Simulation method and system for battlefield joint interaction of light center architecture

Abstract

The invention provides a simulation method and system for battlefield joint interaction of a light center architecture. The simulation method comprises the following steps: establishing a dynamic service cluster election mechanism, and periodically electing a cluster head node through a dynamic efficiency weight adjustment strategy; constructing a hierarchical topology discovery mechanism based on spatio-temporal correlation, and establishing a heartbeat message transmission mechanism between nodes; based on battlefield situation awareness data, constructing a deep reinforcement learning model to dynamically select a transmission path, and implementing a multi-modal routing optimization strategy; and establishing a digital twinning collaborative environment of the cross-mechanism simulation system through a remote digital fitting protocol. According to the method, a double-stage election process and an improved Byzantine fault-tolerant algorithm are introduced, and efficient election and consensus of nodes are ensured; a compressed bitmap coding technology and a space-time diagram convolutional network are adopted to improve the transmission efficiency of node state information and the topological change prediction capability; the flexibility and safety of battlefield information interaction are improved.

Description

Technical Field

[0001] The present invention relates to the field of military communication and simulation technology, and in particular to a simulation method and system for battlefield joint interaction with a light center architecture. Background Art

[0002] In the existing battlefield joint interaction technology, the load rate of the traditional central node forwarding mode exceeds 80% when the node scale exceeds 300, resulting in a sharp increase in response delay; at the same time, the decentralized architecture relies on a flooding node discovery mechanism, and the bandwidth occupancy rate exceeds 40% at a scale of 200 nodes, making it difficult to support high-definition situation data synchronization; due to the heterogeneity of protocols in simulation systems of different military services (such as DIS, HLA, DDS), the protocol conversion efficiency is less than 70%, and the cross-domain authentication takes up to 1.2 seconds, which poses obstacles to cross-institutional interconnection; the traditional AODV protocol only uses the number of hops as an optimization target, and the path survival time is less than 20 seconds in a complex electromagnetic environment, which cannot adapt to the dynamic battlefield environment.

[0003] At the same time, although there are dynamic cluster structures in existing research, there is no relevant means to solve the problems of dynamic weight adjustment and spatiotemporal topology prediction; at the same time, the research and application of protocol conversion and digital twin collaboration are relatively lacking. Therefore, a new way of battlefield joint interaction that takes into account node scale, transmission efficiency and security collaboration is urgently needed. Summary of the invention

[0004] In order to solve the above problems, the present invention proposes a simulation method and system for battlefield joint interaction with a light center architecture.

[0005] The present invention specifically provides the following technical solutions: A simulation method for battlefield joint interaction with a light center architecture is provided, and the following steps are performed for a node cluster deployed with a cross-institutional simulation system: Establish a dynamic service cluster election mechanism in the cross-institutional simulation system, and periodically elect a cluster head node by setting a dynamic performance weight adjustment strategy. The cluster head node is responsible for coordinating data interaction between multiple cross-institutional simulation nodes. Construct a hierarchical topology discovery mechanism based on spatiotemporal correlation, maintain dynamic topological relationships between cross-institutional simulation nodes, and establish a heartbeat message transmission mechanism between cross-institutional simulation nodes; Based on the battlefield situation awareness data of the cross-institutional simulation system, a deep reinforcement learning model is constructed to dynamically select the data transmission path between the cross-institutional simulation nodes. The transmission path is used for tactical command synchronization and simulation data interaction between the cross-institutional simulation nodes to implement a multimodal routing optimization strategy. The input of the deep reinforcement learning model includes: the link quality matrix between the cross-institutional simulation nodes, the load state vector of each cross-institutional simulation node, and the threat assessment index of the cross-institutional simulation node. Through remote digital testing protocols, a digital twin collaborative environment for cross-institutional simulation systems is established to support virtual-reality mapping and joint deduction of multi-institutional simulation resources.

[0006] Optionally, the service cluster election mechanism includes introducing a two-stage election process, specifically: In the pre-selection stage, the candidate nodes are quickly screened through the Bloom filter, and the filtering condition is that the misjudgment rate is less than 0.1%; In the runoff stage: the improved Byzantine Fault Tolerance algorithm BFT-PoS is used to reach consensus; Dynamic performance weight :

[0007] in, is the current computational load, is the load limit, is the available bandwidth, is the total bandwidth, is the historical average transmission delay, is the delay attenuation factor, + + =1, , , All are weight coefficients; The dynamic performance weight adjustment strategy is as follows: when the node failure rate is ≥ 5%, the topology stability weight coefficient Increase to 0.6 ≤ ≤0.8.

[0008] Optionally, the hierarchical topology discovery mechanism includes: In the basic topology layer, compressed bitmap encoding technology is used to compress the state information of cross-institutional simulation nodes to 12 bytes / node. At the same time, the rapid positioning of cross-institutional simulation nodes is achieved based on the improved Kademlia protocol, with a response time of <50ms. In the dynamic perception layer, a spatiotemporal graph convolutional network ST-GCN is constructed to predict the impact of battlefield situation changes on topology. The input feature dimension is: time step × number of nodes × 8, and the output is: topology change probability distribution; In the emergency reconstruction layer, set the topology reorganization trigger threshold in the emergency scenario: packet loss rate>15% or delay>200ms.

[0009] Optionally, the reward function of the multimodal routing optimization strategy is for:

[0010] in, For delayed rewards, is the end-to-end transmission delay of the front path, The maximum tolerable delay allowed by the system is set to 200ms. For safety rewards, is the security level of the i-th path, K is the candidate path, For load balancing rewards, and denote the standard deviation and mean of the node load vector, respectively. For cost penalty, is the energy consumption of a single data transmission, Calculate the cost for the path, is the maximum allowable energy consumption of a single task, , β, and δ are weights.

[0011] Optionally, the remote digital test protocol is specifically RDTP, and the RDTP adopts a triplet protocol conversion engine and a certificateless authentication mechanism; The digital twin collaborative environment includes five core modules: virtual-reality fusion interface, distributed sandbox verification environment, cross-institutional resource mapping engine, security audit and traceability module, and collaborative deduction engine.

[0012] The present invention also provides a battlefield joint interactive simulation system with a light center architecture, the system comprising: Dynamic cluster management subsystem, which integrates the performance evaluation model and resource scheduling engine based on federated learning; Intelligent routing subsystem, including multi-scale feature extraction module and path decision matrix generator; Digital test platform, equipped with virtual-real fusion interface and cross-domain security tunnel; The performance monitoring center is deployed with a time series database and a three-dimensional situation visualization engine.

[0013] Optionally, the dynamic cluster management subsystem further includes: Node profiling module, used to collect node characteristics, including computing power, storage margin, and security level; The fault prediction unit uses a time series prediction model based on the Transformer architecture, with a prediction window of ≥30 minutes; The elastic scaling component is used to support dynamic adjustment of the service cluster scale according to the battlefield situation and adjust the response time to <2s.

[0014] Optionally, the intelligent routing subsystem adopts a multipath transmission controller supporting 6 parallel transmission paths; the intelligent routing subsystem also adopts an adaptive coding strategy to dynamically select a coding scheme according to channel quality, including: In high signal-to-noise ratio scenarios, select Turbo coding with a bit rate of 0.8; In the medium signal-to-noise ratio scenario, select LDPC code with a code rate of 0.6; In low signal-to-noise ratio scenarios, select polar coding with a code rate of 0.4.

[0015] Optionally, the digital test platform further includes: Model fusion engine to support smooth transition from LVC to digital twin; Protocol sandbox environment, providing a set of protocol simulation test tools; Security audit module, used to record complete operation logs and support blockchain evidence storage.

[0016] Optionally, the performance monitoring center further includes: Multi-dimensional evaluation indicator system, including transmission efficiency, resource utilization, and security compliance; Real-time alarm module, with a three-level alarm mechanism, including general alarm, severe alarm, and emergency response; Retrospective analysis tool to locate performance bottlenecks based on causal reasoning models.

[0017] The present invention has the following beneficial technical effects: The present invention provides a simulation method and system for battlefield joint interaction with a light-center architecture. Through the light-center architecture design, the contradiction between the high load of the centralized architecture and the high bandwidth occupancy of the decentralized architecture is balanced for the first time, and the measured bandwidth utilization rate is greatly improved. The advantages of multi-field algorithms such as federated learning, spatiotemporal graph convolutional networks, and national secret algorithms are deeply analyzed, and they are organically integrated across fields to form an autonomous and controllable technical system. At the same time, the coupling design of the dynamic performance weight model and the multi-objective reinforcement learning routing is proposed to solve the system imbalance problem caused by a single optimization goal. This scheme supports real-time interaction among multiple entities in joint exercises, greatly reduces the delay in command transmission, and improves the flexibility and security of battlefield information interaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0019] Figure 1 A flowchart of a simulation method for battlefield joint interaction of a light center architecture provided by an embodiment of the present invention; Figure 2 A block diagram of a battlefield joint interactive simulation system with a light center architecture provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0020] The embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0021] The following describes the implementation methods of the present application through specific examples, and those skilled in the art can easily understand other advantages and effects of the present application from the contents disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The present application can also be implemented or applied through other different specific implementation methods, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present application. It should be noted that, in the absence of conflict, the following embodiments and the features in the embodiments can be combined with each other. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without making creative work belong to the scope of protection of the present application.

[0022] The purpose of the present invention is to provide a simulation method and system for battlefield joint interaction with a light center architecture, aiming to improve the accuracy and immediacy of battlefield joint interaction.

[0023] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0024] Reference Figure 1 , showing a simulation method of battlefield joint interaction of a light center architecture according to an embodiment of the present application, for a node cluster deployed with a cross-organization simulation system, performing the following steps: A dynamic service cluster election mechanism is established in the cross-institutional simulation system. The cluster head node is periodically elected by setting a dynamic performance weight adjustment strategy. The cluster head node is responsible for coordinating data interaction among multiple cross-institutional simulation nodes.

[0025] The service cluster election mechanism includes introducing a two-stage election process, specifically: In the pre-selection stage, the candidate nodes are quickly screened through Bloom filters, and the filtering condition is that the misjudgment rate is less than 0.1%; In the runoff stage: the improved Byzantine Fault Tolerance algorithm BFT-PoS is used to reach consensus.

[0026] Among them, the traditional PBFT algorithm faces two major bottlenecks in the dynamic environment of the battlefield: 1) High communication complexity: when the number of nodes is N, the message complexity is O(N²), which is difficult to expand to 2000+ nodes; 2) Static committee mechanism: the verification nodes are fixed and vulnerable to targeted attacks.

[0027] Specifically, the improved BFT-PoS adopts a dynamic staking rights mechanism. Each round of elections updates the staking amount according to the real-time performance of the nodes, and malicious nodes will be fined and their deposits will be confiscated. At the same time, a two-layer verification committee architecture and activity guarantee mechanism are adopted; SM2 / SM3 is used to replace ECDSA / SHA256, and the signing speed is increased by 40%. TEE (such as Intel SGX) is integrated to protect the committee members' private keys from being stolen, thereby improving security performance. The performance data comparison results of the improved BFT-PoS implementation compared with the traditional PBFT algorithm are shown in Table 1.

[0028] Table 1. Performance comparison results

[0029] Specifically, the dynamic performance weight :

[0030] in, is the current computational load, is the load limit, is the available bandwidth, is the total bandwidth, is the historical average transmission delay, is the delay attenuation factor, + + =1; The dynamic performance weight adjustment strategy is as follows: when the node failure rate is ≥ 5%, the topology stability weight coefficient Increase to 0.6 ≤ ≤0.8.

[0031] about , , The value of is carefully designed by combining the battlefield scene characteristics and system optimization goals. The constraints are + + =1 and , , The values ​​of are all in the range of [0,1].

[0032] For example, Strategy 1: Based on battlefield scenarios , , The value of is statically configured, as shown in Table 2.

[0033] Table 2. Static configuration parameter comparison table

[0034] Test 1: In a 200-node simulation environment, when the real-time communication scenario uses weights of (0.2, 0.3, 0.5), the end-to-end latency is reduced by 42% compared to the balanced weight scenario, and the instruction transmission success rate is increased to 98.7%.

[0035] Strategy 2: Dynamic Adaptive Adjustment Mechanism A weight adjustment function is designed for the highly dynamic and highly interfering battlefield environment. The performance weight coefficient is dynamically adjusted by real-time sensing of computing load, bandwidth status, and latency changes. , , , the weight adjustment function is specifically:

[0036] in, The current cluster computing load, that is, the comprehensive utilization of CPU + memory, with a value range of 0-1; The cluster load warning threshold, i.e. the critical value that triggers emergency adjustment, is generally 0.7-0.8; It is the load sensitivity adjustment factor, which controls the steepness of the S-curve, and is generally set at 2.5-3.5; The load emergency coefficient is the nonlinear amplification factor when the load exceeds the limit, which is generally 0.8-1.2; The minimum bandwidth required for the current task, which is dynamically calculated based on the data type; is the current available bandwidth, which is a real-time monitoring value; is the total channel bandwidth, which is a fixed value of 100MHz; The bandwidth redundancy coefficient is the ratio of reserved bandwidth to prevent sudden traffic congestion. The value is generally 0.1-0.3.

[0037] Test 2: When sudden electromagnetic interference causes a surge in latency The values ​​of the initial state are: =0.65, =55Mbps, Tdelay=120ms; ( , , ) = (0.3, 0.4, 0.3).

[0038] After the interference occurs, the electromagnetic pulse causes Tdelay to increase sharply to 280ms, triggering the delay intervention mechanism and starting to adjust ( , , ) value, the adjustment process is:

[0039] get( , , ) = (0.30, 0.22, 0.56). At this time, in the emergency intervention mode, electromagnetic interference causes a sudden increase in delay, and Tdelay increases from 120ms to 280ms. This is an extreme event and requires a quick response; therefore, in an emergency, the system prioritizes , i.e., the latency stability weight, sacrifices some normalization constraints in exchange for system survival, that is, at this time, the system allows temporary breakthrough of normalization constraints to prioritize key performance indicators, i.e., latency stability. This is a fault-tolerant mechanism in the design, and the constraints are gradually restored through subsequent steps (such as weight redistribution). For example, the system performs weight rebalancing in the next control cycle Δt=50ms, and finally obtains ( / 1.08, / 1.08, / 1.08)=(0.278,0.204,0.518).

[0040] Ultimately, the latency fluctuation dropped from ±50ms to ±21ms, a reduction of 58%, the number of topology reconstructions dropped from 12 times / minute to 5 times / minute, and the stability improved by 42%.

[0041] Strategy 3: The phased weight planning is shown in Table 3.

[0042] Table 3. Phases ( , , ) Value table

[0043] In the performance comparison experiment, the phased strategy increased the task completion rate by 37% and reduced the resource waste rate by 29% compared with the fixed weight scheme.

[0044] Among them, reinforcement learning is used to optimize the parameters. After 100 iterations of evolution, the Pareto optimal solution set is obtained, and the typical solution is ( , , ) = (0.35, 0.28, 0.37).

[0045] Build a hierarchical topology discovery mechanism based on spatiotemporal correlation, maintain the dynamic topological relationship between cross-institutional simulation nodes, and establish a heartbeat message transmission mechanism between cross-institutional simulation nodes.

[0046] The hierarchical topology discovery mechanism includes: In the basic topology layer, compressed bitmap encoding technology is used to compress the status information of cross-institutional simulation nodes to 12 bytes / node. At the same time, the rapid positioning of cross-institutional simulation nodes is achieved based on the improved Kademlia protocol, with a response time of <50ms.

[0047] The improved Kademlia protocol specifically includes: 1) automatic adjustment based on node density, reducing query hops from O(logN) to O(1), and achieving dynamic k-bucket adjustment; 2) random number challenge based on SM3 hash, with verification time <3ms; 3) application of spatiotemporal coding ID = Hash (geolocation + timestamp + device fingerprint) to support fast reconstruction of dynamic topology; 4) compared with the 64-byte coding of the traditional solution, the bitmap coding is compressed to 12 bytes / node, reducing bandwidth usage by 81%.

[0048] In the dynamic perception layer, a spatiotemporal graph convolutional network ST-GCN is constructed to predict the impact of battlefield situation changes on topology. The input feature dimension is: time step × number of nodes × 8, and the output is: probability distribution of topology changes.

[0049] The loss function of the ST-GCN model is a multi-fusion loss that uses cross entropy loss as the dominant classification task combined with KL divergence to constrain the distribution consistency of spatiotemporal attention weights.

[0050] In the experiment, we collected data from multiple exercises, including electromagnetic interference, node failure, emergency tasks, and other multi-scenario exercise data, totaling 500,000 spatiotemporal sequences, and labeled the topology change events. At the same time, we added 20% adversarial samples to improve the robustness of the ST-GCN model in interference environments. The test results are shown in Table 4.

[0051] Table 4. Comparison of measured results of ST-GCN model

[0052] In the emergency reconstruction layer, set the topology reorganization trigger threshold in the emergency scenario: packet loss rate>15% or delay>200ms.

[0053] The hierarchical heartbeat message transmission mechanism specifically includes: setting up a three-level heartbeat message structure, including: basic heartbeat layer: sending 5-byte simplified messages at a fixed period; extended status layer: dynamically triggering the sending of enhanced messages containing resource status matrices; emergency notification layer: event-driven sending of priority messages carrying battlefield situation warnings.

[0054] Based on the battlefield situation awareness data of the cross-agency simulation system, a deep reinforcement learning model is constructed to dynamically select the data transmission path between the cross-agency simulation nodes. The transmission path is used for tactical command synchronization and simulation data interaction between the cross-agency simulation nodes, and implements a multimodal routing optimization strategy. The input of the deep reinforcement learning model includes: the link quality matrix between the cross-agency simulation nodes, the load state vector of each cross-agency simulation node, and the threat assessment index of the cross-agency simulation node.

[0055] The specific steps of building a deep reinforcement learning model are: State space modeling: Define a multidimensional state vector S t = [Q, L, T, H], where the link quality matrix Q ∈ R N×N , element q ij =SNR ij -P ij loss / P max , N is the number of nodes, SNR is the signal-to-noise ratio, PlossPloss is the packet loss rate; the node load vector L[l1,l2,...,l N ], the product of computing and storage resource occupancy; threat assessment index T=[t1,t2,...,t N ], , where D1 is the electromagnetic interference intensity, D2 is the geographical exposure, D3 is the number of historical attacks, weights w1=0.5, w2=0.3, w3=0.2; the historical path state H is the sliding window recording the delay and security level of the most recent 5-hop path.

[0056] Define action space: Construct a hierarchical action space A={A path ,A encode ,A priority}; where A path For path selection, K paths are selected from the candidate path set P, K = 3, and Gumbel-Softmax sampling is used; A encode As the coding strategy, a discrete action set {Turbo code, LDPC code, polar code} is used to dynamically adjust the code rate; A priority For priority marking, set the DSCP field value of the packet.

[0057] Reward function for multimodal routing optimization strategy design:

[0058] in, For delayed rewards, is the end-to-end transmission delay of the front path, The maximum tolerable delay allowed by the system is set to 200ms. For safety rewards, is the security level of the i-th path, K is the candidate path, For load balancing rewards, and denote the standard deviation and mean of the node load vector, respectively. For cost penalty, is the energy consumption of a single data transmission, Calculate the cost for the path, is the maximum allowable energy consumption of a single task. According to multiple tests, the weight values ​​are α=0.4, β=0.3, γ=0.2, and δ=0.1. The test results are shown in Table 5.

[0059] Table 5. Test results

[0060] Through multi-objective coupling design, the problem of "optimizing a single indicator leading to system imbalance" in traditional solutions is solved.

[0061] Through remote digital testing protocols, a digital twin collaborative environment for cross-institutional simulation systems is established to support virtual-reality mapping and joint deduction of multi-institutional simulation resources.

[0062] The remote digital testing protocol is specifically RDTP (Remote Digital Testing Protocol), which uses a triplet protocol conversion engine and a certificateless authentication mechanism. The RDTP protocol implements lossless bidirectional conversion of DIS 2.0.4, HLA 1516-2010, and DDS 1.4 protocols. At the same time, the protocol adopts a pipeline processing architecture, including parsing → semantic mapping → encapsulation, and the conversion delay is ≤3ms. In addition, the protocol supports context-aware conversion. The certificateless authentication mechanism includes an identification password system based on the SM9 algorithm.

[0063] The digital twin collaborative environment includes five core modules: virtual-reality fusion interface, distributed sandbox verification environment, cross-institutional resource mapping engine, security audit and traceability module, and collaborative deduction engine.

[0064] In the virtual-reality fusion interface, the multi-resolution model adapter supports five-level granularity mapping from LVC (real-virtual-construction) to digital twins: entity level (1:1) → unit level (1:N) → system level (N:M) → battlefield level (regional mapping) → strategic level (global deduction); at the same time, the model resolution can be automatically switched according to bandwidth conditions.

[0065] The distributed sandbox verification environment has three dimensions, including spatial dimension, temporal dimension and logical dimension.

[0066] In the cross-institutional resource mapping engine, cosine similarity is used to calculate resource matching, and the ant colony algorithm is used to optimize the resource scheduling path to achieve dynamic load balancing.

[0067] A four-layer protection mechanism is designed in the security audit and traceability module, including an operation audit mechanism using blockchain evidence storage technology, a data integrity verification mechanism based on the full-link verification of the Merkle tree, a privacy protection mechanism using the Paillier algorithm, and an anomaly detection mechanism using the isolation forest combined with time series prediction technology.

[0068] The collaborative simulation engine can realize the decision-making of multiple intelligent agents including command, combat, and support, and can also realize real-time confrontation simulation.

[0069] The problems in cross-institutional simulation systems are solved through the design of lightweight protocols, resource virtualization, and intelligent decision-making.

[0070] In one embodiment of the present application, Figure 2 As shown, a battlefield joint interactive simulation system with a light center architecture is also provided, and the system includes: The dynamic cluster management subsystem is used to integrate the performance evaluation model and resource scheduling engine based on federated learning. The dynamic cluster management subsystem also includes: a node profiling module for collecting the characteristics of cross-institutional simulation nodes, including computing power, storage margin, and security level; a fault prediction unit, which uses a time series prediction model based on the Transformer architecture with a prediction window of ≥30 minutes; and an elastic scaling component to support dynamic adjustment of the service cluster size according to the battlefield situation and adjust the response time to <2s.

[0071] The intelligent routing subsystem includes a multi-scale feature extraction module and a path decision matrix generator. The intelligent routing subsystem adopts a multipath transmission controller that supports 6 parallel transmission paths; the intelligent routing subsystem also adopts an adaptive coding strategy to dynamically select a coding scheme based on channel quality, including: in high signal-to-noise ratio scenarios, Turbo code coding is selected with a code rate of 0.8; in medium signal-to-noise ratio scenarios, LDPC code coding is selected with a code rate of 0.6; in low signal-to-noise ratio scenarios, polar code coding is selected with a code rate of 0.4.

[0072] The digital configuration test platform is equipped with a virtual-reality fusion interface and a cross-domain security tunnel. The digital configuration test platform also includes: a model fusion engine to support a smooth transition from LVC to digital twins; a protocol sandbox environment that provides a DIS / HLA / DDS protocol simulation test tool set; and a security audit module to record complete operation logs and support blockchain evidence storage.

[0073] The performance monitoring center is deployed with a time series database and a three-dimensional situation visualization engine. The performance monitoring center also includes: a multi-dimensional evaluation indicator system, including transmission efficiency, resource utilization, and security compliance; a real-time alarm module, which sets a three-level warning mechanism, including general warning, severe warning, and emergency response; and a retrospective analysis tool to locate performance bottlenecks based on a causal reasoning model.

[0074] Based on the same inventive concept, according to another aspect of the present invention, an embodiment of the present invention further provides a computer device, including: at least one processor; and The memory stores a computer program that can be run on the processor, and the processor executes the steps of any one of the above simulation methods when executing the program.

[0075] Based on the same inventive concept, according to another aspect of the present invention, an embodiment of the present invention further provides a computer-readable storage medium, which stores computer program instructions. When the computer program instructions are executed by a processor, the steps of any one of the above simulation methods are performed. Finally, it should be noted that a person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the storage medium can be a disk, an optical disk, a read-only storage memory (ROM) or a random access memory (RAM), etc. The above-mentioned computer program embodiments can achieve the same or similar effects as the corresponding above-mentioned arbitrary method embodiments.

[0076] In addition, typically, the devices and equipment disclosed in the embodiments of the present invention may be various electronic terminal devices, such as mobile phones, personal digital assistants (PDAs), tablet computers (PADs), smart TVs, etc., or large terminal devices, such as servers, etc. Therefore, the protection scope disclosed in the embodiments of the present invention should not be limited to a certain type of device or equipment. The client disclosed in the embodiments of the present invention may be applied to any of the above electronic terminal devices in the form of electronic hardware, computer software, or a combination of the two. In addition, the method disclosed in the embodiment of the present invention can also be implemented as a computer program executed by a CPU, and the computer program can be stored in a computer-readable storage medium. When the computer program is executed by the CPU, the above functions defined in the method disclosed in the embodiment of the present invention are performed. In addition, the above method steps and system units may also be implemented using a controller and a computer-readable storage medium for storing a computer program that enables the controller to implement the above steps or unit functions. The above are exemplary embodiments disclosed in the present invention, but it should be noted that various changes and modifications may be made without departing from the scope disclosed in the embodiments of the present invention as defined in the claims. The functions, steps and / or actions of the method claims according to the disclosed embodiments described herein do not need to be performed in any marked order. In addition, although the elements disclosed in the embodiments of the present invention may be described or required in individual form, they may also be understood as multiple unless explicitly limited to the singular. In this specification, the same or similar parts between the various embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the embodiments described later, the description is relatively simple, and the relevant parts can be referred to the partial description of the previous embodiments.

[0077] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A simulation method for battlefield joint interaction with a light center architecture, characterized in that: For the node cluster where the cross-organization simulation system is deployed, perform the following steps: exist A dynamic service cluster election mechanism is established in the cross-institutional simulation system. The cluster head node is periodically elected by setting a dynamic performance weight adjustment strategy. The cluster head node is responsible for coordinating data interaction between multiple cross-institutional simulation nodes. Construct a hierarchical topology discovery mechanism based on spatiotemporal correlation, maintain dynamic topological relationships between cross-institutional simulation nodes, and establish a heartbeat message transmission mechanism between cross-institutional simulation nodes; Based on the battlefield situation awareness data of the cross-institutional simulation system, a deep reinforcement learning model is constructed to dynamically select the data transmission path between cross-institutional simulation nodes. The transmission path is used for tactical command synchronization and simulation data interaction between cross-institutional simulation nodes, and a multi-modal routing optimization strategy is implemented. The input of the deep reinforcement learning model includes: a link quality matrix between cross-institutional simulation nodes, a load state vector of each cross-institutional simulation node, and a threat assessment index of the cross-institutional simulation node; Through remote digital testing protocols, a digital twin collaborative environment for cross-institutional simulation systems is established to support virtual-reality mapping and joint deduction of multi-institutional simulation resources.

2. The simulation method of battlefield joint interaction with a light center architecture according to claim 1 is characterized in that: The service cluster election mechanism includes introducing a two-stage election process, specifically: In the pre-selection stage, the candidate nodes are quickly screened through the Bloom filter, and the filtering condition is that the misjudgment rate is less than 0.1%; In the runoff stage: the improved Byzantine Fault Tolerance algorithm BFT-PoS is used to reach consensus; Dynamic performance weight : in, is the current computational load, is the load limit, is the available bandwidth, is the total bandwidth, is the historical average transmission delay, is the delay attenuation factor, + + =1, , , All are weight coefficients; The dynamic performance weight adjustment strategy is as follows: when the node failure rate is ≥ 5%, the topology stability weight coefficient Increase to 0.6 ≤ ≤0.

8.

3. The simulation method of battlefield joint interaction with a light center architecture according to claim 1 is characterized in that: The hierarchical topology discovery mechanism includes: In the basic topology layer, compressed bitmap encoding technology is used to compress the state information of cross-institutional simulation nodes to 12 bytes / node. At the same time, the rapid positioning of cross-institutional simulation nodes is achieved based on the improved Kademlia protocol, with a response time of <50ms. In the dynamic perception layer, a spatiotemporal graph convolutional network ST-GCN is constructed to predict the impact of battlefield situation changes on topology. The input feature dimension is: time step × number of nodes × 8, and the output is: topology change probability distribution; In the emergency reconstruction layer, set the topology reorganization trigger threshold in the emergency scenario: packet loss rate>15% or delay>200ms.

4. The simulation method of battlefield joint interaction with a light center architecture according to claim 1 is characterized in that: The reward function of the multimodal routing optimization strategy for: in, For delayed rewards, is the end-to-end transmission delay of the front path, The maximum tolerable delay allowed by the system is set to 200ms. For safety rewards, is the security level of the i-th path, K is the candidate path, For load balancing rewards, and denote the standard deviation and mean of the node load vector, respectively. For cost penalty, is the energy consumption of a single data transmission, Calculate the cost for the path, is the maximum allowable energy consumption of a single task, , β, and δ are weights.

5. The simulation method of battlefield joint interaction with a light center architecture according to claim 1 is characterized in that: The remote digital test protocol is specifically RDTP, and the RDTP adopts a triplet protocol conversion engine and a certificateless authentication mechanism; The digital twin collaborative environment includes five core modules: virtual-reality fusion interface, distributed sandbox verification environment, cross-institutional resource mapping engine, security audit and traceability module, and collaborative deduction engine.

6. A battlefield joint interaction simulation system with a light center architecture, used to implement a battlefield joint interaction simulation method with a light center architecture as claimed in any one of claims 1 to 5, characterized in that: The system comprises: Dynamic cluster management subsystem, which integrates the performance evaluation model and resource scheduling engine based on federated learning; Intelligent routing subsystem, including multi-scale feature extraction module and path decision matrix generator; Digital test platform, equipped with virtual-real fusion interface and cross-domain security tunnel; The performance monitoring center is deployed with a time series database and a three-dimensional situation visualization engine.

7. A battlefield joint interactive simulation system with a light center architecture according to claim 6, characterized in that: The dynamic cluster management subsystem further includes: The node profiling module is used to collect the characteristics of cross-institutional simulation nodes, including computing power, storage margin, and security level; The fault prediction unit uses a time series prediction model based on the Transformer architecture, with a prediction window of ≥30 minutes; The elastic scaling component is used to support dynamic adjustment of the service cluster scale according to the battlefield situation and adjust the response time to <2s.

8. The battlefield joint interactive simulation system with a light center architecture according to claim 6 is characterized in that: The intelligent routing subsystem adopts a multipath transmission controller that supports 6 parallel transmission paths; The intelligent routing subsystem also adopts an adaptive coding strategy to dynamically select a coding scheme based on channel quality, including: In high signal-to-noise ratio scenarios, select Turbo coding with a bit rate of 0.8; In the medium signal-to-noise ratio scenario, select LDPC code with a code rate of 0.6; In low signal-to-noise ratio scenarios, select polar coding with a code rate of 0.

4.

9. The battlefield joint interactive simulation system with a light center architecture according to claim 6, characterized in that: The digital test platform further includes: Model fusion engine to support smooth transition from LVC to digital twin; Protocol sandbox environment, providing a set of protocol simulation test tools; Security audit module, used to record complete operation logs and support blockchain evidence storage.

10. The battlefield joint interactive simulation system with a light center architecture according to claim 6, characterized in that: The performance monitoring center further includes: Multi-dimensional evaluation indicator system, including transmission efficiency, resource utilization, and security compliance; Real-time alarm module, with a three-level alarm mechanism, including general alarm, severe alarm, and emergency response; Retrospective analysis tool to locate performance bottlenecks based on causal reasoning models.

Images