Radio supervision method based on knowledge driving

Abstract

The invention discloses a radio supervision method based on knowledge driving, which promotes effective fusion of data of different sources by integrating real-time detection data, network data, transmitter index parameters and data of a basic database. According to the method, data islands are broken, a solid foundation is provided for comprehensively understanding and analyzing the radio environment, radio supervision instances are described by using a quintuple model and a triad model, knowledge graph data are managed and stored in combination with a graph database tool, so that the system can process and respond to real-time changes more quickly, and the real-time monitoring effect is improved. And the accuracy and the timeliness of a monitoring result are improved.

Description

Technical Field

[0001] The present invention relates to the field of radio regulation, and in particular to a knowledge-driven radio regulation method. Background Art

[0002] With the actual needs of the radio regulatory industry and the advancement of science and technology, data between existing departments or institutions often cannot be effectively shared, forming "data islands". This limits the overall understanding of the radio environment and makes it difficult to achieve comprehensive monitoring and management. The real-time detection capabilities and data analysis speed of existing systems may be limited, resulting in an inability to respond to rapidly changing electromagnetic environments in a timely manner. In addition, due to the diverse and complex sources of data, ensuring the accuracy and consistency of data is also a challenge. Traditional radio regulation relies more on manual monitoring and experience judgment, lacks automated intelligent analysis capabilities for large amounts of complex data, and is difficult to efficiently identify abnormal behavior or predict potential problems.

[0003] In summary, a knowledge-driven radio regulation approach is needed to address the shortcomings of existing technologies. Summary of the invention

[0004] In view of the deficiencies of the prior art, the present invention provides a knowledge-driven radio supervision method to solve the above problems.

[0005] To achieve the above object, the present invention provides the following technical solution: a knowledge-driven radio supervision method, comprising the following steps:

[0006] Step S1: abstract radio regulation into multiple ontologies, establish a five-tuple model, and describe radio regulation instances;

[0007] Step S2: Based on the quintuple model, further extract the triplet model to express the logical hierarchy;

[0008] Step S3: data integration, integrating the real-time detection data, network data, transmitter index parameters and data of the basic database;

[0009] Step S4: Use graph database tools to map, manage and store radio regulatory knowledge graph data;

[0010] Step S5: Scenario association verification, verifying the triplet model by monitoring and managing the scenario, and performing association analysis of the data pieces;

[0011] Step S6: monitoring and analysis, monitoring radio services, and analyzing electromagnetic environment conditions and overseas radio signals;

[0012] Step S7: Conduct compliance review of the monitored radio stations and equipment, and perform radio service application analysis.

[0013] Optionally, the quintuple model in step S1 is:

[0014] G = {E, R, P, T, S};

[0015] In the formula, E is the entity set, R is the relationship set, which indicates the association mode between entities, P is the attribute combination, which describes the characteristics of the entity or relationship, T is the timestamp, which indicates the time point or time period when the event occurs; S is the spatial information, which indicates the geographical location or geographical area where the event occurs.

[0016] Optionally, the five-tuple model in step S1 is established in the following manner:

[0017] Step A1: extract entities from the data and extract the relationships between entities;

[0018] Step A2: Attribute annotation, adding corresponding attributes to each entity and relationship;

[0019] Step A3: Time and space labeling, label all relevant events with time and space labels;

[0020] Step A4: Map the quintuples into nodes and edges in the database, and both nodes and edges carry attributes.

[0021] Optionally, the triplet model is extracted in step S2 in the following manner:

[0022] Step S21: extracting entities from the quintuple model and defining them;

[0023] Step S22: extracting relations and attributes from the quintuple model, and deriving and defining events from the relations and attributes;

[0024] Step S23: extracting time and space from the five-tuple model, fusing them, and defining them as scenes;

[0025] Step S24: construct the relationship between entities, events and scenes, and establish a triple model.

[0026] Optionally, in step S3,

[0027] Real-time detection data is the instant detection information of spectrum sensing nodes;

[0028] Network data refers to text data obtained from the public network;

[0029] Transmitter indicator data is non-real-time data of transmitter technical specifications and performance parameters;

[0030] The basic database data is quasi-real-time data for the approval of frequency stations and transmitter equipment models.

[0031] Optionally, step S4 is implemented in the following manner:

[0032] Step S41: create a node, create a node for each entity, assign a unique ID and attributes, and create a scene node to represent the monitoring or management scene;

[0033] Step S42: Establish relationship edges, define secondary relationships between entities, establish event edges, connect events and their entities and scenes, and express the occurrence and development process of events;

[0034] Step S43: Build indexes and constraints, create indexes for frequently queried attributes, and make the attributes unique through constraints.

[0035] Optionally, the scene association verification in step S5 is performed in the following manner:

[0036] Step S51: clarifying the verification objectives based on accuracy, effectiveness and scalability;

[0037] Step S52: Select representative scenarios according to monitoring and management requirements;

[0038] Step S53: construct test cases for each selected scenario, provide data sets to the model, predict test results, and evaluate the predicted results using evaluation criteria;

[0039] Step S54: according to the selected scenario, using the triple model to create an instance, define entity nodes, and describe time edges;

[0040] Step S55: Map the prepared data into the graph database to form corresponding node and edge structures, and construct a complete scene graph;

[0041] Step S56: Use algorithms to mine hidden patterns and relationships between data and identify important association rules.

[0042] Optionally, the detection and analysis in step S6 includes the following methods:

[0043] Time series analysis is used to analyze the changing trends of the electromagnetic environment in different time periods and identify possible influencing factors;

[0044] Frequency band analysis is used to analyze different frequency bands separately and count the number of foreign signals and the cumulative bandwidth in each frequency band;

[0045] Dynamic change monitoring is used to observe changes in overseas signals in a short period of time, capture sudden events, analyze changing trends over a longer period of time, and predict possible future situations.

[0046] Beneficial effects of the present invention:

[0047] 1. In the present invention, by integrating real-time detection data, network data, transmitter index parameters and basic database data, the method promotes the effective integration of data from different sources. This not only breaks the "data island", but also provides a solid foundation for comprehensive understanding and analysis of the radio environment. The five-tuple model and the three-tuple model are used to describe radio regulatory instances, and the graph database tool is combined to manage and store knowledge graph data, so that the system can process and respond to real-time changes more quickly, improving the accuracy and timeliness of monitoring results;

[0048] 2. In the present invention, technical means such as time series analysis, frequency band analysis and dynamic change monitoring are introduced to support automated intelligent analysis, so that abnormal behavior can be efficiently identified or potential problems can be predicted, and decision-making support capabilities can be enhanced. Through in-depth analysis of electromagnetic environment conditions and overseas radio signals, this method helps to discover and utilize spectrum holes, promote dynamic allocation of radio spectrum resources, and improve resource utilization efficiency;

[0049] 3. In the present invention, the scenario association verification module can help to better understand the signal interaction in complex scenarios by constructing test cases, evaluating prediction results and mining important association rules, and provide strong support for dealing with cross-border radio signal interference problems. The method adopts advanced data analysis technology and flexible model structure, making it highly adaptable. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 This is a radio regulatory framework diagram of the present invention.

[0051] Figure 2 This is a knowledge graph supervision framework diagram of the present invention.

[0052] Figure 3 This is a triple model framework diagram of the present invention. DETAILED DESCRIPTION

[0053] In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0054] like Figure 1 , 2 As shown in , a knowledge-driven radio regulation method includes the following steps:

[0055] Step S1: abstract radio regulation into multiple ontologies, establish a five-tuple model, and describe radio regulation instances;

[0056] Step S2: Based on the quintuple model, further extract the triplet model to express the logical hierarchy;

[0057] Step S3: data integration, integrating the real-time detection data, network data, transmitter index parameters and data of the basic database;

[0058] Step S4: Use graph database tools to map, manage and store radio regulatory knowledge graph data;

[0059] Step S5: Scenario association verification, verifying the triplet model by monitoring and managing the scenario, and performing association analysis of the data pieces;

[0060] Step S6: monitoring and analysis, monitoring radio services, and analyzing electromagnetic environment conditions and overseas radio signals;

[0061] Step S7: Conduct compliance review of the monitored radio stations and equipment, and perform radio service application analysis.

[0062] The quintuple model in step S1 is:

[0063] G = {E, R, P, T, S};

[0064] In the formula, E is the entity set, R is the relationship set, which indicates the association mode between entities, P is the attribute combination, which describes the characteristics of the entity or relationship, T is the timestamp, which indicates the time point or time period when the event occurs; S is the spatial information, which indicates the geographical location or geographical area where the event occurs.

[0065] The five-tuple model in step S1 is established in the following way:

[0066] Step A1: extract entities from the data and extract the relationships between entities;

[0067] Step A2: Attribute annotation, adding corresponding attributes to each entity and relationship;

[0068] Step A3: Time and space labeling, label all relevant events with time and space labels;

[0069] Step A4: Map the quintuples into nodes and edges in the database, and both nodes and edges carry attributes.

[0070] In step S2, the triplet model is extracted in the following way:

[0071] Step S21: extracting entities from the quintuple model and defining them;

[0072] Step S22: extracting relations and attributes from the quintuple model, and deriving and defining events from the relations and attributes;

[0073] Step S23: extracting time and space from the five-tuple model, fusing them, and defining them as scenes;

[0074] Step S24: construct the relationship between entities, events and scenes, and establish a triple model.

[0075] In step S3,

[0076] Real-time detection data is the instant detection information of spectrum sensing nodes;

[0077] Network data refers to text data obtained from the public network;

[0078] Transmitter indicator data is non-real-time data of transmitter technical specifications and performance parameters;

[0079] The basic database data is quasi-real-time data for the approval of frequency stations and transmitter equipment models.

[0080] Step S4 is implemented in the following manner:

[0081] Step S41: create a node, create a node for each entity, assign a unique ID and attributes, and create a scene node to represent the monitoring or management scene;

[0082] Step S42: Establish relationship edges, define secondary relationships between entities, establish event edges, connect events and their entities and scenes, and express the occurrence and development process of events;

[0083] Step S43: Build indexes and constraints, create indexes for frequently queried attributes, and make the attributes unique through constraints.

[0084] The scene association verification in step S5 is performed in the following manner:

[0085] Step S51: clarifying the verification objectives based on accuracy, effectiveness and scalability;

[0086] Step S52: Select representative scenarios according to monitoring and management requirements;

[0087] Step S53: construct test cases for each selected scenario, provide data sets to the model, predict test results, and evaluate the predicted results using evaluation criteria;

[0088] Step S54: according to the selected scenario, using the triple model to create an instance, define entity nodes, and describe time edges;

[0089] Step S55: Map the prepared data into the graph database to form corresponding node and edge structures, and construct a complete scene graph;

[0090] Step S56: Use algorithms to mine hidden patterns and relationships between data and identify important association rules.

[0091] The detection and analysis in step S6 include the following methods:

[0092] Time series analysis is used to analyze the changing trends of the electromagnetic environment in different time periods and identify possible influencing factors;

[0093] Frequency band analysis is used to analyze different frequency bands separately and count the number of foreign signals and the cumulative bandwidth in each frequency band;

[0094] Dynamic change monitoring is used to observe changes in overseas signals in a short period of time, capture sudden events, analyze changing trends over a longer period of time, and predict possible future situations.

[0095] The present invention integrates real-time detection data, network data, transmitter index parameters and basic database data, and the method promotes the effective integration of data from different sources. This not only breaks the "data island", but also provides a solid foundation for comprehensive understanding and analysis of the radio environment. The five-tuple model and the three-tuple model are used to describe radio regulatory instances, and the knowledge graph data is managed and stored in combination with graph database tools, so that the system can process and respond to real-time changes more quickly, improving the accuracy and timeliness of monitoring results.

[0096] The introduction of technical means such as time series analysis, frequency band analysis and dynamic change monitoring supports automated intelligent analysis, which can efficiently identify abnormal behaviors or predict potential problems, enhance decision-making support capabilities, and through in-depth analysis of electromagnetic environment conditions and overseas radio signals, this method helps to discover and utilize spectrum holes, promote dynamic allocation of radio spectrum resources, and improve resource utilization efficiency;

[0097] The scenario association verification module can help better understand the signal interaction in complex scenarios by constructing test cases, evaluating prediction results and mining important association rules, and provide strong support for dealing with cross-border radio signal interference problems. The method adopts advanced data analysis technology and flexible model structure, making it highly adaptable.

[0098] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modification, equivalent substitution or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A knowledge-driven radio regulation method, characterized in that: The following steps are involved: Step S1: abstract radio regulation into multiple ontologies, establish a five-tuple model, and describe radio regulation instances; Step S2: Based on the quintuple model, further extract the triplet model to express the logical hierarchy; Step S3: data integration, integrating the real-time detection data, network data, transmitter index parameters and data of the basic database; Step S4: Use graph database tools to map, manage and store radio regulatory knowledge graph data; Step S5: Scenario association verification, verifying the triplet model by monitoring and managing the scenario, and performing association analysis of the data pieces; Step S6: monitoring and analysis, monitoring radio services, and analyzing electromagnetic environment conditions and overseas radio signals; Step S7: Conduct compliance review of the monitored radio stations and equipment, and perform radio service application analysis.

2. The knowledge-driven radio regulation method according to claim 1, characterized in that: The five-tuple model in step S1 is: G = {E, R, P, T, S}; In the formula, E is the entity set, R is the relationship set, which indicates the association mode between entities, P is the attribute combination, which describes the characteristics of the entity or relationship, T is the timestamp, which indicates the time point or time period when the event occurs; S is the spatial information, which indicates the geographical location or geographical area where the event occurs.

3. The knowledge-driven radio regulation method according to claim 2 is characterized in that: The five-tuple model in step S1 is established in the following manner: Step A1: extract entities from the data and extract the relationships between entities; Step A2: Attribute annotation, adding corresponding attributes to each entity and relationship; Step A3: Time and space labeling, label all relevant events with time and space labels; Step A4: Map the quintuples into nodes and edges in the database, and both nodes and edges carry attributes.

4. The knowledge-driven radio regulation method according to claim 3 is characterized in that: The triplet model is extracted in step S2 in the following manner: Step S21: extracting entities from the quintuple model and defining them; Step S22: extracting relations and attributes from the quintuple model, and deriving and defining events from the relations and attributes; Step S23: extracting time and space from the five-tuple model, fusing them, and defining them as scenes; Step S24: construct the relationship between entities, events and scenes, and establish a triple model.

5. The knowledge-driven radio regulation method according to claim 1 is characterized in that: In the step S3, Real-time detection data is the instant detection information of spectrum sensing nodes; Network data refers to text data obtained from the public network; Transmitter indicator data is non-real-time data of transmitter technical specifications and performance parameters; The basic database data is quasi-real-time data for the approval of frequency stations and transmitter equipment models.

6. The knowledge-driven radio regulation method according to claim 1 is characterized in that: The step S4 is implemented in the following manner: Step S41: create a node, create a node for each entity, assign a unique ID and attributes, and create a scene node to represent the monitoring or management scene; Step S42: Establish relationship edges, define secondary relationships between entities, establish event edges, connect events and their entities and scenes, and express the occurrence and development process of events; Step S43: Build indexes and constraints, create indexes for frequently queried attributes, and make the attributes unique through constraints.

7. The knowledge-driven radio regulation method according to claim 1 is characterized in that: The scene association verification in step S5 is performed in the following manner: Step S51: clarifying verification goals based on accuracy, effectiveness, and scalability; Step S52: Select representative scenarios according to monitoring and management requirements; Step S53: construct test cases for each selected scenario, provide data sets to the model, predict test results, and evaluate the predicted results using evaluation criteria; Step S54: according to the selected scenario, using the triple model to create an instance, define entity nodes, and describe time edges; Step S55: Map the prepared data into the graph database to form corresponding node and edge structures, and construct a complete scene graph; Step S56: Use algorithms to mine hidden patterns and relationships between data and identify important association rules.

8. The knowledge-driven radio regulation method according to claim 1, characterized in that: The detection and analysis in step S6 include the following methods: Time series analysis is used to analyze the changing trends of the electromagnetic environment in different time periods and identify possible influencing factors; Frequency band analysis is used to analyze different frequency bands separately and count the number of foreign signals and the cumulative bandwidth in each frequency band; Dynamic change monitoring is used to observe changes in overseas signals in a short period of time, capture sudden events, analyze changing trends over a longer period of time, and predict possible future situations.

Images