Finger and control network evolution method in strong denial environment and medium

Abstract

The invention provides a command and control network evolution method in a strong rejection environment and a medium, and the method comprises the steps: carrying out the modeling of nodes and edges in an adversarial network based on a friend or foe adversarial closed-loop theory, and constructing a command and control network model; wherein the command and control network model comprises a reconnaissance node, a command node, a strike node and an enemy target node of our party, different nodes correspond to confrontation entities in different fields in the confrontation network, and edges of the connected nodes represent the own cooperative relationship or the enemy-me confrontation relationship among the different confrontation entities. The our cooperation side weight is measured by a multi-dimensional network communication index, and the friend-or-foe side weight is measured by our efficiency and enemy threat intensity; dividing the current denial level of the adversarial network according to the communication interruption range and the availability of nodes in the command and control network; and generating a command path corresponding to the denial level. According to the method, the anti-damage capability of the command and control network in a strong denial environment can be improved, and the survivability and task completion degree of a confrontation command and control system in an extreme environment are remarkably improved.

Description

Technical Field

[0001] The present invention belongs to the field of charge network evolution, and in particular relates to a charge network evolution method and medium in a strong denial environment. Background Art

[0002] In modern strong denial adversarial environments, command and control networks face challenges such as strong node heterogeneity, time-varying communication links, and complex enemy interference. Command and control system models based on the closed-loop theory of friendly-enemy adversarial networks can integrate the relationships between sensing, decision-making, influence, and enemy target nodes to calculate system capabilities. However, traditional command and control network evolution methods often use static adjacency matrices to describe node relationships. This is ineffective in representing novel adversarial relationships, such as cross-domain collaboration between adversarial entities (e.g., dynamic linkage between reconnaissance and strike nodes), collaboration between enemy target nodes (e.g., information exchange between enemy target nodes), and countermeasures by enemy target nodes against friendly entities (e.g., counter-reconnaissance actions conducted by enemy target nodes against reconnaissance nodes). Furthermore, they lack accurate quantitative modeling of dynamic parameters such as communication capacity, latency, and transmission accuracy. This limits the coordination efficiency of friendly adversarial entities in anti-interference and anti-damage scenarios, making it difficult to adapt to the complex game demands of strong denial environments. Therefore, a command and control network evolution method that integrates multi-dimensional dynamic indicators and supports cross-domain collaboration is urgently needed to enhance the anti-damage capability and system effectiveness of command and control networks in strong denial environments. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a command and control network evolution method and medium in a strong denial environment, so as to improve the anti-damage capability of the command and control network in a strong denial environment, and significantly improve the survivability and mission completion of the counter-command and control system in extreme environments.

[0004] In a first aspect, the present invention provides a method for evolving a charge network in a strong denial environment, the method comprising the following steps: Based on the closed-loop theory of friendly-enemy confrontation, the nodes and edges in the adversarial network are modeled to construct a command and control network model. The command and control network model includes enemy target nodes, reconnaissance nodes, command nodes, and strike nodes. Different nodes correspond to adversarial entities in different adversarial domains in the adversarial network. The edges connecting the nodes represent the friendly collaborative relationship or the friendly-enemy confrontation relationship between different adversarial entities. The weight of the friendly collaborative edge is measured by multidimensional network communication indicators, while the weight of the friendly-enemy confrontation edge is measured by both the friendly effectiveness and the enemy threat intensity. The current denial level of the adversarial network is classified based on the scope of communication interruption and the availability of nodes in the command and control network; Generate the charge path corresponding to the denial level; According to the accusation path, the accusation network model is evolved and updated.

[0005] Optionally, the node set of the accusation network model is represented as ;in, is a set of enemy target nodes, representing enemy confrontation entities that need to be attacked; A collection of reconnaissance nodes, representing our adversarial entities that collect intelligence about the adversarial environment and transmit information to the command nodes; A command node set represents our adversarial entity that can receive adversarial environment intelligence, generate attack plans, and command and control other nodes; Represents a set of attack nodes, representing our opposing entities that can receive and execute the attack plan.

[0006] Optionally, the attributes of our adversarial entity in the accusation network model include denial environment-specific attributes and general attributes; The common attributes of our adversarial entities include node number, node name, node type, node location, the denial level of the area where the node is located, the node's predefined rule base, the domain to which the node belongs, the impact event, the data attributes of the node's input and output, the node's damage status, and the anti-interference capability; The denial environment-specific attributes of the reconnaissance nodes in the reconnaissance node set are expressed as:

[0007] in, Indicates the The attribute set of a reconnaissance node, superscript Indicates time , The network access attribute set of the reconnaissance node, Indicates the number of network access attributes. Indicates the When a reconnaissance node detects multi-dimensional data information of the reconnaissance confrontation environment, enemy targets and enemy situation, its inherent detection success probability attribute set, an element in the set is recorded as , indicating the Reconnaissance nodes detect enemy target nodes The detection success rate when the data information is Indicates the When a reconnaissance node detects multi-dimensional data information, its false alarm rate attribute set, an element in the set is recorded as , indicating the Reconnaissance nodes detect enemy target nodes The false alarm rate when the data is Indicates the A reconnaissance node estimates the probability of the presence of high-value intelligence in the enemy target node based on the predefined rule base. An element in the set is recorded as , indicating the Reconnaissance nodes estimate enemy target nodes The probability of high-value intelligence. represents the amount of information collected by the reconnaissance node, Indicates the The set of enemy target nodes detected by the reconnaissance nodes, Indicates the The set of other scout nodes to which a scout node is connected, Indicates the A collection of command nodes connected to reconnaissance nodes; The denial environment-specific attributes of the command node are expressed as:

[0008] in, The attributes of a command node for network access, Indicates the The theoretical maximum information processing capacity of a command node is Indicates the The delay from receiving the situation to outputting the command for a command node, Indicates the The error in the assignment of combat tasks to command nodes, Indicates the The number of nodes controlled by the command node, Indicates the The maximum number of nodes that can be commanded by a command node, Indicates the A collection of reconnaissance detection nodes connected to the command node, Indicates the The collection of other command nodes connected to the command node, Indicates the A collection of strike nodes connected to command nodes; The denial environment-specific properties of the attack node are expressed as:

[0009] in, Indicates the The attributes of the attack node for network access, Indicates the The probability of damage caused by a strike node attacking other nodes, Indicates the The probability of a strike node being successfully defended against destruction or functional failure through physical defense, electronic countermeasures, or tactical maneuvers when attacked by the enemy. Indicates the The probability of successful interception of a strike node actively launching weapons to intercept and destroy the enemy's incoming targets is Indicates the The maximum attack range of the attack node; The denial environment-specific properties of the enemy target node are expressed as:

[0010]

[0011] in, Indicates the observation identification The type of enemy target node, Indicates the prediction The effective attack time of enemy target nodes, Indicates the first The intrinsic value of an enemy target node, Indicates that the observed The location of the enemy target node, Indicates the prediction The health status of enemy target nodes, Indicates the prediction The stealth performance of enemy target nodes, Indicates the The accuracy rate of enemy target nodes in identifying the signal characteristics of our nodes, Indicates the The duration of the confrontation of an enemy target node when it is subjected to our electronic countermeasures, Indicates the The response speed of an enemy target node to discover a threat, Indicates the The interception rate of enemy target nodes against our attack equipment, Indicates the The firepower intensity of the enemy target node's counterattack against our node, Indicates the The set of other enemy target nodes that an enemy target node can directly connect to, Indicates the The set of our reconnaissance detection nodes that are used for counter-reconnaissance of enemy target nodes, Indicates the The collection of our attack nodes that counter the firepower of enemy target nodes.

[0012] Optionally, the C2 network interaction relationship includes friendly collaborative relationship and enemy confrontation relationship, which can be divided into neighboring interaction and cross-domain interaction according to the combat domain; Optional friendly collaborative relationships focus on closed-loop collaboration between friendly adversarial entities, covering the basic processes of intelligence sharing, command and control, and firepower coordination, including neighboring friendly collaborative relationships and cross-domain friendly collaborative relationships; The coordination relationship between the neighboring parties includes: the reconnaissance node sends battlefield situation intelligence to the command node in the neighborhood; the reconnaissance node sends fire control-level guidance data to the strike node in the neighborhood; the command node issues task redirection instructions to the reconnaissance node in the neighborhood; the command node commands and controls the strike node in the neighborhood; the strike node feeds back the confrontation environment intelligence to the reconnaissance node in the neighborhood; the strike node reports the strike results to the command node in the neighborhood; the reconnaissance nodes in the neighborhood transmit the confrontation environment intelligence; the command nodes in the neighborhood coordinate command or the superior command node commands and controls the subordinate command node in the neighborhood; the strike nodes in the neighborhood coordinate strikes; Cross-domain friendly coordination relationships include: reconnaissance nodes sending adversarial environment intelligence to cross-domain command nodes; reconnaissance nodes sending cross-domain guidance data to cross-domain strike nodes; command nodes exercising command and control over cross-domain reconnaissance nodes; command nodes exercising command and control over cross-domain strike nodes; strike nodes feeding back adversarial environment intelligence to cross-domain reconnaissance nodes; strike nodes reporting strike results to cross-domain command nodes; cross-domain reconnaissance nodes transmitting adversarial environment intelligence; cross-domain command nodes coordinating commands or superior command nodes commanding and controlling cross-domain subordinate command nodes; cross-domain strike nodes coordinating strikes; Optionally, the enemy-friendly confrontation relationship revolves around reconnaissance countermeasures and firepower confrontation, reflecting the game-playing nature of system confrontation, including neighboring enemy-friendly confrontation relationships and cross-domain enemy-friendly confrontation relationships; Neighborhood enemy-to-enemy confrontation relationships include: our reconnaissance nodes implement electromagnetic interference on enemy target nodes in the neighborhood; our strike nodes perform damage missions on enemy target nodes in the neighborhood; enemy target nodes implement counter-reconnaissance on our reconnaissance nodes in the neighborhood; enemy target nodes implement firepower counterattacks on our strike nodes in the neighborhood; and enemy target nodes in the neighborhood conduct coordinated resistance. The cross-domain enemy-friendly confrontation relationship includes: our reconnaissance nodes implement cross-domain interference on cross-domain enemy nodes; our strike nodes perform cross-domain damage missions on cross-domain enemy target nodes; enemy target nodes implement cross-domain counter-reconnaissance on our cross-domain reconnaissance nodes; enemy target nodes implement cross-domain firepower counterattacks on our cross-domain strike nodes; and enemy target nodes conduct cross-domain coordinated resistance.

[0013] Optionally, the multi-dimensional network communication indicators include accuracy, transmission delay, reception delay, propagation delay, link load, and anti-interference capability.

[0014] Optionally, our effectiveness includes the reconnaissance success rate of reconnaissance nodes and the strike success rate of strike nodes, and the enemy threat intensity includes the anti-reconnaissance risk and fire counter-attack risk of enemy target nodes.

[0015] Optionally, the current denial level of the adversarial network is classified based on the scope of communication interruption and the availability of nodes in the command and control network, including: If the command and control network has cooperative nodes in the local topological neighborhood and the communication subnet has a neighborhood communication path that meets the real-time constraints, the current denial level of the adversarial network is classified as level I; If the command and control network has cooperative nodes within a local topological neighborhood, but its communication path relies on global relay links to achieve cross-subnet coordination, the current denial level of the adversarial network is classified as Level II; If the command and control network cannot find a cooperative node in the local topological neighborhood and needs to search across the entire domain to establish an effective command and control path, the current denial level of the adversarial network will be classified as Level III.

[0016] Optionally, generate a charge path corresponding to the denial level, including: When the denial level is level I or II, a dynamic weighting method based on node classification is used to generate a charge path in the neighborhood of the source node; When the denial level is III, the command path is generated in the entire domain based on the complex network measurement method.

[0017] Optionally, a dynamic weighting method based on node classification is used to generate a charge path in the neighborhood of the source node, including: According to the adjacency matrix of the source node, obtain the set of adjacent nodes in the neighborhood of the source node; According to the corresponding types of adjacent nodes, the node capabilities of adjacent nodes and the dynamic edge weights of the interaction relationship between the source node and the adjacent nodes are calculated in real time; the dynamic edge weights are used to measure the strength of the interaction relationship; According to the node capabilities and dynamic edge weights, the shortest path and the second shortest path are obtained by combining the Dijkstra expansion algorithm; If there is a communication path corresponding to the shortest path in the communication subnet, the shortest path is used as the command path; otherwise, it is determined whether there is a communication path corresponding to the second shortest path in the communication subnet; If there is a communication path corresponding to the second shortest path in the communication subnet, the second shortest path is used as the charge path; otherwise, after adjusting the edge weights, the process returns to the step of obtaining the adjacent nodes in the neighborhood of the source node according to the adjacency matrix of the source node.

[0018] Optionally, based on complex network measurement methods, generate command paths in the entire domain, including: According to the source node Adjacency matrix, get all cross-domain node sets directly connected to it ; The path evaluation value corresponding to the source node and each cross-domain node connected to it is calculated based on the pre-built path evaluation function; the expression of the path evaluation function is ;in, is a configurable weight coefficient that satisfies , For nodes At the moment The closeness centrality is used to measure the node The information transmission efficiency in the whole network is defined as , express The set of reachable nodes except itself, for arrive The shortest path length is calculated using the inverse of the edge weight of the adjacency matrix as the path weight and the Floyd-Warshall algorithm is used to solve it. With all reachable nodes The shortest path between For nodes At the moment The betweenness centrality is used to measure the node The degree of impact on network connectivity, which is defined as , Represents a node pair The total number of shortest paths, For passing The number of shortest paths, In the calculation, the inverse of the original edge weight of the adjacency matrix is ​​used as the path length, and the Floyd-Warshall algorithm is used to solve the shortest path distance matrix D between nodes, and the path backtracking statistics are used to calculate the path length. The number of shortest paths, for arrive Direct link quality, whose value is taken from the adjacency matrix arrive The original edge weight of Based on the Fibonacci heap-optimized Dijkstra extension algorithm and the path evaluation value, the shortest path segment is obtained. The node with the shortest path to the source node is selected from the cross-domain node set, and the global shortest command path is recursively generated. It is also determined whether there is a communication path corresponding to the shortest path in the communication subnet. If there is a communication path corresponding to the shortest path in the communication subnet, the communication path is used as the charge path; otherwise, after adjusting the edge weights, the process returns to the step of obtaining the cross-domain nodes connected to the source node according to the adjacency matrix of the source node.

[0019] In a second aspect, the present invention provides a computer-readable storage medium storing a computer program, which implements the above method when executed by a processor.

[0020] The beneficial effects of the present invention are: The present invention provides a command and control network evolution method in a strong denial environment. The edges connecting nodes in the constructed command and control network model represent the collaborative or adversarial relationships between different combat entities. Compared with traditional technologies that focus on a single command and control network architecture or the construction of an internal collaborative closed-loop process, the present invention incorporates the modeling of active countermeasure behavior of enemy target entities, further enriching the interactive dimension of adversarial entities. The multi-domain nodes of the enemy target nodes, which are integrated into the reconnaissance-decision-strike-countermeasure system, can achieve flexible combat "damage in one domain, replacement in another domain", thereby improving anti-damage capabilities. The current denial level of the adversarial network is divided according to the scope of communication interruption and the availability of command and control network nodes. By analyzing the dependency relationship between the communication network and the command and control network in a strong denial environment, command and control paths adapted to different denial levels are generated. In a strong denial environment, alternative links can be automatically generated based on real-time events (such as node failure and interference escalation). This significantly improves the timeliness and environmental adaptability of command and control path generation, enhances the anti-damage capability of the command and control network in a strong denial environment, and significantly enhances the survivability and mission completion of the adversarial command and control system in extreme environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 This is a flowchart of a method for evolving a charge network in a strong denial environment in one embodiment of the present application; Figure 2 This is a schematic diagram of the corresponding relationship between the charging network and the communication network in one embodiment of the present application; Figure 3 A schematic diagram of charge path generation in a denial level I denial environment in one embodiment of the present application; Figure 4 A schematic diagram of charge path generation in a denial level II environment in one embodiment of the present application; Figure 5 A schematic diagram of charge path generation in a denial level III environment in one embodiment of the present application. DETAILED DESCRIPTION

[0022] Aiming at the problems of low coordination efficiency and weak anti-damage capability of traditional methods in anti-interference and anti-damage scenarios, the present invention discloses a command and control network evolution method and medium in a strong denial environment, wherein the edges connecting nodes in the constructed command and control network model represent the coordination or confrontation relationship between different adversarial entities. Compared with traditional traditional technologies that focus on a single command and control network architecture or the construction of our internal collaborative closed-loop process, the present invention incorporates the active counter-action behavior modeling of enemy target entities, further enriching the interactive dimension of adversarial entities; it integrates the multi-domain nodes of enemy target nodes of reconnaissance-decision-strike-counterattack, which can realize " The system uses flexible operations such as "if one domain is damaged, another domain will fill in the gap" to improve anti-damage capabilities; the system divides the current denial level of the adversarial network according to the scope of communication interruption and the availability of command and control network nodes; by analyzing the dependency relationship between the communication network and the command and control network in a strong denial environment, it generates command and control paths that are adapted to different denial levels. In a strong denial environment, it can automatically generate alternative links based on real-time events (such as node failure and interference escalation), significantly improving the timeliness and environmental adaptability of command and control path generation, enhancing the anti-damage capability of the command and control network in a strong denial environment, and significantly improving the survivability and mission completion of the adversarial command and control system in extreme environments.

[0023] The following describes the charge network evolution method in a strong denial environment provided by the present invention.

[0024] like Figure 1 As shown, the method includes the following steps: Step 11: Model the nodes and edges in the adversarial network based on the enemy-friendly closed-loop theory to construct a command and control network model.

[0025] Among them, the command and control network model includes enemy target nodes, reconnaissance nodes, command nodes and attack nodes. Different nodes correspond to adversarial entities in different adversarial fields in the adversarial network. The edges connecting the nodes represent the friendly collaborative relationship or the enemy-enemy confrontation relationship between different adversarial entities. The weight of our collaborative edge is measured by multi-dimensional network communication indicators, and the weight of the enemy-enemy confrontation edge is measured by our effectiveness and the enemy's threat intensity.

[0026] In the embodiment of the present invention, the node set of the charge network model is represented as .in, is a set of enemy target nodes, representing enemy confrontation entities that need to be attacked; A collection of reconnaissance nodes, representing our adversarial entities (such as radars, optoelectronic detectors, etc.) that collect intelligence about the adversarial environment and transmit information to command nodes. A set of command nodes, representing our adversarial entities (such as command centers and command platforms at all levels) that can receive adversarial environment intelligence, generate attack plans, and command and control other nodes; Represents a set of attack nodes, representing our counter-attack entities (such as various forms of attack counter-attack equipment) that can receive and execute attack plans.

[0027] In a strong denial environment, the attributes of our adversarial entities in the accusation network model include denial environment-specific attributes and common attributes.

[0028] In an embodiment of the present invention, the common attributes of our antagonistic entity include node number, node name, node type, node location, denial level of the area where the node is located, node predefined rule base, node domain, impact event, node input and output data attributes (including data type, data format, encryption and decryption method, and input and output time), node collaborative gain coefficient, node damage status and anti-interference capability.

[0029] The following describes the domains to which different types of nodes correspond in the embodiments of the present invention.

[0030] Reconnaissance nodes are categorized by their deployment location, reconnaissance capabilities, and the intelligence gathering role they perform within combat missions. For example, a drone equipped with electronic reconnaissance equipment might be assigned to the electromagnetic signal reconnaissance domain, focusing on intercepting and analyzing enemy communications, radar, and other electromagnetic signals. The domain assigned to a reconnaissance node clarifies the scope and focus of its intelligence contribution within the combat system, facilitating information exchange and mission coordination with other nodes.

[0031] For command nodes, their domains are divided according to their physical deployment location, the command and control functions they undertake, and the decision-making and coordination roles they play in combat missions. For example, the node that makes decisions and commands for the overall combat situation belongs to the strategic decision-making and command. The domain to which a command node belongs reflects its command and control authority and scope in the combat system. The domain to which a command node belongs defines the boundaries of its command functions, ensuring orderly command and efficient coordination within the combat system. Strike nodes are categorized by their domains based on their deployment location, the strike capabilities of their countermeasure equipment, and the strike role they play in combat missions. A strike node's domain represents its strike mission scope and capabilities within the combat system. This domain clarifies its strike mission positioning within the combat system, facilitating the rational allocation of strike resources and achieving coordinated countermeasures and precision strikes within the combat system.

[0032] In a feasible implementation, the denial environment-specific attributes of the reconnaissance nodes in the reconnaissance node set are expressed as:

[0033] in, Indicates the The attribute set of a scout node at time t, superscript Indicates time , The network access attribute set of the reconnaissance node, Indicates the number of network access attributes. Indicates the When a reconnaissance node detects multi-dimensional data information of the reconnaissance confrontation environment, enemy targets and enemy situation, its inherent detection success probability attribute set, an element in the set is recorded as , indicating the Reconnaissance nodes detect enemy target nodes The detection success rate when the data information is Indicates the When a reconnaissance node detects multi-dimensional data information, its false alarm rate attribute set, an element in the set is recorded as , indicating the Reconnaissance nodes detect enemy target nodes The false alarm rate when the data is Indicates the A reconnaissance node estimates the probability of the presence of high-value intelligence in the enemy target node based on the predefined rule base. An element in the set is recorded as , indicating the Reconnaissance nodes estimate enemy target nodes The probability of high-value intelligence. represents the amount of information collected by the reconnaissance node, Indicates the The set of enemy target nodes detected by the reconnaissance nodes, Indicates the The set of other scout nodes to which a scout node is connected, Indicates the The set of command nodes connected to the reconnaissance nodes. For example, the number of network access attributes of the reconnaissance nodes is 8, Indicates the The network type and specific port number of each reconnaissance node, Indicates the The communication capacity of each network accessed by a reconnaissance node when processing information is used to reflect the bandwidth or data carrying capacity of the reconnaissance node when interacting with different networks when collecting intelligence in the adversarial environment; , Indicates the The scout node transmits the The rate of class data information, used to quantify the speed of the reconnaissance node output data, , Indicates the The reconnaissance node will The accuracy of class data information transmitted to other nodes is used to reflect the reliability of data transmission; , Indicates the The duration of access to this type of data information is used to reflect the The time characteristics of a reconnaissance node occupying the network when processing information; Represents a collection of information access levels, used to describe the positioning of information in the adversarial network hierarchy (such as the adversarial strategy layer, the adversarial campaign layer, and the adversarial tactical layer). It represents the actual throughput of each type of information, which means the amount of effective data actually processed or transmitted by the reconnaissance node within a certain period of time. Represents the set of reconnaissance ranges when a reconnaissance node collects intelligence about the adversarial environment.

[0034] It should be noted that, in the embodiment of the present invention, the multi-dimensional data information of the reconnaissance node on the confrontation environment, enemy targets and our situation includes geographical environment information (such as terrain obstruction, climate, etc.) and electromagnetic environment information (electromagnetic signal frequency, signal strength, etc.); enemy target node type, enemy model and various enemy status information; the location information and deployment status of our combat units, and various equipment statuses (for example, the reconnaissance node can monitor the key performance parameters of our equipment, such as the durability of the confrontation equipment, the communication distance, bandwidth, and anti-interference capability of the communication equipment, etc.).

[0035] In one feasible implementation, the denial environment-specific attributes of the command node are expressed as:

[0036] in, The attributes of a command node for network access, Indicates the The theoretical maximum information processing capacity of a command node is Indicates the The delay from receiving the situation to outputting the command for a command node, Indicates the The error in the assignment of combat tasks to command nodes, Indicates the The number of nodes controlled by the command node, Indicates the The maximum number of nodes that can be commanded by a command node, Indicates the A collection of reconnaissance detection nodes connected to the command node, Indicates the The collection of other command nodes connected to the command node, Indicates the A set of attack nodes connected to the command node. The number of attributes for a command node to access the network is 7. Indicates the type of network accessed and its specific port number. Indicates the communication capacity of each network accessed when processing information, Indicates the rate at which each type of data information is transmitted to other nodes, Indicates the accuracy of transmitting each type of data information to other nodes, Indicates the duration of access to each type of information. Indicates the level of access to each type of information, Indicates the actual throughput of each type of information.

[0037] In a feasible implementation, the denial environment-specific attributes of the attack node are expressed as:

[0038] in, Indicates the The attributes of the attack node for network access, Indicates the The probability of damage caused by a strike node attacking other nodes, Indicates the The probability of a strike node being successfully defended against destruction or functional failure through physical defense, electronic countermeasures, or tactical maneuvers when attacked by the enemy. Indicates the The probability of successful interception of a strike node actively launching weapons to intercept and destroy the enemy's incoming targets is Indicates the The maximum attack range of the attack node. The number of attributes for attacking nodes to access the network is 5. Indicates the type of network accessed and its specific port number. Indicates the communication capacity of each network for information access, Indicates the rate at which each type of data is transmitted to other attack nodes for collaboration. Indicates the accuracy rate of transmitting each type of data information to other attack nodes for collaboration, Indicates the actual throughput of each type of information.

[0039] In one feasible implementation, the denial environment-specific attributes of the enemy target node are expressed as:

[0040]

[0041] in, Indicates the observation identification The type of enemy target node, Indicates the prediction The effective attack time of enemy target nodes, Indicates the first The intrinsic value of an enemy target node, Indicates that the observed The location of the enemy target node, Indicates the prediction The health status of enemy target nodes, Indicates the prediction The stealth performance of enemy target nodes, Indicates the The accuracy rate of enemy target nodes in identifying the signal characteristics of our nodes, Indicates the The duration of the confrontation of an enemy target node when it is subjected to our electronic countermeasures, Indicates the The response speed of an enemy target node to discover a threat, Indicates the The interception rate of enemy target nodes against our attack equipment, Indicates the The firepower intensity of the enemy target node's counterattack against our node, Indicates the The set of other enemy target nodes that an enemy target node can directly connect to, Indicates the The set of our reconnaissance detection nodes that are used for counter-reconnaissance of enemy target nodes, Indicates the The collection of our attack nodes that counter the firepower of enemy target nodes.

[0042] To adapt to the new strongly denied confrontation environment involving unmanned systems, in the embodiment of the present invention, the command and control network interaction relationship includes friendly collaborative relationships and enemy confrontation relationships, which are divided into neighboring domain interactions and cross-domain interactions according to the combat domain. Among them, the friendly collaborative relationship focuses on the closed-loop collaboration between friendly and confrontational entities, covering the basic processes of intelligence sharing, command and control, and firepower coordination, including neighboring domain friendly collaborative relationships and cross-domain friendly collaborative relationships; The coordination relationship between the neighboring parties includes: the reconnaissance node sends battlefield situation intelligence to the command node in the neighborhood; the reconnaissance node sends fire control-level guidance data to the strike node in the neighborhood; the command node issues task redirection instructions to the reconnaissance node in the neighborhood; the command node commands and controls the strike node in the neighborhood; the strike node feeds back the confrontation environment intelligence to the reconnaissance node in the neighborhood; the strike node reports the strike results to the command node in the neighborhood; the reconnaissance nodes in the neighborhood transmit the confrontation environment intelligence; the command nodes in the neighborhood coordinate command or the superior command node commands and controls the subordinate command node in the neighborhood; the strike nodes in the neighborhood coordinate strikes; Cross-domain friendly coordination relationships include: reconnaissance nodes sending adversarial environment intelligence to cross-domain command nodes; reconnaissance nodes sending cross-domain guidance data to cross-domain strike nodes; command nodes exercising command and control over cross-domain reconnaissance nodes; command nodes exercising command and control over cross-domain strike nodes; strike nodes feeding back adversarial environment intelligence to cross-domain reconnaissance nodes; strike nodes reporting strike results to cross-domain command nodes; cross-domain reconnaissance nodes transmitting adversarial environment intelligence; cross-domain command nodes coordinating commands or superior command nodes commanding and controlling cross-domain subordinate command nodes; cross-domain strike nodes coordinating strikes; The enemy-friendly confrontation relationship, centered around reconnaissance countermeasures and firepower confrontation, reflects the game-playing nature of system confrontation, including neighboring enemy-friendly confrontation relationships and cross-domain enemy-friendly confrontation relationships; The neighborhood enemy-friendly confrontation relationship includes: our reconnaissance nodes implement electromagnetic interference on enemy target nodes in the neighborhood; our strike nodes perform damage missions on enemy target nodes in the neighborhood; enemy target nodes implement counter-reconnaissance on our reconnaissance nodes in the neighborhood; enemy target nodes implement firepower counterattacks on our strike nodes in the neighborhood; and enemy target nodes in the neighborhood carry out coordinated resistance.

[0043] The cross-domain enemy-friendly confrontation relationship includes: our reconnaissance nodes implement cross-domain interference on cross-domain enemy nodes; our strike nodes perform cross-domain damage missions on cross-domain enemy target nodes; enemy target nodes implement cross-domain counter-reconnaissance on our cross-domain reconnaissance nodes; enemy target nodes implement cross-domain firepower counterattacks on our cross-domain strike nodes; and enemy target nodes conduct cross-domain coordinated resistance.

[0044] It is worth mentioning that compared with the traditional relationship modeling method, in order to adapt to the new strong denial confrontation environment with the participation of unmanned systems, a series of cross-domain collaborative enhancement mechanisms and flexible relationship topologies for in-depth modeling of enemy confrontation have been constructed to achieve the dynamic reconstruction capability of "damage to this domain and replacement by that domain", providing model-level support for the system's anti-interference and anti-damage capabilities. First, the cross-domain collaborative relationship of our confrontation nodes has been increased (such as 、 、 ) to support the construction of distributed anti-destruction topology; added unmanned systems to guide the attack on the node, and then feed the intelligence back to the unmanned system for dynamic closed-loop collaboration ( ), to enhance the application of unmanned systems and shorten the strike chain; secondly, to expand the dimensions of enemy target nodes to counter our nodes (such as 、 ) to accurately quantify the interference intensity of enemy denial behavior on the combat loop, and increase the modeling of enemy coordinated defense relationships ( ) to improve the accuracy of damage effectiveness prediction.

[0045] In order to address the technical defects of traditional methods in adversarial network edge weight modeling, which only focus on single indicators (such as connectivity and delay) between our adversarial entities and lack multi-dimensional comprehensive modeling capabilities, the present invention proposes a multi-parameter integrated adversarial network edge weight correction method.

[0046] Specifically, we first extract the multi-dimensional network communication indicators of the interaction between our entities, including information transmission accuracy, sending delay, receiving delay, propagation delay, link load rate and anti-interference capability, and construct a basic edge weight evaluation model; secondly, for the enemy confrontation scenario, we introduce adversarial parameters such as the firepower countermeasure effectiveness, anti-reconnaissance effectiveness, and coordination effectiveness between enemy target nodes, and dynamically correct the enemy confrontation edge weights; finally, based on the corrected edge weights, we reconstruct the confrontation path to achieve an accurate characterization of the dynamic and time-varying characteristics of the battlefield, and provide support for the elastic evolution of the command and control network in a strong denial environment.

[0047] In a feasible embodiment, the calculation expression of the collaborative edge weight between our nodes is: ,in, Indicates the sending delay, Indicates the receiving delay, represents the propagation delay, Indicates the link load rate, Representation node The accuracy of information transmission between Representation node The anti-interference ability between them.

[0048] In one feasible implementation, For example, the transmission delay of a data m at a type S node i is .in, Indicates the data frame length.

[0049] Propagation delay For example, the propagation delay of a data m between an S-type node i and a D-type node j is:

[0050] in Representation data The speed of propagation.

[0051] Receiving delay For example, the delay of data m being received by node j after it is transmitted from node i of type S to node j of type D is: .

[0052] End-to-end transmission accuracy between two nodes For example, the probability that a data m is correctly transmitted between an S-type node i and a D-type node j is: .

[0053] Anti-interference capability of the link between two nodes Take the minimum value of the anti-interference ability attribute values ​​of these two nodes. For example, the anti-interference ability when data is transmitted between S-type node i and D-type node j is: .

[0054] In a feasible embodiment, the calculation expression of the edge weight between the enemy and friendly nodes is: ,in, , represents the enemy target node, Represents our node ability, Represents our node The collaborative gain coefficient when collaborating with similar nodes is based on the node The coefficient is preset based on the type and deployment conditions and is dynamically adjusted according to battlefield events (such as enemy interference and node damage). Represents the enemy target node Passive counter-attack capability, Representation node and The intensity of environmental interference between them.

[0055] In one feasible implementation, the scout node and enemy target nodes The edge weight calculation between the involved scout nodes Reconnaissance capability value of reconnaissance enemy target node j And the anti-reconnaissance capability value of the enemy target node j Scouting Node Ability Value Taking into account detection capability, false alarm penalties, and high-value intelligence utility gains, it is defined as: ,in and They represent the correct detection probability and false alarm rate of reconnaissance node i to enemy target node j, It represents the probability that the reconnaissance node i judges that the enemy target node j has high-value intelligence data based on the predefined rule base. Anti-reconnaissance capability value Defined as: ,in , , , is the weight coefficient and satisfies , 、 and They are respectively the stealth performance of enemy target node j, the accuracy of identifying our node signal characteristics, and the combat duration when suffering from our electronic countermeasures. Indicates the number of our reconnaissance nodes countered by the enemy target node j, Indicates the total number of our reconnaissance nodes. Indicates that in the multi-source reconnaissance fusion process, other nodes detect enemy target nodes for reconnaissance node i The synergy gain coefficient of the ability.

[0056] In one feasible implementation, attacking nodes and enemy targets The edge weight calculation between them involves the ability value of attacking node i to attack enemy target node j And the firepower counterattack capability value of the enemy target node j The ability value of attacking node i The inherent destructive capability to the target under ideal conditions and the defensive survivability and active interception self-defense capability to maintain destructive effectiveness in a confrontation environment are defined as: ,in , is the weight coefficient and satisfies , 、 and They represent the probability of attacking node i to damage enemy target node j, the probability of successful defense when attacked by enemy target node j, the probability of successful interception of incoming weapons from enemy target node j, and the maximum attack range, respectively. represents the actual distance between attack node i and enemy target node j, is an indicator function, which takes 1 (can be attacked) when the condition is met, and takes 0 (cannot be attacked) otherwise. Defined as: ,in , , , is the weight coefficient and satisfies , 、 and They are the response speed of the enemy target node j to detect threats, the success rate of intercepting our attack weapons, and the firepower intensity attribute value of counterattack against our node. Indicates the number of attack nodes countered by the enemy target node j, Indicates the number of nodes we attack. Indicates that in the coordinated attack process, other nodes attack node i to attack the enemy target node The synergistic gain coefficient of performance.

[0057] It should be noted that in the command and control network, when encountering non-preset events such as node failure, sudden interference, or sudden changes in mission requirements, the dynamic reconstruction and functional recovery of the network topology must be achieved through the collaborative mechanism of the command and control system evolution algorithm and the communication system optimization algorithm. Existing command and control network path planning technologies mostly use global search strategies or fixed neighborhood rules, which are difficult to cope with dynamic adversarial environments under different denial levels. In view of this, the present invention adopts a hierarchically triggered layered heap optimized Dijkstra expansion algorithm to achieve adaptive generation of command and control paths from the neighborhood to the entire domain, realizing the dynamic evolution of the command and control network under strong denial environments, as shown in steps 12 to 14.

[0058] Step 12: Classify the current denial level of the adversarial network based on the scope of communication interruption and the availability of nodes in the command and control network.

[0059] Specifically, if the command and control network can find available cooperative nodes in the neighborhood, and there is a neighborhood communication path in the communication subnet, the current denial level of the adversarial network is classified as level I. It should be noted that in the combat system architecture, the command and control network and the communication network are two key components, which are interdependent and work in coordination. In the embodiment of the present invention, the corresponding relationship between the command and control network and the communication network is as follows: Figure 2 As shown in the figure. The command and control network is mainly composed of S\D\T\I. Its core function is to analyze, make decisions and issue commands based on the battlefield situation information. In contrast, the communication network not only covers S\D\I, but also introduces relay communication nodes (such as Figure 2 Relay communication nodes play a crucial role in communication networks. The complexity of battlefield environments, such as terrain obstruction and electromagnetic interference, can hinder signal transmission. Relay communication nodes receive, amplify, and forward signals, expanding communication coverage and ensuring smooth communication links. This ensures accurate and efficient transmission of situational awareness information and decision-making instructions between different nodes, thereby supporting the effective implementation of various functions of the command and control network.

[0060] If the command and control network has cooperative nodes in the local topological neighborhood and the communication subnet has a neighborhood communication path that meets the real-time constraints, the current denial level of the adversarial network is classified as level I; If the command and control network has cooperative nodes within a local topological neighborhood, but its communication path relies on global relay links to achieve cross-subnet coordination, the current denial level of the adversarial network is classified as Level II; If the command and control network cannot find a cooperative node in the local topological neighborhood and needs to search across the entire domain to establish an effective command and control path, the current denial level of the adversarial network will be classified as Level III.

[0061] Step 13: Generate the charge path corresponding to the denial level.

[0062] Specifically include step 13.1 and step 13.2.

[0063] Step 13.1: When the denial level is level I or II, a charge path is generated in the neighborhood of the source node based on a dynamic weighting method of node classification.

[0064] In a feasible embodiment, as Figure 3 As shown in the figure, in a denial environment with a denial level of I, D22 in the accusation network issues an order to I2 to attack T2. At this time, D22 can find I2 in the neighborhood. D22 in the communication network can also find I2 in the neighborhood. The path for D22 to find I2 is based on the dynamic weight method of node classification. Its activation link is as follows Figure 3 Indicated by the red arrow.

[0065] In a feasible embodiment, as Figure 4 As shown in the figure, in a denial environment with a denial level of II, D22 in the command network issues an order to I2 to attack T2. At this time, D22 can find I2 in the neighborhood, while D22 in the communication network can find I2 across domains. The path for D22 to find I2 in the command network is a dynamic weight method based on node classification, and its activation link is as follows: Figure 4 Indicated by the red arrow.

[0066] Specifically, it includes steps 13.1.1 to 13.1.5.

[0067] Step 13.1.1, according to the source node Adjacency matrix, get the set of all adjacent nodes in the neighborhood of the source node .

[0068] Step 13.1.2: Dynamically calculate the adjacent nodes based on their functional types. The node capacity weight is calculated and the inverse of the dynamic edge weight of the adjacency matrix is ​​summed to generate the comprehensive evaluation function value of the path between the source node and the adjacent nodes.

[0069] In the embodiment of the present invention, the capability of the reconnaissance node is the battlefield situation awareness capability; the node capability of the command node is the command decision-making and control capability; and the node capability of the strike node is the target precision destruction capability.

[0070] In one feasible implementation, the definition and quantitative calculation of the capabilities of the three types of nodes in the network are as follows: The battlefield situational awareness capability of a reconnaissance node is represented by the reconnaissance node weight. Its value is based on the balance between the probability of target detection and the probability of false alarm, combined with high-value intelligence and collaborative utility, and is calculated according to the following formula:

[0071] in, Represents a scout node The damage status (1 is completely damaged, 0 is healthy), Represents a scout node The collaborative gain coefficient of other nodes cooperating to detect the target node j at time t (1 means that the collaboration of other nodes is particularly strong, and 0 means that no other nodes are cooperating to detect), and Represents the reconnaissance nodes at time t The correct detection probability and false alarm rate of target node j, Represents a scout node The probability of high-value intelligence data existing in target node j is determined based on the predefined rule base.

[0072] The command decision-making and control capability is represented by the weight of the command node. Its value is based on the node damage status, theoretical maximum information processing capacity, command load, decision-making speed and accuracy, and is calculated according to the following formula:

[0073] in, represents the weight coefficient of the command node load, Command Node The damage status (1 is completely damaged, 0 is healthy), Indicates the theoretical maximum information processing capacity of the command node, Indicates the number of nodes controlled by command node c2, Indicates the maximum number of nodes that a command node can command. Command Node The delay from receiving the situation to outputting the command, Command Node Errors in combat mission assignment.

[0074] The target precision destruction capability of the strike node is represented by the strike node weight, whose value is based on the destruction probability, successful defense probability, successful interception probability and effective strike range, and is calculated according to the following formula:

[0075] in , is the weight coefficient and satisfies , 、 and Represents the attack nodes The probability of damaging the enemy target node j, the probability of successfully defending the enemy target node j when attacked, the probability of successfully intercepting the incoming weapons of the enemy target node j, and the maximum attack range, Indicates the attack node The actual distance between the enemy target node j, It is an indicator function, which takes 1 (strikeable) when the condition is met, and 0 (non-strikeable) otherwise.

[0076] In an embodiment of the present invention, a feasible implementation of the comprehensive evaluation function value of the path between a node and an adjacent node in the neighborhood is specifically calculated as follows:

[0077] in is the preset coefficient, for Ability value, for and The interaction strength of The edge weight of .

[0078] Step 13.1.3: Based on the comprehensive evaluation values ​​of the paths between the node and its adjacent nodes, the shortest path and the second shortest path are obtained using the Dijkstra expansion algorithm.

[0079] In a feasible implementation, based on the comprehensive evaluation values ​​of the paths between a node and its adjacent nodes, the process of obtaining the shortest path and the second shortest path by using the Dijkstra expansion algorithm includes steps 13.1.3.1 to 13.1.3.3.

[0080] Step 13.1.3.1, initialize the data structure.

[0081] Initial distance list: For each node v in the graph, create two variables to record the shortest distance dist1[v] from the source node s to this node and the second shortest distance dist2[v]. Initially, dist1[s] = 0, and for other nodes v, dist1[v] = ∞, dist2[v] = ∞.

[0082] Initial queue: Create a priority queue (usually implemented using a min-heap) Q, and add the source node s and its distance 0 to the priority queue Q in the form of (0, s). Here, the first element of the tuple represents the distance, and the second element represents the node.

[0083] Initial access markers: Create two lists to record the predecessor nodes for each node v, namely prev1[v] (to record the predecessor node on the shortest path) and prev2[v] (to record the predecessor node on the second shortest path), both initially empty.

[0084] Step 13.1.3.2, when the priority queue Q is not empty, loop through the following steps.

[0085] Extract the top element: Extract the node u with the smallest distance and its corresponding distance d from the priority queue Q, that is, extract the top element (d, u).

[0086] Skip invalid paths: If d > dist2[u], it means that the current extracted path is neither the shortest path nor the second shortest path for node u (because there already exists a better second shortest path value), so directly skip this node and continue the next iteration.

[0087] Update adjacent nodes: For each adjacent node v of node u: Calculate the new distance newDist = d + Path1(u, v) from the source node s through node u to node v, where Path1(u, v) is obtained based on the dynamic weight path based on node classification proposed above, that is, first calculate the ability of this node according to the node attribute value, and then combine the edge weight from node u to node v.

[0088] Update the shortest distance: If newDist < dist1[v], then set dist2[v] = dist1[v], prev2[v] = prev1[v]; then update dist1[v] = newDist, prev1[v] = u. If node v is already in the priority queue Q, update its distance value; if not in the priority queue Q, add (newDist, v) to the priority queue Q.

[0089] Update the second shortest distance: If dist1[v] < newDist < dist2[v], then update dist2[v] = newDist and prev2[v] = u. If node v is in the priority queue Q, update its distance value; if not, add (newDist, v) to the priority queue Q.

[0090] Step 13.1.3.3, path backtracking.

[0091] When the priority queue Q is empty, for the enemy target node t in the accusation network: Shortest path: Backtrack through the prev1[t] list. Starting from node t, continuously find the previous node according to prev1[t] until reaching the source node s. These nodes are connected in sequence to form the shortest path from the source node s to node t.

[0092] Second shortest path: Backtrack through the prev2[t] list. Starting from node t, in the same way, find the corresponding predecessor nodes according to prev2[t] until reaching the source node s, so as to obtain the second shortest path from the source node s to node t.

[0093] Step 13.1.4, if there is a communication path corresponding to the shortest path in the communication subnet, then use the shortest path as the accusation path. Otherwise, judge whether there is a communication path corresponding to the second shortest path in the communication subnet.

[0094] Step 13.1.5, if there is a communication path corresponding to the second shortest path in the communication subnet, then use the second shortest path as the accusation path. Otherwise, after adjusting the edge weights, return to execute the step of obtaining the adjacent nodes in the neighborhood of the source node according to the adjacency matrix of the source node.

[0095] In a feasible implementation, before adjusting the edge weights in Step 13.1.5, it also includes checking data synchronization. If the data synchronization is abnormal, first perform data synchronization, and then return to judge whether there is a communication path corresponding to the second shortest path in the communication subnet; if the data synchronization is normal, then adjust the edge weights. Exemplarily, the check of data synchronization includes: confirming whether the time sources of each node in the confrontation network are accurate or whether the time synchronization error is within the allowable range.

[0096] Step 13.2, when the denial level is level III, generate an accusation path in the whole domain based on the complex network measurement method.

[0097] In a feasible embodiment, such as Figure 5As shown in the figure, in a denial environment with a denial level of III, D22 in the command network issues an order to I2 to attack T2. At this time, D22 needs to cross domains to find I2. D22 in the communication network needs to cross domains to find I2. The path for D22 in the command network to find I2 is based on the complex network measurement method. Its activation link is as follows: Figure 5 Indicated by the red arrow.

[0098] Specifically, it includes steps 13.2.1 to 13.2.3.

[0099] Step 13.2.1: Calculate the path evaluation value corresponding to each cross-domain node based on the pre-built path evaluation function.

[0100] In a feasible implementation, the expression of the path evaluation function is: ;in, is a configurable weight coefficient that satisfies ; For nodes At the moment The closeness centrality, characterizing the node In terms of information transmission efficiency of the entire network, the smaller the value, the The longer the average path to other nodes, the weaker the network connectivity, which is defined as , express The set of reachable nodes of (excluding itself), for arrive The shortest path length is calculated using the inverse of the edge weight of the adjacency matrix as the path weight and the Floyd-Warshall algorithm is used to solve it. With all reachable nodes The shortest path between For nodes At the moment The betweenness centrality of The network bottleneck risk is greater. The larger the value, the more critical paths pass through the node. The greater the impact on the entire network when attacked. It is defined as , Represents a reachable node pair The total number of shortest paths, For passing The number of shortest paths, In the calculation of , the inverse of the edge weight of the adjacency matrix is ​​used as the path length, and the Floyd-Warshall algorithm is used to solve arrive The shortest path distance matrix D, and based on the path backtracking statistics The number of shortest paths; the original edge weights in the adjacency matrix The positive value represents the quality of the physical link. The larger the value, the better the link performance. , the original link quality is uniformly converted into the path cost dimension, so that this item maintains the same direction as other cost items in the function.

[0101] This path evaluation function characterizes the efficiency of network information dissemination through node proximity centrality, quantifies network structural vulnerability through node betweenness centrality, and reflects link communication reliability through the quality of links from source nodes to adjacent nodes. It comprehensively considers the global topology and link status of the command and control network. When a confrontational situation triggers dynamic evolution of the network topology (e.g., node damage or link interruption), these three indicators exhibit time-varying characteristics, providing a multi-objective optimization decision-making basis for command and control network reconstruction. Its mathematical essence is a weighted minimization problem.

[0102] By adjusting the weight coefficient in the path evaluation function , which can realize dynamic path optimization oriented to task requirements: in high-timeliness strike tasks (such as time-sensitive target elimination), increasing Weight value, give priority to the adjacent node with the largest centrality (that is, the node with the shortest average path in the whole network), shorten the end-to-end transmission delay, generate the minimum diameter subgraph, and accelerate the closure of the collaborative ring; in the anti-destruction survival mission (such as anti-saturation attack), increase Weight, avoid nodes with high betweenness centrality (i.e., key bottleneck nodes in the network), reduce the risk of single point failure, improve the network algebraic connectivity and overall anti-destruction capability; in harsh environment communication tasks (such as sandstorm areas), increase Weight value, ensuring the minimum available link quality.

[0103] In a highly denial environment, the combat situation is constantly changing. A path evaluation function calculates the path evaluation value from a source node to adjacent nodes in real time. Based on this, the Dijkstra extension algorithm, optimized with a Fibonacci heap, dynamically generates shortest and second-shortest path sets. These path sets provide specific path plans for updating the command and control network topology. Based on these plans, the network replans link connections and data transmission routes, achieving real-time topology updates and ensuring the continuous and efficient operation of the command and control network in a complex and ever-changing combat system.

[0104] Step 13.2.2, based on the Fibonacci heap optimized Dijkstra expansion algorithm and path evaluation value, obtain the shortest path segment, select the optimal node in the cross-domain node set, recursively generate the global shortest command path, and determine whether there is a communication path corresponding to the shortest path in the communication subnet.

[0105] When the traditional Dijkstra algorithm uses a normal priority queue, the time complexity is , where V is the number of nodes and E is the number of edges. After using Fibonacci heap optimization, the time complexity of Dijkstra's algorithm is reduced to The extended algorithm's time complexity for updating neighbor nodes is O(k). Each time a node's neighbors are processed, the additional time complexity for inserting or updating the distance list and for heap operations (primarily for distance list operations) is O(k). This approach remains applicable to solving the top-k shortest paths problem in large-scale networks, especially when k is relatively small, with manageable performance impact.

[0106] The following describes a process for calculating the shortest path based on the Fibonacci heap-optimized Dijkstra expansion algorithm and path evaluation values ​​in one embodiment of the present invention, specifically including steps 13.2.2.1 to 13.2.2.3.

[0107] Step 13.2.2.1, initialize the data structure.

[0108] Initial distance list: For each node v in the graph, create two variables to record the shortest distance dist1[v] and the second shortest distance dist2[v] from the source node s to the node. Initially, dist1[s]=0. For other nodes v, dist1[v]=∞ and dist2[v]=∞. Initial queue: Create a priority queue Q (implemented using a Fibonacci heap) and add the source node s and its shortest distance 0 to the priority queue Q in the form of (s, 1, 0). The first element of the tuple represents the node, the second element represents the type of path, i.e., the shortest path (1) or the second shortest path (2), and the third element is the key value of the second element, i.e., the distance corresponding to the path type.

[0109] Initial access mark: Create two predecessor node record lists for each node v, namely prev1[v] (recording the predecessor node on the shortest path) and prev2[v] (recording the predecessor node on the second shortest path), both of which are initially empty.

[0110] Step 13.2.2.2, when the priority queue Q is not empty, loop the following steps: Take out the top element of the heap: take out the node u with the smallest distance and its corresponding distance d from the priority queue Q, that is, take out the top element of the heap.

[0111] Update adjacent nodes: For each adjacent node v of node u: Calculate the new distance newDist=d+Path(u,v) from the source node s through the node u to the node v, where Path(u,v) is obtained according to the path based on the complex network measurement method proposed above, that is, first based on the closeness centrality and betweenness centrality of node v, and then combined with the edge weight from node u to node v.

[0112] Update the shortest distance: If newDist < dist1[v], then set dist2[v] = dist1[v] and prev2[v] = prev1[v]; then update dist1[v] = newDist and prev1[v] = u. Insert (v, 1, dist1[v]) and (v, 2, dist2[v]) into the priority queue Q.

[0113] Update the second shortest distance: If dist1[v] < newDist < dist2[v], then update dist2[v] = newDist, prev2[v] = u, and insert (v, 2, dist2[v]) into the priority queue Q.

[0114] Step 13.2.2.3, when the priority queue Q is empty, for the enemy target node t in the accusation network.

[0115] Shortest path: Backtrack through the prev1[t] list. Starting from node t, continuously find the previous node according to prev1[t] until reaching the source node s. These nodes connected in sequence are the shortest path from the source node s to node t.

[0116] Step 13.2.3, if there is a communication path corresponding to the shortest path in the communication subnet, then use this communication path as the accusation path; otherwise, after adjusting the edge weights, return to execute the step of obtaining the cross-domain nodes connected to the source node according to the adjacency matrix of the source node.

[0117] Step 14, evolve and update the accusation network model according to the accusation path.

[0118] In the embodiment of the present invention, the evolution of the accusation network model is essentially a dynamic weighted co-evolution process of each path.

[0119] In summary, the present invention provides a command and control network evolution method for a strongly denied environment. The edges connecting nodes in the constructed command and control network model represent the collaborative or antagonistic relationships between different adversarial entities. Compared to traditional technologies that focus on a single command and control network architecture or the construction of an internal collaborative closed-loop process, the present invention incorporates the modeling of active countermeasure behaviors of enemy target entities, further enriching the interactive dimensions of adversarial entities. The multi-domain integration of enemy target nodes in the reconnaissance, decision-making, strike, and countermeasure systems enables flexible operations where "one domain is damaged and another domain fills the gap," enhancing damage resistance. The current denial level of the adversarial network is determined based on the scope of communication interruption and the availability of nodes in the command and control network. By analyzing the dependencies between the communication network and the command and control network in a strongly denied environment, command and control paths adapted to different denial levels are generated. In a strongly denied environment, alternative links can be automatically generated based on real-time events (such as node failures and interference escalation), significantly improving the timeliness and environmental adaptability of command and control path generation, enhancing the damage resistance of the command and control network in a strongly denied environment, and significantly enhancing the survivability and mission completion of the adversarial command and control system in extreme environments.

[0120] An embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps in the above-mentioned various method embodiments can be implemented.

[0121] An embodiment of the present application provides a computer program product. When the computer program product is run on a terminal device, the terminal device can implement the steps in the above-mentioned method embodiments when executing the computer program product.

[0122] Those skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of protection of the present application is limited to these examples. In line with the present application, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of different aspects of one or more embodiments of the present application as described above, which are not provided in detail for the sake of simplicity.

[0123] The one or more embodiments of this application are intended to encompass all such substitutions, modifications, and variations that fall within the broad scope of this application. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of one or more embodiments of this application should be included in the scope of protection of this application.

Claims

1. A method for evolving a charge network in a strong denial environment, characterized by: include: Based on the closed-loop theory of friendly-enemy confrontation, the nodes and edges in the adversarial network are modeled to construct a command and control network model. The command and control network model includes friendly reconnaissance nodes, command nodes, strike nodes, and enemy target nodes. Different nodes correspond to adversarial entities in different adversarial fields in the adversarial network. The edges connecting the nodes represent friendly collaborative relationships or enemy-enemy adversarial relationships between different adversarial entities. The weight of friendly collaborative edges is measured by multidimensional network communication indicators, and the weight of enemy-enemy adversarial edges is measured by both friendly effectiveness and enemy threat intensity. Classify the current denial level of the adversarial network based on the scope of communication disruption and the availability of nodes in the command and control network; Generating a charge path corresponding to the denial level; According to the accusation path, the accusation network model is evolved and updated.

2. The charge and control network evolution method according to claim 1, characterized in that: The node set of the accusation network model is represented as ;in, is a set of enemy target nodes, representing hostile entities that need to be attacked; A collection of reconnaissance nodes, representing our adversarial entities that collect intelligence about the adversarial environment and transmit information to the command nodes; A command node set represents our adversarial entity that can receive adversarial environment intelligence, generate attack plans, and command and control other nodes; Represents a set of attack nodes, representing our opposing entities that can receive and execute the attack plan.

3. The charge and control network evolution method according to claim 2, characterized in that: The attributes of our adversarial entities in the accusation network model include attributes specific to the denial environment and general attributes; The common attributes of our adversarial entity include node number, node name, node type, node location, denial level of the area where the node is located, node predefined rule base, node domain, impact event, node input and output data attributes, node damage status and anti-interference capability; The denial environment-specific attributes of the reconnaissance nodes in the reconnaissance node set are expressed as: in, Indicates the The attribute set of a reconnaissance node, superscript Indicates time , The network access attribute set of the reconnaissance node, Indicates the number of network access attributes. Indicates the When a reconnaissance node detects multi-dimensional data information of the reconnaissance confrontation environment, enemy targets and enemy situation, its inherent detection success probability attribute set, an element in the set is recorded as , indicating the Reconnaissance nodes detect enemy target nodes The detection success rate when the data information is Indicates the When a reconnaissance node detects the multi-dimensional data information, its false alarm rate attribute set, an element in the set is recorded as , indicating the Reconnaissance nodes detect enemy target nodes The false alarm rate when the data is Indicates the The reconnaissance node estimates the probability of the enemy target node having high-value intelligence based on the predefined rule base. An element in the set is recorded as , indicating the Reconnaissance nodes estimate enemy target nodes The probability of high-value intelligence. represents the amount of information collected by the reconnaissance node, Indicates the The set of enemy target nodes detected by the reconnaissance nodes, Indicates the The set of other scout nodes to which a scout node is connected, Indicates the A collection of command nodes connected to reconnaissance nodes; The denial environment-specific attributes of the command node are expressed as: in, The attributes of a command node for network access, Indicates the The theoretical maximum information processing capacity of a command node is Indicates the The delay from receiving the situation to outputting the command for a command node, Indicates the The error in the assignment of combat tasks to command nodes, Indicates the The number of nodes controlled by the command node, Indicates the The maximum number of nodes that can be commanded by a command node, Indicates the A collection of reconnaissance detection nodes connected to the command node, Indicates the The collection of other command nodes connected to the command node, Indicates the A collection of strike nodes connected to command nodes; The denial environment-specific properties of the attack node are expressed as: in, Indicates the The attributes of the attack node for network access, Indicates the The probability of damage caused by a strike node attacking other nodes, Indicates the The probability of a strike node being successfully defended against destruction or functional failure through physical defense, electronic countermeasures, or tactical maneuvers when attacked by the enemy. Indicates the The probability of successful interception of a strike node actively launching weapons to intercept and destroy the enemy's incoming targets is Indicates the The maximum attack range of the attack node, Indicates the The enemy target node attacked by the attack node, Indicates the A collection of command nodes connected to the strike nodes, Indicates the The set of other attack nodes connected to the attack node; The denial environment-specific properties of the enemy target node are expressed as: in, Indicates the observation identification The type of enemy target node, Indicates the prediction The effective attack time of enemy target nodes, Indicates the first The intrinsic value of an enemy target node, Indicates that the observed The location of the enemy target node, Indicates the prediction The health status of enemy target nodes, Indicates the prediction The stealth performance of enemy target nodes, Indicates the The accuracy rate of enemy target nodes in identifying the signal characteristics of our nodes, Indicates the The duration of the confrontation of an enemy target node when it is subjected to our electronic countermeasures, Indicates the The response speed of an enemy target node to discover a threat, Indicates the The interception rate of enemy target nodes against our attack equipment, Indicates the The firepower intensity of the enemy target node's counterattack against our node, Indicates the The set of other enemy target nodes that an enemy target node can directly connect to, Indicates the The set of our reconnaissance detection nodes that are used for counter-reconnaissance of enemy target nodes, Indicates the The collection of our attack nodes that counter the firepower of enemy target nodes.

4. The charge and control network evolution method according to claim 3, characterized in that: The interactive relationship of the command and control network includes friendly cooperative relationship and enemy confrontation relationship, which is divided into neighborhood interaction and cross-domain interaction according to the combat domain; The friendly collaborative relationship focuses on the closed-loop collaboration between friendly adversary entities, covering the basic processes of intelligence sharing, command and control, and firepower coordination, including neighboring friendly collaborative relationships and cross-domain friendly collaborative relationships; The neighborhood self-side collaborative relationship includes: reconnaissance nodes sending battlefield situation intelligence to command nodes in the neighborhood; reconnaissance nodes sending fire control-level guidance data to strike nodes in the neighborhood; command nodes issuing task redirection instructions to reconnaissance nodes in the neighborhood; command nodes commanding and controlling strike nodes in the neighborhood; strike nodes feeding back adversarial environment intelligence to reconnaissance nodes in the neighborhood; strike nodes reporting strike results to command nodes in the neighborhood; reconnaissance nodes in the neighborhood transmitting adversarial environment intelligence; command nodes in the neighborhood coordinating commands or superior command nodes commanding and controlling subordinate command nodes in the neighborhood; and strike nodes in the neighborhood coordinating strikes; The cross-domain self-side collaborative relationship includes: reconnaissance nodes sending counter-environment intelligence to cross-domain command nodes; reconnaissance nodes sending cross-domain guidance data to cross-domain strike nodes; command nodes commanding and controlling cross-domain reconnaissance nodes; command nodes commanding and controlling cross-domain strike nodes; strike nodes feeding back counter-environment intelligence to cross-domain reconnaissance nodes; strike nodes reporting strike results to cross-domain command nodes; cross-domain reconnaissance nodes transmitting counter-environment intelligence; cross-domain command nodes coordinating commands or superior command nodes commanding and controlling cross-domain subordinate command nodes; cross-domain strike nodes coordinating strikes; The enemy-friendly confrontation relationship revolves around reconnaissance countermeasures and firepower confrontation, reflecting the game nature of system confrontation, including neighboring enemy-friendly confrontation relationships and cross-domain enemy-friendly confrontation relationships; The neighborhood enemy-friendly confrontation relationship includes: our reconnaissance nodes implement electromagnetic interference on enemy target nodes in the neighborhood; our strike nodes perform damage tasks on enemy target nodes in the neighborhood; enemy target nodes implement counter-reconnaissance on our reconnaissance nodes in the neighborhood; enemy target nodes implement firepower counterattacks on our strike nodes in the neighborhood; and enemy target nodes in the neighborhood carry out coordinated resistance. The cross-domain enemy-friendly confrontation relationship includes: our reconnaissance nodes implement cross-domain interference on cross-domain enemy nodes; our strike nodes perform cross-domain damage missions on cross-domain enemy target nodes; enemy target nodes implement cross-domain counter-reconnaissance on our cross-domain reconnaissance nodes; enemy target nodes implement cross-domain firepower counterattacks on our cross-domain strike nodes; and enemy target nodes conduct cross-domain collaborative resistance.

5. The charge and control network evolution method according to claim 4, characterized in that: The multi-dimensional network communication indicators include accuracy, transmission delay, reception delay, propagation delay, link load and anti-destruction performance; the our effectiveness includes the reconnaissance success rate of the reconnaissance node and the strike success rate of the strike node; the enemy threat intensity includes the anti-reconnaissance risk and fire counter-attack risk of the enemy target node.

6. The method for evolving a command and control network according to claim 5, characterized in that: The current denial level of the adversarial network is divided into the following levels based on the scope of communication interruption and the availability of nodes in the command network: If the command and control network has cooperative nodes in the local topological neighborhood and the communication subnet has a neighborhood communication path that meets the real-time constraints, the current denial level of the adversarial network is classified as level I; If the command and control network has cooperative nodes within a local topological neighborhood, but its communication path relies on global relay links to achieve cross-subnet coordination, the current denial level of the adversarial network is classified as Level II; If the command and control network cannot find a cooperative node in the local topological neighborhood and needs to search across the entire domain to establish an effective command and control path, the current denial level of the adversarial network will be classified as Level III.

7. The method for evolving a command and control network according to claim 6, wherein: Generating a charge path corresponding to the denial level includes: When the denial level is level I or II, a dynamic weighting method based on node classification is used to generate a charge path in the neighborhood of the source node; When the denial level is III, the command path is generated in the entire domain based on the complex network measurement method.

8. The method for evolving a command and control network according to claim 7, wherein: The dynamic weight method based on node classification generates a charge path in the neighborhood of a source node, including: According to the adjacency matrix of the source node, obtain the set of adjacent nodes in the neighborhood of the source node; Calculate in real time the node capabilities of the adjacent nodes and the dynamic edge weights of the interaction relationships between the source node and the adjacent nodes based on the types of the adjacent nodes; the dynamic edge weights are used to measure the strength of the interaction relationships; Obtaining the shortest path and the second shortest path based on the node capabilities and the dynamic edge weights in combination with the Dijkstra expansion algorithm; If a communication path corresponding to the shortest path exists in the communication subnet, the shortest path is used as the charge path; otherwise, determining whether a communication path corresponding to the second shortest path exists in the communication subnet; If a communication path corresponding to the second shortest path exists in the communication subnet, the second shortest path is used as the charge path; otherwise, after adjusting the edge weights, the step of returning to the step of obtaining adjacent nodes in the neighborhood of the source node according to the adjacency matrix of the source node is performed.

9. The charge and control network evolution method according to claim 8, characterized in that: The operation of generating a charge path in the entire domain based on the complex network measurement method includes: According to the source node Adjacency matrix, get all cross-domain node sets directly connected to it ; The path evaluation value between the source node and each cross-domain node is calculated based on the pre-built path evaluation function; the expression of the path evaluation function is ;in, is a configurable weight coefficient that satisfies , For nodes At the moment The closeness centrality is defined as , express The set of reachable nodes except itself, for arrive The shortest path length is calculated using the inverse of the edge weight of the adjacency matrix as the path weight and the Floyd-Warshall algorithm is used to solve it. With all reachable nodes The shortest path between For nodes At the moment The betweenness centrality of , Represents a node pair The total number of shortest paths, For passing The number of shortest paths, In the calculation of , the inverse of the edge weight of the adjacency matrix is ​​used as the path length, and the Floyd-Warshall algorithm is used to solve arrive The shortest path distance matrix D, and based on the path backtracking statistics The number of shortest paths, for arrive Direct link quality, whose value is taken from the adjacency matrix arrive The original edge weight of Based on the Fibonacci heap optimized Dijkstra extension algorithm and the path evaluation value, the shortest path segment is obtained, the optimal node is selected from the cross-domain node set, and the global shortest command path is recursively generated, and it is determined whether there is a communication path corresponding to the shortest path in the communication subnet; If there is a communication path corresponding to the shortest path in the communication subnet, the communication path is used as the charge path; otherwise, after adjusting the edge weight, return to the step of obtaining all cross-domain nodes directly connected to the source node according to the adjacency matrix of the source node.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 9 is implemented.

Images