Search

CN-121809282-B - Electromagnetic environment simulation method based on armored vehicle trunking communication

CN121809282BCN 121809282 BCN121809282 BCN 121809282BCN-121809282-B

Abstract

The invention discloses an electromagnetic environment simulation method based on armored vehicle trunking communication, which comprises the steps of constructing a structured behavior library containing various attack behaviors and defense strategies, defining a cross-layer state vector penetrating through three layers of physics, network and task, performing game decision by a multi-agent reinforcement learning driven attack and defense agent in the simulation process according to a payment function associated with success and failure of the task, selecting actions in the behavior library, executing the attack and defense event through a discrete event simulation engine, precisely evolving electromagnetic environment and cross-layer conduction containing the influence of the antenna attitude, analyzing mass time sequence data by using a causal inference algorithm after the simulation is finished, and constructing a causal relationship graph to realize automatic root diagnosis of task failure. The method aims to solve the problems that the traditional simulation cannot simulate intelligent game, hierarchical fracture exists and the depth diagnosis capability is lacking, and meanwhile, the intelligent game simulation with high fidelity and causality mechanism insight can be realized.

Inventors

  • LI GANG
  • LIU BAOLI
  • ZHU QIANG

Assignees

  • 徐州九鼎机电总厂

Dates

Publication Date
20260512
Application Date
20260304

Claims (10)

  1. 1. An electromagnetic environment simulation method based on armored vehicle trunking communication is characterized by comprising the following steps: Constructing a structured behavior library comprising a plurality of attack behaviors and defensive strategies configured to modify a cross-layer state vector of armored vehicle nodes, the cross-layer state vector comprising state parameters characterizing an armored vehicle physical layer, a network layer and a task layer; In the simulation process, the intelligent agents representing the attack party and the defending party select the attack behavior and the defending strategy of the current round from the structured behavior library according to the respectively observed state information and a preset payment function related to task success and failure; Converting the attack behavior and the defending strategy into simulation events, executing the simulation events by a simulation engine, updating the cross-layer state vector of the armored vehicle node, and evolving an electromagnetic environment in the simulation process; After the simulation is finished, based on the cross-layer state vector time sequence data recorded in the simulation process and the selected attack behavior and defense strategy, a causal inference algorithm is applied to construct a directed acyclic graph representing causal relation among the state parameters, and a root simulation event causing failure of a preset task is identified according to the directed acyclic graph.
  2. 2. The electromagnetic environment simulation method based on armored vehicle trunked communication of claim 1, wherein in constructing the structured behavior library, each attack behavior or defending strategy is defined as a data structure comprising: An identifier specifying the attack behavior or defense strategy validation protocol layer; a set of preconditions defining execution of the attack or defense strategy; a set of preset rules for modifying specific state parameters in the cross-layer state vector.
  3. 3. The method for simulating an electromagnetic environment based on armored vehicle trunking communication of claim 1, wherein the payment function value is determined according to a weighted combination of a task success rate indicator, a quantized value of simulated resource consumption and a quantized value of probability of communication behavior being detected by an enemy, wherein the task success rate indicator has a positive weight and the other two have negative weights.
  4. 4. The electromagnetic environment simulation method based on armored vehicle trunking communication according to claim 3, wherein the quantized value of the probability of detecting the communication behavior by the enemy is obtained by calculating information divergence between the communication behavior characteristic distribution at the current moment and the communication behavior characteristic distribution at the historical moment.
  5. 5. The method for simulating an electromagnetic environment based on armored vehicle trunked communication of claim 1, wherein the process of selecting the attack behavior and the defending strategy by the agents is realized through a multi-agent reinforcement learning algorithm, and each agent outputs the deterministic action of the current round by utilizing a pre-trained strategy network based on the partial observable state information of each agent.
  6. 6. The electromagnetic environment simulation method based on armored vehicle trunking communication of claim 1, wherein the electromagnetic environment in the evolution simulation process comprises: when the simulation event is physical layer interference, an interference power item is added outside the original noise item in the signal-to-interference-and-noise ratio calculation process of the nodes in the affected area according to the power spectrum density model of the interference.
  7. 7. The electromagnetic environment simulation method based on armored vehicle trunked communication of claim 6, wherein the calculating process of the signal-to-interference-and-noise ratio further comprises: according to the real-time gesture of the armored vehicle node, inquiring and applying corresponding antenna gain values from a database which is pre-stored with antenna three-dimensional radiation patterns of the vehicle under different gestures.
  8. 8. The electromagnetic environment simulation method based on armored vehicle trunked communication of claim 1, wherein the simulation engine performs the process of simulating events, comprising: When the state parameter of one protocol layer changes due to the execution of the simulation event and the change meets the triggering condition of one simulation event of another protocol layer, a new simulation event of the other protocol layer is automatically generated and scheduled.
  9. 9. The method of electromagnetic environment simulation based on armored vehicle trunked communication of claim 1 wherein the causal inference algorithm employs a constraint-based causal discovery algorithm which determines the correlation between state parameters by conditional independence checking of the time series data and applies a directional rule to determine causal directions for the correlation to form the directed acyclic graph.
  10. 10. The electromagnetic environment simulation method based on armored vehicle trunked communication of claim 9, wherein the process of identifying the root cause simulation event that causes the failure of the preset task comprises: And in the directed acyclic graph, starting from the node representing task failure, performing backward tracing along a path with causality until tracing to one or more root nodes without upstream input nodes, and generating a diagnosis report containing the root nodes, the tracing path and the task failure result.

Description

Electromagnetic environment simulation method based on armored vehicle trunking communication Technical Field The invention relates to the technical field of wireless communication network simulation, in particular to an electromagnetic environment simulation method based on armored vehicle trunking communication. Background With the advanced development of modern military informatization, ad-hoc networks (Ad-hoc networks) with armored vehicles as nodes have become an infrastructure for realizing battlefield situation sharing, command coordination and joint operations. In order to ensure the reliability and viability of such networks in a practical environment, it is critical to conduct high-fidelity simulation tests during the development and train-loading stages. Current network simulation techniques, such as NS-3, OPNET, etc. based approaches, have been able to model and evaluate physical layer channel characteristics, MAC layer protocol efficiency, and network layer routing performance of mobile ad hoc networks. Part of advanced simulation even merges a three-dimensional Geographic Information System (GIS) to simulate the influence of terrain shielding on signal propagation, and realizes the simulation of a simple and static electronic interference scene to a certain extent, thereby providing a powerful tool for basic performance verification of a communication system. However, the limitations of the existing simulation technology are increasingly prominent when dealing with future intelligent and collaborative battlefield environments. Firstly, most of traditional simulation methods adopt scripted attack models, intelligent countermeasure actions of an attacker, which are dynamically adjusted according to countermeasures of a defender, cannot be simulated, so that simulation results are seriously disjointed with a complex game process of a real attack and defense scene. Second, the evaluation dimension of the simulation is relatively single, and is generally limited to network quality of service (QoS) indexes such as throughput and time delay, and it is difficult to directly relate the microscopic change of network performance to the macroscopic combat effectiveness of the cluster "task completion", that is, the core problem of "whether the network supports the task successfully" cannot be effectively answered. Third, the existing simulation method has hierarchical fracture on the model, and attack means such as interference of a physical layer, route spoofing of a network layer, data pollution of an application layer and the like are usually analyzed in an isolated manner, so that a 'combined killing chain' effect generated under the synergistic effect of attacks of different levels cannot be revealed, and the comprehensive effectiveness of a cross-layer defense strategy cannot be evaluated. Disclosure of Invention This section is intended to outline some aspects of embodiments of the application and to briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section as well as in the description of the application and in the title of the application, which may not be used to limit the scope of the application. The present invention has been made in view of the above-described problems occurring in the prior art. Therefore, the invention provides an electromagnetic environment simulation method based on armored vehicle trunking communication, which is used for solving the problems in the background technology. In order to solve the technical problems, the invention provides the following technical scheme that the electromagnetic environment simulation method based on armored vehicle trunking communication comprises the following steps: Constructing a structured behavior library comprising a plurality of attack behaviors and defensive strategies configured to modify a cross-layer state vector of armored vehicle nodes, the cross-layer state vector comprising state parameters characterizing an armored vehicle physical layer, a network layer and a task layer; In the simulation process, the intelligent agents representing the attack party and the defending party select the attack behavior and the defending strategy of the current round from the structured behavior library according to the respectively observed state information and a preset payment function related to task success and failure; Converting the attack behavior and the defending strategy into simulation events, executing the simulation events by a simulation engine, updating the cross-layer state vector of the armored vehicle node, and evolving an electromagnetic environment in the simulation process; After the simulation is finished, based on the cross-layer state vector time sequence data recorded in the simulation process and the selected attack behavior and defense strategy, a causal inference algorithm is applied to construct a directed acyclic graph representing causal relation among the state p