Network information system model based on multi-mode network environment

Abstract

The invention relates to a network information system model, and discloses a network information system model based on a multi-modal network environment, which is characterized in that the network information system model is in two-way communication with a war game deduction system through an interactive interface, and operational factor data of the war game deduction system is converted into network topology and application modality of the multi-modal network environment; comprising a parameter conversion module, a topology generation module, a modal generation module, a data acquisition module and a business evaluation module. The parameter conversion module converts a combat scene and a capability parameter into network topology data according to a military knowledge graph, and the topology generation module generates a network topology structure and applies the network topology structure to a multi-modal network environment; and the modal generation module selects a network modal according to a combat scene and a use demand, and applies the network modal to the multi-modal network element equipment. The mapping of the network information capability parameters of the combined combat scene of the war game deduction system and the network arrangement data of the multi-modal network environment is realized, and the efficient, reliable and safe combined combat network environment is obtained through virtual-real combination and twin operation of the network information capability parameters and the network arrangement data.

Description

Technical Field

[0001] The present invention relates to a network information system model, and in particular to a network information system model based on a multimodal network environment. Background Art

[0002] The cybersecurity and informatization system, or network information system, is a combat system that reshapes the information age. It is a complex system that links various combat units in the multi-dimensional battlefield space of land, sea, air, space, electricity, and the Internet. Joint operations are a multi-service, multi-domain, multi-regional, multi-task, and multi-level combat form, which puts forward higher requirements on the network information system. It needs to have multi-domain coordination, diversified communications, high reliability and security, rapid response capabilities, and intelligent management capabilities. In the joint combat mission scenario, the quality of network communication is a key factor affecting the combat mission. How to efficiently dispatch network facility resources and provide an efficient, reliable, and secure network environment is the top priority of the current joint combat scenario.

[0003] Wargame simulation is a simulation of the military war process on a map based on predetermined simulation rules. By simulating the battlefield environment, alternate decision-making and command confrontation are carried out on the envisioned military actions. Combining the joint combat scenarios of the wargame simulation system with the orchestration and deployment of the multimodal network environment is of great significance to the establishment and verification of an efficient, reliable and secure network environment.

[0004] However, there is currently a lack of a technical solution that can effectively integrate wargame simulation with multimodal network environment orchestration, which seriously restricts the optimization and verification effect of the joint combat network environment. Therefore, it is urgent to develop a new network and information system model to achieve an effective combination of wargame simulation system and multimodal network environment. Summary of the invention

[0005] The present invention aims to solve the technical problem that the prior art still lacks a technology that can effectively integrate war game simulation with multimodal network environment orchestration, which seriously restricts the optimization and verification effect of the joint combat network environment. It provides a network information system model based on the multimodal network environment, which can combine the war game simulation system with the multimodal network orchestration to generate an efficient, reliable and secure joint combat network environment.

[0006] The technical solution adopted by the present invention to solve the above technical problems is: a network information system model based on a multimodal network environment, which communicates and transmits bidirectionally with a war game simulation system for simulating combat scenarios and processes through an interactive interface, and converts the combat element data of the war game simulation system into the network topology and application mode of the multimodal network environment; the network information system model includes a parameter conversion module, a topology generation module, a mode generation module, a data acquisition module and a business evaluation module.

[0007] The parameter conversion module converts the combat scenarios and capability parameters obtained from the war game simulation system into network topology data and communication parameter data according to the military knowledge graph.

[0008] The topology generation module generates a corresponding network topology structure according to the network topology data and communication parameter data provided by the parameter conversion module, and applies the generated network topology structure to the multimodal network environment.

[0009] The modality generation module is based on the multimodal network environment applying the network topology structure, selects the network modality according to the combat scenario and usage requirement information obtained from the parameter conversion module, and applies the network modality to the corresponding multimodal network element device in the multimodal network environment.

[0010] The data acquisition module collects operating data in a multimodal network environment and feeds the collected data back to the business evaluation module and the war game simulation system.

[0011] The service evaluation module evaluates the performance of the topology structure of the multimodal network environment according to the operation data in the multimodal network environment, so as to optimize the network topology structure and mode.

[0012] The military knowledge graph is a knowledge graph that includes military knowledge, equipment, communication methods and tactics, constructed based on offline mined military field-related knowledge data combined with multimodal network characteristics.

[0013] The parameter conversion module finds the corresponding nodes and elements in the military knowledge graph based on the combat scenario description and capability parameters obtained from the war game simulation system, determines the entities and the relationships between them, quantifies the identified combat elements including facilities and communication nodes, and determines the required quantity and configuration.

[0014] The topology generation module sets the initialized network topology parameters according to the information received from the parameter conversion module, constructs a graph neural network model, sets the loss function according to the network characteristics and requirements, trains the graph neural network model, and generates a network topology.

[0015] The network topology parameters include the number of nodes, node types, number of connections, bandwidth and delay requirements.

[0016] The topology generation module uses the characteristic representation of the current network topology structure as the state, the adjustment operation of the network topology structure as the action, and adopts the reinforcement learning algorithm to select the optimal action according to the current state to adjust and optimize the network topology structure.

[0017] The modality generation module uses a machine learning model to estimate the network element modality type based on information obtained from the network environment, user needs and historical data.

[0018] The modality generation module adopts a multi-classification machine learning model, takes the relevant data of network performance and user needs as the model input, takes the network element modality type as the output, and then deploys the corresponding modality attributes on the corresponding multi-modal network element equipment.

[0019] The data collection module uses a data collection agent, a standard protocol collector, and a flow data collector to collect attack and defense data, actual installation data, and device network operation data in a multimodal network.

[0020] The business evaluation module uses network technology and tactics, joint combat communication capabilities and the impact on joint combat operations as indicators to quantify network topology and modal performance.

[0021] The beneficial effects of the present invention are as follows: the network information system model communicates bidirectionally with the war game simulation system through an interactive interface, uses the war game simulation system to simulate the joint combat scenario, and converts the simulated battlefield combat elements into the network topology and application mode of the multimodal network environment, thereby realizing the mapping of the network information capability parameters of the joint combat scenario of the war game simulation system with the network orchestration configuration data of the multimodal network environment. Through the interaction of the virtual and real combination and twin operation of the two, an efficient, reliable and secure joint combat network environment is obtained, and a reliable and secure communication environment test is provided for the joint combat scenario. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a schematic diagram of the interaction and mapping structure of the network information system model of the present invention.

[0023] Figure 2 It is a flow chart of establishing a multimodal network by the network information system model of the present invention.

[0024] Figure 3 It is a schematic diagram of the multimodal military communication knowledge graph.

[0025] Figure 4 It is a schematic diagram of network topology generation based on the graph model. DETAILED DESCRIPTION

[0026] The technical solution of the present invention is described clearly and completely below in conjunction with the accompanying drawings and specific implementation methods. The specific contents listed in the following embodiments are not limited to the technical features necessary for the technical problems to be solved by the technical solutions recorded in the claims. At the same time, the enumerated embodiments are only part of the present invention, not all embodiments.

[0027] The network information system model based on the multimodal network environment of the present invention communicates bidirectionally with the war game simulation system through an interactive interface, obtains combat element data in simulated combat scenarios and processes from the war game simulation system, and adopts a certain mapping structure to convert combat element data such as the type and quantity of air defense missile positions, reconnaissance and early warning facilities, communications, and command and control nodes into network topology data such as communication network nodes and terminal nodes in the multimodal network environment. According to the workflow of target discovery, track tracking, information communication, command and control, and implementation of strikes, combat methods such as engagement rules, real-time instruction formulation, and rule-based automated actions are converted into network communication workflows to support the generation of military application modalities in the multimodal network environment.

[0028] The network information system model includes a parameter conversion module, a topology generation module, a modality generation module, a data acquisition module and a business evaluation module.

[0029] The parameter conversion module is used for data conversion between the network information system model and the war game simulation system. The war game simulation system simulates the real battlefield environment and war layout according to the needs of specific combat scenarios. The network information system model obtains combat scenarios and capability parameters from the war game simulation system through the interactive interface, and converts them through the parameter conversion module. Among them, the combat scenario includes combat requirements: such as mission type (defense, attack, reconnaissance, etc.), combat time frame, target area and basic information of both the enemy and us; environmental information: such as geographical features (topography, climate, landform), enemy deployment, and existing resources and deployment locations of friendly forces. Capability parameters include force deployment: such as the number of troops, arms (infantry, armor, air force) and other information reflecting the overall level of combat effectiveness; resource allocation: such as the quantity and distribution of various support materials such as ammunition, fuel, logistics support, and medical resources. The network information system model also obtains historical data from the war game simulation system. For example, past combat data includes success and failure cases of historical battles, combat results, loss statistics and resource usage, providing experience support for the model. By analyzing historical data, effective tactics and strategies are extracted to help the model identify the best course of action in similar scenarios.

[0030] The parameter conversion module converts the combat scenarios and capability parameters obtained from the war game simulation system into network topology data and communication parameter data by using the military knowledge graph, and quantifies and stores the combat elements such as facilities, communications, and command and control nodes and quantities required for the combat scenarios. The military knowledge graph is a knowledge graph that includes military knowledge, equipment, communication methods, and tactics, based on offline military field-related knowledge data combined with multimodal network characteristics. For example Figure 3As shown in the figure, the knowledge entities of the military knowledge graph include military confrontation, exercise confrontation, communication equipment, modal deployment, command organization, time, location, etc. The parameter conversion module finds the corresponding nodes and elements in the military knowledge graph based on the combat scenario description and capability parameters obtained from the war game simulation system, and determines the entities and their relationships through the multi-modal military communication knowledge graph, such as switch A is deployed in location A, switch B deploys TSN mode, switch A is connected to switch C, etc. The parameter conversion module quantifies the identified combat elements including facilities, communication nodes, etc., determines the required quantity and configuration, and stores them for subsequent use.

[0031] The data acquisition module collects the operation data in the multimodal network environment and feeds the collected data back to the business evaluation module and the war game simulation system. The data acquisition module uses technologies such as data acquisition agent, standard protocol collector, and flow data collector to collect attack and defense data, actual installation data, device network flow mirroring and other device network operation data in the multimodal network, and performs random collection. The data acquisition module provides a data acquisition gateway to realize acquisition configuration, acquisition control, and data encapsulation and forwarding.

[0032] The topology generation module combines the graph neural network model and the reinforcement learning algorithm to obtain a better network topology. In the initialization phase, the parameters of the initialized network topology structure, such as the number of nodes, connection type, number of connections, bandwidth and delay requirements, are set according to the information received from the parameter conversion module. A graph neural network model is constructed, and the network topology is initialized in the graph neural network model. In the network topology generation phase, based on the initialized network topology, feature extraction and information propagation are performed through the graph neural network model, and a suitable loss function is designed according to the network characteristics and task requirements for model training, optimization of the network topology, and generation of the network topology structure. For example Figure 4As shown in the figure, there are 6 multimodal network element devices, which are initially interconnected in pairs. After being optimized by the graph model, a new network topology is formed. In the reinforcement learning optimization stage, the topology generation module uses the characteristics of the current network topology as the state and the adjustment operation of the network topology as the action. The reinforcement learning algorithm is used to select the optimal action according to the current state, execute the action to adjust the network topology, and monitor the network performance indicators (such as delay, bandwidth utilization, packet loss rate, etc.) in real time to evaluate the network state. The data acquisition system automatically records relevant performance data and uses the performance evaluation algorithm to analyze these indicators. According to the observed performance changes, the reinforcement learning agent obtains feedback and gives rewards or penalties according to the network performance to accurately reflect the effect of network topology optimization. Repeat the steps of the reinforcement learning optimization stage, and gradually optimize the network topology through continuous trial and error and learning. In each iteration, the strategy or parameters are adjusted according to the results of the previous iteration to improve the learning efficiency. Set the convergence condition, stop the iteration when the condition is met, and check whether the final network topology meets the task requirements, and make necessary fine-tuning as needed. The optimal network topology output obtained by reinforcement learning optimization is applied to the multimodal network environment. After application, performance testing and verification can be carried out, and necessary adjustments and optimizations can be made based on the test results to adapt to different network environments and task requirements.

[0033] The modality generation module is based on the multimodal network environment that applies the network topology structure. According to the combat scenarios and usage requirement information obtained from the parameter conversion module, a multi-classification model is constructed using a machine learning model. The appropriate network modality is selected through the model and applied to the corresponding multimodal network element device in the multimodal network environment.

[0034] The modality generation module uses a machine learning model to estimate the network element modality type based on information obtained from the network environment, user needs and historical data. The machine learning model uses a multi-classification machine learning model, such as a random forest, to perform feature selection and parameter adjustment through training of the collected data to achieve efficient prediction. The input feature vector X of the model is data related to network performance and user needs, including combat application scenarios, communication duration, network throughput, communication quality, transmission delay, and number of network element connections. The output category Y includes network element modality types such as GEO network modality, TSN network modality, NDN network modality, IPV4 network modality, and IPV6 network modality.

[0035] The machine learning model uses the softmax loss function, where the softmax function converts the model output into a probability distribution, and the loss function is used to measure the difference between the predicted probability distribution and the true label. The loss function is as follows:

[0036]

[0037] in, refers to the label value of the real sample, It refers to the probability value obtained after the actual output is calculated by softmax. This formula can measure the distance between the true distribution and the actual output distribution.

[0038] The modality generation module uses a machine learning model to convert input features X into corresponding modality type output Y, and deploys corresponding modality attributes on corresponding multimodal network element devices, such as multimodal network element A deploys GEO network modality, multimodal network element B deploys NDN network modality, multimodal network element C deploys TSN network modality, and other multimodal network elements deploy IP network modality. The combination of multiple selected basic network modalities constitutes the application network modality to meet different combat application scenarios and usage requirements. The application network modalities of different combat application scenarios and usage requirements can be kept in historical data to provide experience reference for machine learning models, which can be used as a reference for network deployment in similar scenarios.

[0039] The business evaluation module evaluates the performance of the multimodal network environment topology structure according to the operation data in the multimodal network environment, so as to optimize the network topology structure and mode. The data acquisition module collects network throughput, broadband occupancy rate, information transmission delay, network recovery time, communication quality level, communication network transmission delay, communication quality repair rate, command communication efficiency, inter-troop communication delay, missile and artillery shell launch time, Lanchester damage coefficient, etc. in the network data. The business evaluation module uses network technology and tactics, joint combat communication capabilities and the impact on joint combat operations as indicators to quantify network topology structure and modal performance, and adopts an evaluation method based on efficiency entropy to evaluate the network information efficiency of the collected network data information, obtain network information efficiency entropy, complete the evaluation of the multimodal network empowerment effect, and feed back the evaluation result information to the war game simulation system through the parameter conversion module, so as to re-update the combat scenario.

[0040] The above description of the specific implementation mode is only used to help understand the technical concept and core idea of ​​the present invention. Although the technical solution is described and illustrated using a specific preferred embodiment, it should not be understood as a limitation of the present invention itself. Those skilled in the art may make various changes in form and details without departing from the technical concept of the present invention. These easily conceived changes or substitutions should all be included in the protection scope of the present invention.

Claims

1. A network information system model based on a multimodal network environment, characterized by: It communicates and transmits bidirectionally with the war game simulation system for simulating combat scenarios and processes through an interactive interface, and converts the combat element data of the war game simulation system into the network topology and application mode of the multi-modal network environment; the network information system model includes a parameter conversion module, a topology generation module, a mode generation module, a data acquisition module and a business evaluation module; The parameter conversion module converts the combat scenarios and capability parameters obtained from the war game simulation system into network topology data and communication parameter data according to the military knowledge graph; The topology generation module generates a corresponding network topology structure according to the network topology data and communication parameter data provided by the parameter conversion module, and applies the generated network topology structure to the multimodal network environment; The modality generation module selects a network modality based on the multimodal network environment applying the network topology structure and the combat scenario and usage requirement information obtained from the parameter conversion module, and applies the network modality to the corresponding multimodal network element device in the multimodal network environment; The network topology information generated by the topology generation module and the communication parameter data provided by the parameter conversion module generate a network modality, and apply the generated network modality to the corresponding multimodal network element device in the multimodal network environment; The data acquisition module collects the operation data in the multimodal network environment and feeds the collected data back to the business evaluation module and the war game simulation system; The service evaluation module evaluates the performance of the topology structure of the multimodal network environment according to the operation data in the multimodal network environment, so as to optimize the network topology structure and mode.

2. A network information system model based on a multimodal network environment as claimed in claim 1, characterized in that: The military knowledge graph is a knowledge graph that includes military knowledge, equipment, communication methods and tactics, constructed based on offline mined military field-related knowledge data combined with multimodal network characteristics.

3. A network information system model based on a multimodal network environment as claimed in claim 1, characterized in that: The parameter conversion module finds the corresponding nodes and elements in the military knowledge graph based on the combat scenario description and capability parameters obtained from the war game simulation system, determines the entities and the relationships between them, quantifies the identified combat elements including facilities and communication nodes, and determines the required quantity and configuration.

4. The network information system model based on a multimodal network environment as claimed in claim 1, characterized in that: The topology generation module sets the initialized network topology parameters according to the information received from the parameter conversion module, constructs a graph neural network model, sets the loss function according to the network characteristics and requirements, trains the graph neural network model, and generates a network topology structure.

5. A network information system model based on a multimodal network environment as claimed in claim 4, characterized in that: The network topology parameters include the number of nodes, node types, number of connections, bandwidth and delay requirements.

6. A network information system model based on a multimodal network environment as claimed in claim 4, characterized in that: The topology generation module uses the characteristic representation of the current network topology structure as the state, the adjustment operation of the network topology structure as the action, and adopts a reinforcement learning algorithm to select the optimal action according to the current state to adjust and optimize the network topology structure.

7. The network information system model based on a multimodal network environment as claimed in claim 1, characterized in that: The modality generation module uses a machine learning model to estimate the network element modality type based on information obtained from the network environment, user needs and historical data.

8. A network information system model based on a multimodal network environment as claimed in claim 7, characterized in that: The modality generation module adopts a multi-classification machine learning model, takes the relevant data of network performance and user needs as the model input, takes the network element modality type as the output, and then deploys the corresponding modality attributes on the corresponding multi-modal network element equipment.

9. The network information system model based on a multimodal network environment as claimed in claim 1, characterized in that: The data collection module uses a data collection agent, a standard protocol collector, and a flow data collector to collect attack and defense data, actual installation data, and device network operation data in a multimodal network.

10. The network information system model based on a multimodal network environment as claimed in claim 1, characterized in that: The business evaluation module uses network technology and tactics, joint combat communication capabilities and the impact on joint combat operations as indicators to quantify network topology and modal performance.

Images