Method, device and electronic equipment for handling low altitude network interference

Abstract

The application provides a low-altitude network interference processing method and device and electronic equipment, wherein the method comprises: searching for to-be-processed interference information in a scheme library knowledge graph, and determining a solution to the to-be-processed interference information based on the obtained search information; a scheme library knowledge graph construction method comprises: acquiring wireless side data collected by a low-altitude network monitoring platform; constructing a plurality of knowledge graphs based on entities and relationships identified in the wireless side data; performing semantic analysis on the plurality of knowledge graphs, determining cross-graph correlation, and constructing a scheme library knowledge graph based on the cross-graph correlation. Based on the constructed scheme library knowledge graph, deep correlations between interference data, sources, solutions, and effects are mined, the interference positioning and solution matching efficiency are improved, the solution to the to-be-processed interference information can be quickly determined, the processing time is greatly shortened, and the processing efficiency is improved.

Description

Technical Field

[0001] This invention relates to the field of computer technology, and in particular to a method, apparatus, and electronic device for processing low-altitude network interference. Background Technology

[0002] Low-altitude networks (such as drone communication and low-altitude broadband access) face problems such as frequent signal interference, difficulty in locating interference sources, and low efficiency in matching solutions due to scarce spectrum resources and complex and changeable environment.

[0003] In existing technologies, the analysis process for low-altitude network interference generally relies on human experience or single-dimensional data analysis, resulting in long interference analysis cycles and insufficient accuracy in determining solutions. Summary of the Invention

[0004] This invention provides a method, apparatus, and electronic device for processing low-altitude network interference, which addresses the shortcomings of existing technologies in analyzing low-altitude network interference, which rely on manual experience or single-dimensional data analysis, resulting in long interference analysis cycles and insufficient accuracy in determining solutions. This invention improves the efficiency and accuracy of low-altitude network interference processing.

[0005] This invention provides a method for handling low-altitude network interference, comprising the following steps: Retrieve the interference information to be processed from the solution database knowledge graph, and determine the solution to the interference information to be processed based on the retrieved information; The method for constructing the solution base knowledge graph includes: Acquire wireless side data collected by the low-altitude network monitoring platform; the wireless side data includes low-altitude network interference data, interference source data, solution data, and processing effect data; Based on the entities and relationships identified in the wireless side data, multiple knowledge graphs are constructed; the multiple knowledge graphs include an interference data knowledge graph, a solution knowledge graph, and a processing effect knowledge graph. Semantic analysis is performed on the multiple knowledge graphs to determine the cross-graph associations between them, and the solution library knowledge graph is constructed based on the cross-graph associations.

[0006] According to a method for processing low-altitude network interference provided by the present invention, the step of performing semantic analysis on the plurality of knowledge graphs to determine the cross-graph association relationships between the plurality of knowledge graphs includes: Semantic analysis is performed on the multiple knowledge graphs to determine the semantic similarity between different knowledge graph nodes in the multiple knowledge graphs; Based on the semantic similarity between the different knowledge graph nodes, cross-graph paths between the different knowledge graph nodes are generated, and the cross-graph paths are used as the cross-graph association relationships.

[0007] According to a method for processing low-altitude network interference provided by the present invention, the step of performing semantic analysis on the plurality of knowledge graphs to determine the semantic similarity between different knowledge graph nodes in the plurality of knowledge graphs includes: Based on graph embedding technology, the nodes of each knowledge graph in the multiple knowledge graphs are mapped to low-dimensional vectors to obtain multiple low-dimensional vector representations; The multiple low-dimensional vector representations are projected onto the same vector space, and the semantic similarity of the multiple low-dimensional vectors is calculated in the vector space to determine the semantic similarity between different knowledge graph nodes in the multiple knowledge graphs.

[0008] According to a method for handling low-altitude network interference provided by the present invention, the step of constructing the scheme base knowledge graph based on the cross-graph association relationship includes: Based on the cross-graph paths of the different knowledge graph nodes, multiple cross-graph relation triples are constructed; Based on the multiple cross-graph relation triples, the solution base knowledge graph is constructed.

[0009] According to a method for handling low-altitude network interference provided by the present invention, after constructing the scheme base knowledge graph, the method further includes: The low-altitude network interference data is input into the prediction model to obtain the solution prediction result of the low-altitude network interference data output by the prediction model. The prediction model is trained based on historical interference data and the corresponding solution labels for the historical interference data. Based on the semantic similarity between the prediction results of the solution and each node in the solution knowledge graph, multiple inference associations associated with the low-altitude network interference data are determined. Based on the multiple reasoning relationships, the solution base knowledge graph is supplemented to obtain a supplemented solution base knowledge graph.

[0010] According to a method for processing low-altitude network interference provided by the present invention, after constructing multiple knowledge graphs based on entities and relationships identified in the wireless side data, the method further includes: Semantic analysis is performed on the nodes of each knowledge graph in the plurality of knowledge graphs to determine the synonym nodes in the nodes of each knowledge graph; Delete the synonym node.

[0011] According to a method for processing low-altitude network interference provided by the present invention, the step of constructing multiple knowledge graphs based on entities and relationships identified in the radio side data includes: Based on the low-altitude network interference data and the entities and relationships identified in the interference source data, an interference data knowledge graph is constructed. Based on the entities and relationships identified in the solution data, a solution knowledge graph is constructed; Based on the entities and relationships identified in the processing effect data, a processing effect knowledge graph is constructed.

[0012] According to a method for handling low-altitude network interference provided by the present invention, after constructing the scheme base knowledge graph, the method further includes: Receive the updated data from the wireless side; Based on the updated data, the cross-graph association is updated, and based on the updated cross-graph association, the solution base knowledge graph is updated.

[0013] The present invention also provides a low-altitude network interference processing device, comprising the following modules: An interference processing module is used to retrieve interference information to be processed from the solution database knowledge graph, and determine a solution for the interference information to be processed based on the retrieved information. The method for constructing the solution base knowledge graph includes: Acquire wireless side data collected by the low-altitude network monitoring platform; the wireless side data includes low-altitude network interference data, interference source data, solution data, and processing effect data; Based on the entities and relationships identified in the wireless side data, multiple knowledge graphs are constructed; the multiple knowledge graphs include an interference data knowledge graph, a solution knowledge graph, and a processing effect knowledge graph. Semantic analysis is performed on the multiple knowledge graphs to determine the cross-graph associations between them, and the solution library knowledge graph is constructed based on the cross-graph associations.

[0014] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the program to implement the low-altitude network interference processing method as described above.

[0015] The low-altitude network interference processing method, apparatus, and electronic equipment provided by this invention construct multiple knowledge graphs and achieve cross-graph associations based on collected wireless data. This establishes a connection between interference information, solutions, and processing effects, facilitating rapid problem localization and finding the optimal solution, thus improving processing efficiency. Based on the cross-graph associations of multiple knowledge graphs, a solution library knowledge graph is constructed. Using this constructed solution library knowledge graph, deep correlations between interference data and its source, solutions, and effects are mined, improving the efficiency of interference localization and solution matching. This enables rapid determination of solutions for the interference information to be processed, significantly shortening processing time and improving processing efficiency. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0017] Figure 1 This is a flowchart illustrating the low-altitude network interference handling method provided by the present invention.

[0018] Figure 2 This is a schematic diagram of the knowledge graph expansion process provided by the present invention.

[0019] Figure 3 This is a schematic diagram of the low-altitude network interference analysis and processing system provided by the present invention.

[0020] Figure 4 This is a schematic diagram of the low-altitude network interference processing device provided by the present invention.

[0021] Figure 5 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0023] Figure 1 This is a flowchart illustrating the low-altitude network interference handling method provided by the present invention, as shown below. Figure 1 As shown, the method includes the following: Step 110: Retrieve the interference information to be processed in the solution database knowledge graph, and determine the solution for the interference information to be processed based on the retrieved information; The method for constructing the solution base knowledge graph includes: Acquire wireless side data collected by the low-altitude network monitoring platform; the wireless side data includes low-altitude network interference data, interference source data, solution data, and processing effect data; Based on the entities and relationships identified in the wireless side data, multiple knowledge graphs are constructed; the multiple knowledge graphs include an interference data knowledge graph, a solution knowledge graph, and a processing effect knowledge graph. Semantic analysis is performed on the multiple knowledge graphs to determine the cross-graph associations between them, and the solution library knowledge graph is constructed based on the cross-graph associations.

[0024] The execution subject of the low-altitude network interference processing method provided by this invention can be an electronic device, a component in the electronic device, an integrated circuit, or a chip. The electronic device can be a mobile electronic device or a non-mobile electronic device. For example, a mobile electronic device can be a mobile phone, tablet computer, laptop computer, PDA, ultra-mobile personal computer (UMPC), netbook, or personal digital assistant (PDA), etc., while a non-mobile electronic device can be a server, network attached storage (NAS), or personal computer (PC), etc. This invention does not impose specific limitations.

[0025] The technical solution of this invention will be described in detail below using the computer executing the low-altitude network interference processing method provided by this invention as an example.

[0026] In step 110, the interference information to be processed is retrieved from the solution knowledge graph, and a solution for the interference information to be processed is determined based on the retrieved information.

[0027] Before determining a solution for the interference information to be processed, a solution knowledge graph is pre-built.

[0028] Data is collected from the low-altitude network monitoring platform. The main function of this platform is to monitor wireless signals in low-altitude networks, including key indicators such as signal quality, interference, and network coverage. It can collect and analyze wireless data in low-altitude networks in real time, such as interference data, interference source data, solution data, and processing effect data, providing data support for network management and optimization.

[0029] The acquired wireless data forms the foundation for building the knowledge graph, containing various key information about the low-altitude network environment. Specifically, the wireless data covers the following aspects: Low-altitude network interference data: Records characteristic parameters of various interference signals appearing in low-altitude networks, such as interference frequency, intensity, and duration. This data reflects the severity and scope of the interference. Interference source data: Traces the source of the interference signal, including information such as the type, location, and operating status of the interfering device, helping to locate the interference source and take corresponding countermeasures. Solution data: Collects solutions for different types of interference, including solution descriptions, implementation steps, and required resources, providing a reference for handling interference problems. Effectiveness data: Records the effectiveness evaluation results after solution implementation, such as whether the interference has been eliminated and whether network performance has been restored, used to verify the effectiveness of the solution.

[0030] The collected wireless data undergoes preprocessing operations such as cleaning, deduplication, and format conversion to improve data quality and consistency. Key entities, such as interference types, interference sources, and solutions, are identified from the preprocessed data. These entities form the foundational nodes for building the knowledge graph. Relationships between entities are analyzed, such as the association between interference types and interference sources, and the causal relationship between solutions and treatment effects. These relationships connect the entities, forming the edges of the knowledge graph.

[0031] Based on the identified entities and relationships, a knowledge graph of interference data, a knowledge graph of solutions, and a knowledge graph of processing effects are constructed respectively.

[0032] After constructing the interference data knowledge graph, the solution knowledge graph, and the processing effect knowledge graph, semantic analysis is performed on each knowledge graph, and the cross-graph associations between them are determined.

[0033] Specifically, cross-graph relationships can be determined by analyzing the semantic connections between different knowledge graphs. For example, a certain type of interference in the interference data knowledge graph may be related to a certain solution in the solution knowledge graph because the solution is designed to target that type of interference.

[0034] Optionally, cross-graph relationships between multiple knowledge graphs can be mined using graph neural networks. Graph neural networks can identify implicit bridging points between three graphs based on semantic relationships. For example: interference features (such as "high-frequency interference") → triggering a specific scheme (such as "frequency band switching") → achieving the desired effect (such as "signal strength increased by 30%).

[0035] By leveraging node similarity, implicit bridging points between different knowledge graphs are linked. For example, if the vector of interference event A is close to the vector of solution B, and the vector of solution B is close to the vector of effect C, then the path A→B→C is constructed. Cross-graph paths are generated through graph walks (such as random walks), for example: (interference data) high-frequency interference → (solution) frequency band switching → (processing effect) signal strength increase of 30%.

[0036] Graph neural networks utilize message passing mechanisms to aggregate information from neighboring nodes and update node features. For example, a node in a "frequency band switching" scheme simultaneously receives information from both interference maps (such as "high-frequency interference" nodes) and effect maps (such as "signal strength enhancement" nodes).

[0037] Through multi-layer propagation, the graph neural network automatically learns the relationship strength weights between the three types of graphs (e.g., the association weight of "high-frequency interference → frequency band switching" is higher than that of "environmental noise → frequency band switching").

[0038] By using multi-layer propagation of graph neural networks, indirect correlations across graphs (such as "vegetation density → multipath interference → anti-interference technology configuration → bit error rate reduction") are captured, thus obtaining cross-graph correlation relationships.

[0039] Based on established cross-graph relationships, multiple knowledge graphs are integrated and merged to construct a unified solution database knowledge graph. This knowledge graph not only includes knowledge and relationships from various fields but also enables cross-domain knowledge association and reasoning.

[0040] After constructing the solution knowledge graph, when dealing with interference problems in low-altitude networks, the first step is to retrieve relevant interference information from the solution knowledge graph.

[0041] Determine search keywords or entities: Based on the characteristics of the interference problem, determine the search keywords or entities. For example, if the interference problem manifests as a degraded signal quality, then the search keywords might be "degraded signal quality" or related entities such as "interference signal" or "network performance".

[0042] Perform a search using the knowledge graph of the solution database. Searches can be conducted through keyword matching, entity linking, or semantic queries. Obtain interference information related to the search keywords or entities from the knowledge graph of the solution database. This information may include the type, source, and scope of the interference.

[0043] After obtaining the interference information to be processed, appropriate solutions need to be determined based on the relationships within the solution knowledge graph. A detailed analysis of the retrieved interference information is conducted to understand its characteristics and impact. Reasoning analysis is performed based on cross-graph relationships within the solution knowledge graph. For example, solutions targeting specific interference types can be found based on the relationship between them. The found solutions are evaluated, considering factors such as feasibility, effectiveness, and cost. Multiple solutions can be compared and optimized. Based on the evaluation results, the final solution is determined. This solution should provide effective measures to address the interference problem.

[0044] The low-altitude network interference processing method provided by this invention constructs multiple knowledge graphs based on collected wireless data and achieves cross-graph association, establishing connections between interference information, solutions, and processing effects. This facilitates rapid problem localization and the identification of optimal solutions, improving processing efficiency. Based on the cross-graph associations of multiple knowledge graphs, a solution library knowledge graph is constructed. Using this constructed solution library knowledge graph, deep relationships between interference data and its source, solutions, and effects are mined, improving interference localization and solution matching efficiency. This enables rapid determination of solutions to the interference information to be processed, significantly shortening processing time and increasing efficiency.

[0045] In one embodiment, semantic analysis is performed on the plurality of knowledge graphs to determine cross-graph associations between the plurality of knowledge graphs, including: performing semantic analysis on the plurality of knowledge graphs to determine the semantic similarity between different knowledge graph nodes in the plurality of knowledge graphs; generating cross-graph paths between different knowledge graph nodes based on the semantic similarity between the different knowledge graph nodes, and using the cross-graph paths as the cross-graph associations.

[0046] Semantic parsing is performed on nodes in the interference data knowledge graph, solution knowledge graph, and processing effect knowledge graph to extract semantic feature vectors of entities.

[0047] The cosine similarity algorithm can be used to calculate the semantic similarity between different graph nodes, quantifying the strength of semantic association between entities. For example, the interference type node "co-frequency interference" and the solution node "spectrum resource reallocation" may be identified as a potential association pair due to their high semantic correlation.

[0048] Valid associations are filtered based on similarity thresholds, and cross-graph paths are generated through path search algorithms, such as a complete inference chain of "interference data → interference localization → solution → effect evaluation". Finally, these cross-graph paths are embedded as associations into the solution base knowledge graph, enabling rapid tracing of interference information and accurate matching of solutions, which can significantly improve the intelligence level of interference handling.

[0049] In one embodiment, semantic analysis is performed on the plurality of knowledge graphs to determine the semantic similarity between different knowledge graph nodes in the plurality of knowledge graphs, including: mapping each node of the plurality of knowledge graphs into a low-dimensional vector based on graph embedding technology to obtain a plurality of low-dimensional vector representations; projecting the plurality of low-dimensional vector representations onto the same vector space, and calculating the semantic similarity of the plurality of low-dimensional vectors in the vector space to determine the semantic similarity between different knowledge graph nodes in the plurality of knowledge graphs.

[0050] Based on graph embedding technology, each node in a knowledge graph is mapped to a low-dimensional vector. Graph embedding technology can capture the structural and semantic information of nodes in the graph, transforming nodes from a high-dimensional graph structure space to a low-dimensional vector space, while preserving the important features of the nodes.

[0051] Projecting multiple low-dimensional vector representations onto the same vector space allows node vectors from different graphs to be compared within the same space. In this vector space, the semantic similarity of multiple low-dimensional vectors can be determined by calculating indices such as distance or angle between vectors.

[0052] In one embodiment, constructing the solution library knowledge graph based on the cross-graph association relationship includes: constructing multiple cross-graph relationship triples based on the cross-graph paths of different knowledge graph nodes; and constructing the solution library knowledge graph based on the multiple cross-graph relationship triples.

[0053] By mining cross-spectral paths, directly related cross-spectral relationship triples are generated. The generated cross-spectral relationship triples can be: <High-frequency interference, recommended solution, frequency band switching>, <Frequency band switching, resulting in effect, signal strength improvement of 30%>, etc.

[0054] A solution library knowledge graph is constructed based on cross-graph relation triples. Specifically, entities in the cross-graph relation triples can be used as nodes in the knowledge graph, and relations can be used as edges between nodes, which are then stored and managed using a graph database.

[0055] In one embodiment, after constructing the solution knowledge graph, the method further includes: inputting the low-altitude network interference data into a prediction model to obtain the solution prediction result of the low-altitude network interference data output by the prediction model; the prediction model is trained based on historical interference data and the solution labels corresponding to the historical interference data; based on the semantic similarity between the solution prediction result and each node in the solution knowledge graph, determining multiple inference relationships associated with the low-altitude network interference data; and supplementing the solution knowledge graph based on the multiple inference relationships to obtain a supplemented solution knowledge graph.

[0056] The predictive model is trained based on historical interference data and its corresponding solution labels. The historical interference data encompasses feature information from various interference scenarios, such as interference frequency, intensity, and duration, while the solution labels record effective measures taken to address these interferences. Through learning from a large amount of historical data, the predictive model can determine the potential mapping relationship between interference features and solutions, thereby providing solution prediction results for new interference data.

[0057] Based on a trained prediction model, low-altitude network interference data is predicted, and the predicted solution results are output. It should be noted that the output predicted solution results are not necessarily recorded in the solution knowledge graph, but are obtained based on a prediction model trained on historical data.

[0058] Calculate the semantic similarity between the predicted solution results and each node in the solution knowledge graph. The predicted results and the text of the knowledge graph nodes can be converted into vector representations, and then algorithms such as cosine similarity can be used to measure the degree of semantic association between them.

[0059] Based on semantic similarity, multiple inferential associations related to low-altitude network interference data are determined. When the semantic similarity between a prediction result and a knowledge graph node exceeds a set threshold, an inferential association is considered to exist between them.

[0060] These reasoning relationships are added to the solution base knowledge graph. This expands the knowledge graph's content and enhances its reasoning capabilities, enabling it to provide solutions more quickly and accurately when faced with new disruptive situations.

[0061] The process of expanding the content in the solution base knowledge graph can be as follows: Figure 2 The schematic diagram of the solution library knowledge graph expansion process provided by this invention illustrates the following steps: Wireless data is collected through a low-altitude network monitoring platform, from which entities and relationships are extracted. These extracted information are used to construct three knowledge graphs: an interference data knowledge graph, a solution knowledge graph, and a processing effect knowledge graph. These are then merged into a solution library knowledge graph, integrating multi-source knowledge to form a more comprehensive knowledge system. New relationships are generated through reasoning to refine the knowledge graph, uncovering potential knowledge and improving its richness and accuracy. Finally, based on the expanded solution library knowledge graph, services such as intelligent retrieval, recommendation, question answering, and network optimization support are provided.

[0062] In one embodiment, after constructing multiple knowledge graphs based on the entities and relationships identified in the wireless side data, the method further includes: performing semantic analysis on the nodes of each knowledge graph to determine synonym nodes in the nodes of each knowledge graph; and deleting the synonym nodes.

[0063] After constructing multiple knowledge graphs, semantic analysis of each knowledge graph node can be performed to identify and delete synonymous nodes, thereby improving the quality and efficiency of the knowledge graph.

[0064] We perform deep analysis on the text of each node in the knowledge graph to extract its semantic features, such as word meaning and syntactic structure. We then use a word vector model to convert the node text into vector representations, which facilitates subsequent similarity calculations.

[0065] Based on semantic similarity algorithms (such as cosine similarity), the semantic similarity between nodes is calculated. A reasonable similarity threshold is set, and when the semantic similarity between two nodes exceeds the threshold, they are determined to be synonymous nodes.

[0066] The identified synonym nodes are deleted to eliminate redundant information in the knowledge graph, reduce data storage, and improve query efficiency and reasoning accuracy.

[0067] In one embodiment, multiple knowledge graphs are constructed based on entities and relationships identified in the wireless side data, including: constructing an interference data knowledge graph based on entities and relationships identified in the low-altitude network interference data and the interference source data; constructing a solution knowledge graph based on entities and relationships identified in the solution data; and constructing a processing effect knowledge graph based on entities and relationships identified in the processing effect data.

[0068] This approach constructs a knowledge graph of interference data by identifying entities and relationships based on low-altitude network interference data and interference source data. For example, it can clearly identify the entity of interference type (such as same-channel interference, adjacent-channel interference, etc.) and its relationship with interference sources (such as specific base stations, drones, etc.), and can also display attributes such as the time and location of the interference, providing a foundation for subsequent analysis.

[0069] By extracting entities and relationships from solution data to construct a solution knowledge graph, it is possible to systematically organize response strategies. Solution entities such as "filter deployment" and "power adjustment," as well as their compatibility with different types of interference, can be intuitively presented in the graph, facilitating quick selection of suitable solutions.

[0070] The processing effect knowledge graph is constructed based on processing effect data and can record the effect indicators after the implementation of the solution, such as signal strength recovery and interference elimination ratio. Through the construction of these three knowledge graphs, a structured construction of the entire process of low-altitude network interference processing has been achieved.

[0071] In one embodiment, after constructing the solution library knowledge graph, the method further includes: receiving updated data from the wireless side data; updating the cross-graph association based on the updated data; and updating the solution library knowledge graph based on the updated cross-graph association.

[0072] After constructing the solution knowledge graph, a low-altitude network monitoring platform continuously receives update streams of data from the wireless side, including real-time information such as new interference events, solution optimization records, and feedback on processing effects. This data serves as the input source for the evolution of the knowledge graph, ensuring that the graph content is synchronized with the actual network status.

[0073] Iterative optimization of cross-graph associations is triggered based on updated data. For example, when a new type of interference source appears, the system automatically parses its semantic association with existing interference types and solutions, recalculates node similarity using a graph embedding algorithm, and updates cross-graph paths. If a solution proves ineffective in multiple scenarios, the connection strength between it and related interference types is weakened through the association propagation mechanism.

[0074] Based on the updated cross-graph relationships, an incremental update strategy is adopted to adjust the knowledge graph of the scheme base.

[0075] This invention also provides a low-altitude network interference analysis and processing system, such as... Figure 3 The schematic diagram of the low-altitude network interference analysis and processing system provided by the present invention is shown. The system includes a knowledge storage module, a knowledge construction module, a knowledge interaction module, a knowledge management module, and a knowledge application module.

[0076] The knowledge storage module includes ontology storage (defining ontology models of the interference domain, such as concepts and relationships of interference types, device types, environmental parameters, etc.), entity storage (storing entity data such as interference events, devices, and environmental factors), and data index (supporting fast retrieval of entities and their relationships).

[0077] The knowledge building module includes data extraction (extracting entities, attributes, and relationships from multi-source data) and knowledge fusion (eliminating data conflicts and merging heterogeneous data).

[0078] Knowledge Interaction Module: Provides visualization of interference data analysis, multi-dimensional search (such as searching by interference type, region, and time), and interactive browsing functions.

[0079] Knowledge management module: Supports knowledge modeling (custom ontology extension), knowledge editing (manual data calibration), knowledge tracing (tracking data source and processing process), version management and access control.

[0080] Knowledge Application Module: Enables intelligent retrieval (semantic interference case query), intelligent recommendation (matching solutions based on interference features), and intelligent question answering (answering questions related to interference handling).

[0081] The specific method steps implemented based on the above modules include: Data Acquisition: Collect wireless data from the low-altitude network monitoring platform, including: interference data (frequency, intensity, duration, range of impact, etc.); interference source data (equipment failure, spectrum conflict, environmental noise, third-party interference sources, etc.); solution data (parameter adjustment, equipment replacement, frequency band switching, anti-interference technology configuration, etc.); and processing effect data (interference elimination rate, signal quality improvement indicators, network performance recovery data, etc.).

[0082] Information extraction: Utilizing natural language processing (NLP) and machine learning techniques, entities (such as "drone communication module", "5G base station", "electromagnetic noise"), attributes (such as "equipment model", "interference frequency", "ambient temperature"), and relationships between entities (such as "interference is caused by equipment failure" and "solution includes parameter adjustment") are extracted from structured / unstructured data to form ontological knowledge representation.

[0083] Knowledge fusion: Through entity disambiguation (distinguishing heterogeneous entities with the same name, such as different models of "microwave transmitter") and coreference resolution (merging different expressions of the same entity, such as "base station" and "communication base station"), the extracted entities are linked to the corresponding correct objects in the knowledge base, forming an interference data knowledge graph (containing interference characteristics and associated sources), a solution knowledge graph (containing solution configuration and implementation strategies), and a treatment effect knowledge graph (containing effect indicators and influencing factors).

[0084] Graph Neural Networks (GNNs) are introduced to mine cross-graph associations between three types of graphs (e.g., "high-frequency interference → recommended frequency band switching scheme → signal strength improved by 30% after processing"), and deep fusion is used to form a scheme base knowledge graph.

[0085] Knowledge Reasoning and Application: Based on the solution base knowledge graph, new relationships (such as "the correlation between vegetation density and multipath interference in a specific area") are generated through rule-based reasoning (e.g., "when the interference intensity > 80dB and the duration > 30 minutes, the equipment inspection process is triggered first") and machine learning reasoning (predicting interference development trends and evaluating solution effectiveness), thus improving the knowledge graph. This supports cloud service providers in interference cause analysis, intelligent solution matching, processing effect prediction, and network optimization strategy formulation.

[0086] The low-altitude network interference processing apparatus provided by the present invention will be described below. The low-altitude network interference processing apparatus described below and the low-altitude network interference processing method described above can be referred to in correspondence.

[0087] like Figure 4 As shown, the device includes: Interference processing module 410 is used to retrieve interference information to be processed in the solution database knowledge graph, and determine a solution for the interference information to be processed based on the retrieved information. The method for constructing the solution base knowledge graph includes: Acquire wireless side data collected by the low-altitude network monitoring platform; the wireless side data includes low-altitude network interference data, interference source data, solution data, and processing effect data; Based on the entities and relationships identified in the wireless side data, multiple knowledge graphs are constructed; the multiple knowledge graphs include an interference data knowledge graph, a solution knowledge graph, and a processing effect knowledge graph. Semantic analysis is performed on the multiple knowledge graphs to determine the cross-graph associations between them, and the solution library knowledge graph is constructed based on the cross-graph associations.

[0088] The low-altitude network interference processing device provided by this invention constructs multiple knowledge graphs and achieves cross-graph associations based on collected wireless data. This establishes a connection between interference information, solutions, and processing effects, facilitating rapid problem localization and finding the optimal solution, thus improving processing efficiency. Based on the cross-graph associations of multiple knowledge graphs, a solution library knowledge graph is constructed. Using this constructed solution library knowledge graph, deep correlations between interference data and its source, solutions, and effects are mined, improving interference localization and solution matching efficiency. This enables rapid determination of solutions for the interference information to be processed, significantly shortening processing time and improving processing efficiency.

[0089] In one embodiment, the interference processing module 410 is specifically used for: The step of performing semantic analysis on the multiple knowledge graphs to determine the cross-graph associations between the multiple knowledge graphs includes: Semantic analysis is performed on the multiple knowledge graphs to determine the semantic similarity between different knowledge graph nodes in the multiple knowledge graphs; Based on the semantic similarity between the different knowledge graph nodes, cross-graph paths between the different knowledge graph nodes are generated, and the cross-graph paths are used as the cross-graph association relationships.

[0090] In one embodiment, the interference processing module 410 is further configured to: The step of performing semantic analysis on the multiple knowledge graphs to determine the semantic similarity between different knowledge graph nodes includes: Based on graph embedding technology, the nodes of each knowledge graph in the multiple knowledge graphs are mapped to low-dimensional vectors to obtain multiple low-dimensional vector representations; The multiple low-dimensional vector representations are projected onto the same vector space, and the semantic similarity of the multiple low-dimensional vectors is calculated in the vector space to determine the semantic similarity between different knowledge graph nodes in the multiple knowledge graphs.

[0091] In one embodiment, the interference processing module 410 is further configured to: The construction of the solution base knowledge graph based on the cross-graph association relationship includes: Based on the cross-graph paths of the different knowledge graph nodes, multiple cross-graph relation triples are constructed; Based on the multiple cross-graph relation triples, the solution base knowledge graph is constructed.

[0092] In one embodiment, the interference processing module 410 is further configured to: After constructing the solution library knowledge graph, the process also includes: The low-altitude network interference data is input into the prediction model to obtain the solution prediction result of the low-altitude network interference data output by the prediction model. The prediction model is trained based on historical interference data and the corresponding solution labels for the historical interference data. Based on the semantic similarity between the prediction results of the solution and each node in the solution knowledge graph, multiple inference associations associated with the low-altitude network interference data are determined. Based on the multiple reasoning relationships, the solution base knowledge graph is supplemented to obtain a supplemented solution base knowledge graph.

[0093] In one embodiment, the interference processing module 410 is further configured to: After constructing multiple knowledge graphs based on the entities and relationships identified in the wireless side data, the process further includes: Semantic analysis is performed on the nodes of each knowledge graph in the plurality of knowledge graphs to determine the synonym nodes in the nodes of each knowledge graph; Delete the synonym node.

[0094] In one embodiment, the interference processing module 410 is further configured to: Based on the entities and relationships identified in the wireless data, multiple knowledge graphs are constructed, including: Based on the low-altitude network interference data and the entities and relationships identified in the interference source data, an interference data knowledge graph is constructed. Based on the entities and relationships identified in the solution data, a solution knowledge graph is constructed; Based on the entities and relationships identified in the processing effect data, a processing effect knowledge graph is constructed.

[0095] In one embodiment, the interference processing module 410 is further configured to: After constructing the solution library knowledge graph, the process also includes: Receive the updated data from the wireless side; Based on the updated data, the cross-graph association is updated, and based on the updated cross-graph association, the solution base knowledge graph is updated.

[0096] Figure 5 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 5 As shown, the electronic device may include a processor 510, a communications interface 520, a memory 530, and a communication bus 540, wherein the processor 510, communications interface 520, and memory 530 communicate with each other via the communication bus 540. The processor 510 can invoke logical instructions in the memory 530 to execute a method for handling low-altitude network interference. This method includes: retrieving interference information to be processed from a solution database knowledge graph, and determining a solution for the interference information to be processed based on the obtained retrieval information. The method for constructing the solution base knowledge graph includes: Acquire wireless side data collected by the low-altitude network monitoring platform; the wireless side data includes low-altitude network interference data, interference source data, solution data, and processing effect data; Based on the entities and relationships identified in the wireless side data, multiple knowledge graphs are constructed; the multiple knowledge graphs include an interference data knowledge graph, a solution knowledge graph, and a processing effect knowledge graph. Semantic analysis is performed on the multiple knowledge graphs to determine the cross-graph associations between them, and the solution library knowledge graph is constructed based on the cross-graph associations.

[0097] Furthermore, the logical instructions in the aforementioned memory 530 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0098] On the other hand, the present invention also provides a computer program product, the computer program product including a computer program, the computer program being able to be stored on a non-transitory computer-readable storage medium, the computer program being executed by a processor, the computer being able to execute the low-altitude network interference processing method provided by the above methods, the method including: retrieving interference information to be processed in a solution base knowledge graph, and determining a solution for the interference information to be processed based on the obtained retrieval information; The method for constructing the solution base knowledge graph includes: Acquire wireless side data collected by the low-altitude network monitoring platform; the wireless side data includes low-altitude network interference data, interference source data, solution data, and processing effect data; Based on the entities and relationships identified in the wireless side data, multiple knowledge graphs are constructed; the multiple knowledge graphs include an interference data knowledge graph, a solution knowledge graph, and a processing effect knowledge graph. Semantic analysis is performed on the multiple knowledge graphs to determine the cross-graph associations between them, and the solution library knowledge graph is constructed based on the cross-graph associations.

[0099] In another aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a method for processing low-altitude network interference provided by the above methods, the method comprising: retrieving interference information to be processed in a solution base knowledge graph, and determining a solution for the interference information to be processed based on the retrieved information; The method for constructing the solution base knowledge graph includes: Acquire wireless side data collected by the low-altitude network monitoring platform; the wireless side data includes low-altitude network interference data, interference source data, solution data, and processing effect data; Based on the entities and relationships identified in the wireless side data, multiple knowledge graphs are constructed; the multiple knowledge graphs include an interference data knowledge graph, a solution knowledge graph, and a processing effect knowledge graph. Semantic analysis is performed on the multiple knowledge graphs to determine the cross-graph associations between them, and the solution library knowledge graph is constructed based on the cross-graph associations.

[0100] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0101] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0102] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for handling low-altitude network interference, characterized in that, include: Retrieve the interference information to be processed from the solution database knowledge graph, and determine the solution to the interference information to be processed based on the retrieved information; The method for constructing the solution base knowledge graph includes: Acquire wireless side data collected by the low-altitude network monitoring platform; the wireless side data includes low-altitude network interference data, interference source data, solution data, and processing effect data; Based on the entities and relationships identified in the wireless side data, multiple knowledge graphs are constructed; the multiple knowledge graphs include an interference data knowledge graph, a solution knowledge graph, and a processing effect knowledge graph. Semantic analysis is performed on the multiple knowledge graphs to determine the cross-graph associations between the multiple knowledge graphs, and the solution library knowledge graph is constructed based on the cross-graph associations; The step of performing semantic analysis on the multiple knowledge graphs to determine the cross-graph associations between the multiple knowledge graphs includes: Semantic analysis is performed on the multiple knowledge graphs to determine the semantic similarity between different knowledge graph nodes in the multiple knowledge graphs; Based on the semantic similarity between the different knowledge graph nodes, cross-graph paths between the different knowledge graph nodes are generated, and the cross-graph paths are used as the cross-graph association relationships. The step of performing semantic analysis on the multiple knowledge graphs to determine the semantic similarity between different knowledge graph nodes includes: Based on graph embedding technology, the nodes of each knowledge graph in the multiple knowledge graphs are mapped to low-dimensional vectors to obtain multiple low-dimensional vector representations; The multiple low-dimensional vector representations are projected onto the same vector space, and the semantic similarity of the multiple low-dimensional vectors is calculated in the vector space to determine the semantic similarity between different knowledge graph nodes in the multiple knowledge graphs.

2. The method for handling low-altitude network interference according to claim 1, characterized in that, The construction of the solution base knowledge graph based on the cross-graph association relationship includes: Based on the cross-graph paths of the different knowledge graph nodes, multiple cross-graph relation triples are constructed; Based on the multiple cross-graph relation triples, the solution base knowledge graph is constructed.

3. The method for handling low-altitude network interference according to claim 1, characterized in that, After constructing the solution base knowledge graph, the process also includes: The low-altitude network interference data is input into the prediction model to obtain the solution prediction result of the low-altitude network interference data output by the prediction model. The prediction model is trained based on historical interference data and the corresponding solution labels for the historical interference data. Based on the semantic similarity between the prediction results of the solution and each node in the solution knowledge graph, multiple inference associations associated with the low-altitude network interference data are determined. Based on the multiple reasoning relationships, the solution base knowledge graph is supplemented to obtain a supplemented solution base knowledge graph.

4. The method for handling low-altitude network interference according to claim 1, characterized in that, After constructing multiple knowledge graphs based on the entities and relationships identified in the wireless side data, the process further includes: Semantic analysis is performed on the nodes of each knowledge graph in the plurality of knowledge graphs to determine the synonym nodes in the nodes of each knowledge graph; Delete the synonym node.

5. The method for handling low-altitude network interference according to claim 1, characterized in that, Based on the entities and relationships identified in the wireless data, multiple knowledge graphs are constructed, including: Based on the low-altitude network interference data and the entities and relationships identified in the interference source data, an interference data knowledge graph is constructed. Based on the entities and relationships identified in the solution data, a solution knowledge graph is constructed; Based on the entities and relationships identified in the processing effect data, a processing effect knowledge graph is constructed.

6. The method for handling low-altitude network interference according to claim 1, characterized in that, After constructing the solution base knowledge graph, the process also includes: Receive the updated data from the wireless side; Based on the updated data, the cross-graph association is updated, and based on the updated cross-graph association, the solution base knowledge graph is updated.

7. A device for processing low-altitude network interference, characterized in that, include: An interference processing module is used to retrieve interference information to be processed from the solution database knowledge graph, and determine a solution for the interference information to be processed based on the retrieved information. The method for constructing the solution base knowledge graph includes: Acquire wireless side data collected by the low-altitude network monitoring platform; the wireless side data includes low-altitude network interference data, interference source data, solution data, and processing effect data; Based on the entities and relationships identified in the wireless side data, multiple knowledge graphs are constructed; the multiple knowledge graphs include an interference data knowledge graph, a solution knowledge graph, and a processing effect knowledge graph. Semantic analysis is performed on the multiple knowledge graphs to determine the cross-graph associations between the multiple knowledge graphs, and the solution library knowledge graph is constructed based on the cross-graph associations; The step of performing semantic analysis on the multiple knowledge graphs to determine the cross-graph associations between the multiple knowledge graphs includes: Semantic analysis is performed on the multiple knowledge graphs to determine the semantic similarity between different knowledge graph nodes in the multiple knowledge graphs; Based on the semantic similarity between the different knowledge graph nodes, cross-graph paths between the different knowledge graph nodes are generated, and the cross-graph paths are used as the cross-graph association relationships. The step of performing semantic analysis on the multiple knowledge graphs to determine the semantic similarity between different knowledge graph nodes includes: Based on graph embedding technology, the nodes of each knowledge graph in the multiple knowledge graphs are mapped to low-dimensional vectors to obtain multiple low-dimensional vector representations; The multiple low-dimensional vector representations are projected onto the same vector space, and the semantic similarity of the multiple low-dimensional vectors is calculated in the vector space to determine the semantic similarity between different knowledge graph nodes in the multiple knowledge graphs.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the low-altitude network interference processing method as described in any one of claims 1 to 6.

Images