected state changes are kept within the preset continuous change range, ensuring that the device nodes maintain a continuous change relationship during the time evolution process. By transmitting the corrected unified state result to the local result write-back unit, the local update results can be kept consistent in the time dimension, thereby providing a continuous state data foundation for the overall fusion generation.

[0056] In a preferred embodiment of the present invention, the preset continuous variation range, used to constrain the magnitude of state changes of the same device node between adjacent time slices, includes: Continuous variation range constraint rules are established for both signal coupling relationship state and spectrum occupancy state. The continuous variation range of signal coupling relationship state is determined by limiting the amplitude level of signal strength change, the number of connection relationship changes, and the range of change types. The range of continuous change in spectrum occupancy status is determined by limiting the changes in the start and end positions of the frequency band, the changes in the number of frequency band occupancy intervals, and the range of frequency band migration. In actual processing, the current time slice state is compared with the previous time slice state item by item. When any change exceeds the set range, it is determined to be out of the continuous change range. By constraining the change amplitude, the device node state exhibits continuous change characteristics in the time dimension, avoiding abrupt changes that do not conform to the time evolution law.

[0057] In a preferred embodiment of the present invention, the time continuity constraint unit includes: Read the unified status result set output by the unified path result unit, and extract the unique identifier and current time slice mark corresponding to each device node; In the digital twin electromagnetic environment model, the historical state record of the corresponding node in the previous time slice is found based on the unique identifier of the device node, and its signal coupling relationship status, spectrum occupancy status and time stamp information are extracted. The unified state result of the current time slice is matched one by one with the historical state of the previous time slice according to the unique identifier of the device node, and a state correspondence relationship at the node level is established. For each matched device node, a state alignment record is generated, which includes at least the current time slice state, the previous time slice state, and the corresponding time interval information. Device nodes for which no corresponding historical state could be found in the previous time slice are marked and treated as new nodes; The status alignment records of all device nodes are summarized to form a status alignment result set, which is then output for the status deviation identification unit to call.

[0058] In a preferred embodiment of the present invention, the continuity correction unit includes: Read the set of deviation nodes output by the state deviation identification unit, and extract the current unified state result and the historical state of the previous time slice corresponding to each deviation node; For each deviation node, its signal coupling relationship is corrected. The correction method includes reducing the part of the change that exceeds the range, so that the degree of change falls back to the preset change range, while retaining the change direction information. After one correction process, the correction result is directly output without repeated iterative correction. For each deviation node's spectrum occupancy status, the changes in the frequency band occupancy range are adjusted to ensure that the adjusted frequency band range changes do not exceed a preset continuous change range, and to maintain the continuity of frequency band usage. During the correction process, multiple changes in the same device node are handled in a unified and coordinated manner to ensure consistency between changes in signal coupling and changes in spectrum occupancy. Generate a revised status record for the device nodes that have been corrected, and attach a correction mark and a description of the basis for the correction; The corrected status records are merged with the status of device nodes that are not marked as deviation nodes to form a corrected unified status result set, which is then output for the result transmission unit to call.

[0059] In a preferred embodiment of the present invention, the overall fusion generation module includes: The region division unit is used to divide the model space into updated and unupdated regions based on the coverage of the influence subgraph in the digital twin electromagnetic environment model. The boundary recognition unit is used to identify the boundary position between the updated area and the unupdated area, obtain the status information of the device nodes on both sides of the boundary, and generate boundary association data. The boundary state adjustment unit is used to compare the state differences of device nodes on both sides of the boundary based on the boundary association data, and adjust the state of device nodes at the boundary position in the updated area so that the difference between the adjusted device node state and the corresponding device node state in the unupdated area meets the preset difference range, and generates the boundary adjustment result. The fusion result generation unit is used to fuse the local update results of the updated region with the previous time result of the unupdated region based on the boundary adjustment results, and generate the overall electromagnetic environment simulation result at the current time.

[0060] In this embodiment of the invention, by dividing the model space according to the coverage of the influence subgraph in the digital twin electromagnetic environment model, the model can be divided into updated and unupdated regions, thereby clarifying the processing methods for different regions. By identifying the boundary positions between the updated and unupdated regions and obtaining the state information of the device nodes on both sides of the boundary, boundary association data for subsequent processing can be formed. By comparing the state differences of the device nodes on both sides of the boundary and adjusting the state of the device nodes at the boundary position in the updated region so that the difference between them and the corresponding device node states in the unupdated region meets a preset difference range, the state changes at the boundary can be constrained. Furthermore, by fusing the adjusted updated region results with the previous time-to-time results of the unupdated region, an overall electromagnetic environment deduction result that maintains spatial continuity can be formed, thereby creating a consistent overall expression between the local updated results and the unupdated region.

[0061] In a preferred embodiment of the present invention, the preset difference range is used to constrain the difference in device node states between the updated region and the unupdated region at the boundary position, including: Difference range constraint rules are established for the differences in signal coupling relationship status and the differences in spectrum occupancy status, respectively. The differences in signal coupling relationship status are determined by comparing the differences in signal strength level, number of connections and connection type between the nodes on both sides of the boundary. Differences in spectrum occupancy status are determined by comparing the differences between nodes on both sides of the boundary in frequency band range, number of occupied intervals, and frequency band overlap. During the boundary adjustment process, the nodes on the updated region side are compared one by one with the nodes on the unupdated region side. When the difference exceeds the set range, the state of the node on the updated region side is adjusted to reduce the change range to within the allowable range. This method ensures that the state changes of the nodes on both sides of the boundary maintain a smooth transition in space, thereby avoiding inconsistent state differences between the local update results and the unupdated region.

[0062] In a preferred embodiment of the present invention, the region division unit includes: Read the influence subgraph output by the influence subgraph construction unit, and extract the spatial location information and coverage information of the device nodes in it; Based on the spatial location of the device nodes, the spatial range involved in the influence subgraph is mapped to the spatial grid or regional cells of the digital twin electromagnetic environment model to determine the spatial range covered by the influence subgraph. The covered spatial range is marked as the updated region, and the uncovered spatial range is marked as the unupdated region; Identify the device nodes within the update area so that they correspond to the write range of subsequent partial update results; Identify the device nodes in the unupdated area to maintain their state data from the previous time slice; The results of dividing the updated and unupdated regions are organized and region identification information and time slice markers are added to form the region division results, which are then output for the boundary recognition unit to use.

[0063] In a preferred embodiment of the present invention, the boundary recognition unit includes: Read the region division results output by the region division unit and extract the spatial distribution information of the updated and unupdated regions; Based on the spatial adjacency relationship between the updated and unupdated regions, the boundary position between them is identified, and the spatial unit where the boundary position is located is marked as the boundary region; Within the boundary area, find the device nodes belonging to the updated area and the non-updated area respectively, and establish the node correspondence so that each boundary position corresponds to one or more pairs of cross-region nodes. Extract the status information of the device nodes on both sides of the boundary, including at least the signal coupling relationship status, spectrum occupancy status, and time stamp information; The boundary location, the unique identifier of the corresponding device node, and its status information are combined to form a boundary association data set; Perform an integrity check on the boundary-related data to ensure that there is valid node information on both sides of the boundary, and output the results for the boundary state adjustment unit to call.

[0064] In a preferred embodiment of the present invention, the boundary state adjustment unit includes: Read the boundary association data set output by the boundary recognition unit, and extract the status information of the device nodes on both sides of the boundary one by one; The signal coupling relationship status of each pair of boundary nodes is compared to determine the difference between the current time slice and the previous time slice, and whether the difference exceeds the allowable range based on the preset difference range. The spectrum occupancy status of each pair of boundary nodes is compared to determine the differences in the frequency band occupancy range and the number of occupancy intervals, and to determine whether the differences exceed the preset difference range. If the difference exceeds the preset difference range, the status of the device nodes on the update area side is adjusted. The adjustment methods include reducing the change amplitude, gradually reducing the change amplitude of the frequency band occupancy range according to the preset difference range, and adjusting the changes in coupling relationship in stages so that the adjusted difference falls back to the preset difference range. During the adjustment process, the signal coupling relationship and spectrum occupancy status of the same node are processed in a coordinated manner to ensure the consistency of the adjustment results across multiple state dimensions. The adjusted device node status is summarized with the unadjusted node status to form the boundary adjustment result, and the boundary location identifier and adjustment record information are attached. The result is then output for the fusion result generation unit to call.

[0065] In a preferred embodiment of the present invention, the fusion result generation unit includes: Read the boundary adjustment results output by the boundary state adjustment unit and extract the adjusted boundary node states; Read the local update results of the updated region output by the local result write-back unit, as well as the previous time slice state data of the unupdated region in the digital twin electromagnetic environment model; Based on the region division results, the device node status in the updated region is replaced with the local update result, while the device node status in the unupdated region remains as it was in the previous time slice. The status of device nodes within the boundary area is overwritten with the boundary adjustment results to ensure a consistent state transition between the updated and unupdated areas. The device node statuses in the updated region, boundary region, and non-updated region are uniformly integrated, and the node data in the model is reorganized according to spatial location relationships. A consistency check is performed on the fused overall state data. The consistency check includes at least whether the node state is complete, whether the time slice markings are consistent, and whether there are conflicts in the spectrum occupancy relationship. After the inspection is completed, the integrated device node status is written into the digital twin electromagnetic environment model to generate the overall electromagnetic environment simulation results for the current time slice, and output for subsequent analysis or display.

[0066] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A system for dynamically constructing and extrapolating complex electromagnetic environments based on digital twins, characterized in that, The system includes: The digital twin base model construction module is used to associate and organize geographic information, equipment node information and spectrum resource information in the physical space, establish a digital twin electromagnetic environment model that maps to the physical space, and configure a unique identifier and corresponding historical operating status sequence for each equipment node in the digital twin electromagnetic environment model. The state timing modeling module is used to continuously receive state data streams reported by communication devices based on unique identifiers, accumulate historical operating state sequences by time stamping, and generate state change records based on the state change amplitude between adjacent time slices. The trigger interval determination module is used to divide the environmental change trigger interval according to the correspondence between the state change record and the preset change determination condition, and generate the target trigger interval that meets the preset change determination condition. The influence subgraph generation module is used to extract the device nodes whose state changes from the digital twin electromagnetic environment model according to the target triggering interval, and combine the signal propagation correlation and spectrum occupancy conflict relationship between the device nodes to construct an influence subgraph consisting of the set of affected device nodes and their associated spectrum resource subsets. The local evolution update module is used to update the signal coupling relationship and spectrum occupancy status between the affected device node sets within the scope of the influence subgraph, generate local update results, and write the local update results back to the digital twin electromagnetic environment model. The overall fusion generation module is used to retain the previous time-instance results of the unaffected subgraphs in the digital twin electromagnetic environment model, and to stitch and fuse the local update results with the previous time-instance results to generate the overall electromagnetic environment simulation results for the current time-instance.

2. The system for dynamic construction and deduction of complex electromagnetic environments based on digital twins according to claim 1, characterized in that, The trigger interval determination module includes: The change sequence generation unit is used to collect the change content of each device node in each time slice based on the differences in device operation status, spectrum occupancy status and spatial location information of each device node in the state change record between adjacent time slices, and generate the corresponding multidimensional change quantity sequence. The composite identifier determination unit is used to match the changes in device operating status, spectrum occupancy status, and spatial location in each time slice with the corresponding preset determination conditions based on the multidimensional change sequence, and to combine and associate the matching results to generate composite change identifiers for each time slice. The trigger interval construction unit is used to perform continuous connection processing on each time slice according to the composite change identifier and in chronological order, and to combine the time slices with continuous composite change identifiers to construct an environmental change trigger interval. The target trigger interval generation unit is used to filter the constructed environmental change trigger intervals based on the duration and intensity of the change corresponding to the environmental change trigger interval, and generate the target trigger interval.

3. The system for dynamic construction and deduction of complex electromagnetic environments based on digital twins according to claim 1, characterized in that, The influence subgraph generation module includes: The initial node extraction unit is used to extract the device nodes whose states have changed from the digital twin electromagnetic environment model according to the target triggering interval, and generate an initial node set. The propagation range extension unit is used to extend the propagation range node by node according to the initial node set, along the signal propagation path relationship between device nodes, and to constrain the propagation path depth and propagation influence range during the extension process, thereby generating an extended node set; The conflict chain generation unit is used to identify pairs of device nodes with overlapping spectrum occupancy based on the spectrum occupancy status of each device node in the extended node set, and extend along the conflict association path between the device node pairs to generate a set of conflict node chains. The influence subgraph construction unit is used to organize associated device nodes, spectrum resource subsets, and connection relationships between nodes based on the initial node set, the extended node set, and the conflict node chain set, and generate an influence subgraph with propagation path constraints and spectrum conflict chain structure.

4. The system for dynamic construction and deduction of complex electromagnetic environments based on digital twins according to claim 1, characterized in that, The local evolution update module includes: The propagation hierarchy generation unit is used to generate multiple propagation hierarchy with sequential dependencies by dividing the device nodes into layers according to the connection order and path depth of the propagation path, based on the signal propagation path relationship between each device node in the influence subgraph. The hierarchical recursive correction unit is used to perform state updates on each level of device nodes in the order from the starting level to the ending level according to the propagation level, and to perform transmission correction on the signal coupling relationship and spectrum occupancy conflict state of the current level of device nodes according to the update results of the previous level of device nodes, thereby generating a hierarchical update result sequence that is recursively applied layer by layer. The path result unification unit is used to perform consistency processing on the update results of the same device node under different propagation paths according to the hierarchical update result sequence, and generate a unified state result for each device node. The local result write-back unit is used to integrate the device node status after each level of update based on the unified status result, generate local update results, and write the local update results into the corresponding device node positions in the digital twin electromagnetic environment model.

5. The system for dynamic construction and deduction of complex electromagnetic environments based on digital twins according to claim 3, characterized in that, The propagation range extension unit includes: The path depth control subunit is used to mark the node level on each propagation path level by level when expanding nodes along the signal propagation path, and terminate the expansion of the corresponding propagation path when the node level reaches the preset path depth, thereby generating a candidate node set with path depth limited. The influence accumulation judgment subunit is used to accumulate and record the propagation influence of each node along the propagation path based on the signal influence transmission of each node in the candidate node set, and to filter the nodes whose cumulative influence meets the preset influence judgment conditions to generate an influence constraint node set. The spectrum association filtering subunit is used to filter nodes with spectrum associations based on the correspondence between the spectrum occupancy information of each node in the set of influence constraint nodes and the spectrum occupancy information of the nodes in the initial node set, and generate a set of spectrum association nodes. The direction consistency constraint subunit is used to retain nodes with continuous and consistent propagation directions based on the signal propagation direction relationship between nodes in the spectrum-associated node set, and generate an extended node set.

6. The system for dynamic construction and deduction of complex electromagnetic environments based on digital twins according to claim 3, characterized in that, The conflict chain generation unit includes: The conflict node pair identification subunit is used to identify device node pairs that occupy the same frequency band and have signal coverage overlap in space based on the spectrum occupancy information of each device node in the extended node set, and generate an initial conflict node pair set. The frequency band continuous extension subunit is used to extend the device nodes along the continuous frequency band coverage direction according to the spectrum occupancy interval relationship of each node in the initial conflict node set, and to include the device nodes with continuous frequency band occupancy relationship into the frequency band extension node set; The interference path construction subunit is used to organize the interference transmission order between nodes according to the signal interference transmission relationship between each device node in the frequency band extension node set, and generate an interference path set. The conflict chain generation subunit is used to chain device nodes according to the interference transmission order based on the set of interference paths, and to constrain the number of nodes in the chain to generate a set of conflict node chains with path length constraints.

7. The system for dynamic construction and deduction of complex electromagnetic environments based on digital twins according to claim 4, characterized in that, The hierarchical recursive correction unit includes: The recursive order determination subunit is used to determine the update order of device nodes according to the propagation level generated by the propagation level generation unit, the propagation path depth, and the node connection relationship, and generate a recursive update sequence. The path weight transfer subunit is used to transfer the signal influence status of the previous level device node according to the influence weight corresponding to the propagation path in the recursive update sequence, and form the corresponding path influence record. The coupling relationship correction subunit is used to perform path-by-path correction processing on the signal coupling relationship of the current level device node based on the path influence record, and generate coupling update results. The conflict state coordination adjustment subunit is used to synchronously adjust the spectrum occupancy conflict state of the current level device node according to the conflict correlation while generating the coupled update result, and generate the hierarchical recursive update result.

8. A system for dynamic construction and deduction of complex electromagnetic environments based on digital twins according to claim 4, characterized in that, The path result unification unit includes: The path result collection subunit is used to summarize the update results of the same device node under different propagation paths, classify and identify them according to the propagation path, and generate a path result set. The conflict intensity determination subunit is used to determine the conflict intensity of each update result based on the spectrum occupancy conflict and signal impact degree corresponding to each update result in the path result set, and generate a conflict intensity sequence. The priority sorting subunit is used to sort the update results in the path result set according to the conflict intensity sequence and the order of propagation paths, and generate a priority sequence. The result adjudication generation subunit is used to select the update result in the path result set according to the priority sequence, and determine the update result with the highest priority as the unified state result for generation.

9. A system for dynamic construction and deduction of complex electromagnetic environments based on digital twins according to claim 4, characterized in that, The local evolution update module also includes: The time continuity constraint unit is used to match and align the unified state result corresponding to the current time slice with the historical state of the device node corresponding to the previous time slice based on the unique identifier of the device node after the unified state result is generated by the path result unification unit, and generate a state alignment result. The state deviation identification unit is used to compare the state changes of each device node between the current time slice and the previous time slice according to the state alignment result, identify device nodes that exceed the preset continuous change range, and generate a set of deviation nodes. The continuity correction unit is used to correct the unified state result of the corresponding device node according to the deviation node set, so that the change between the corrected device node state and the historical state of the corresponding device node in the previous time slice is kept within a preset continuous change range, and the corrected unified state result is generated. The result transmission unit is used to transmit the corrected unified state result to the local result write-back unit, so that the local result write-back unit can generate a local update result.

10. A system for dynamic construction and deduction of complex electromagnetic environments based on digital twins according to claim 1, characterized in that, The overall fusion generation module includes: The region division unit is used to divide the model space into updated and unupdated regions based on the coverage of the influence subgraph in the digital twin electromagnetic environment model. The boundary recognition unit is used to identify the boundary position between the updated area and the unupdated area, obtain the status information of the device nodes on both sides of the boundary, and generate boundary association data. The boundary state adjustment unit is used to compare the state differences of device nodes on both sides of the boundary based on the boundary association data, and adjust the state of device nodes at the boundary position in the updated area so that the difference between the adjusted device node state and the corresponding device node state in the unupdated area meets the preset difference range, and generates the boundary adjustment result. The fusion result generation unit is used to fuse the local update results of the updated region with the previous time result of the unupdated region based on the boundary adjustment results, and generate the overall electromagnetic environment simulation result at the current time.

Images