Large-scale virtuality and reality combined network simulation architecture based on lightweight virtualization technology

Abstract

The invention relates to the technical field of communication network simulation, and particularly discloses a large-scale virtuality and reality combined network simulation architecture based on a lightweight virtualization technology, which comprises a clock module, a communication node equivalent device, a physical node, a simulation host and a radio frequency switching matrix, the clock module is used for providing time synchronization signals for the physical nodes, the communication node equivalents and the simulation host through one-time timing or periodic timing; the simulation host is used for calculating link quality, controlling link on-off of the radio frequency switching matrix according to the link quality, and running the digital node simulator; the radio frequency switching matrix is used for constructing radio frequency channels between the physical nodes and between the physical nodes and the communication node equivalents; the physical node comprises to-be-tested equipment of a real communication terminal machine and is used for transmitting a radio frequency signal through a radio frequency channel; and the communication node equivalents are used for analyzing the radio frequency signals into data frames of a link layer and converting the data frames sent by the simulation host into the radio frequency signals.

Description

Technical Field

[0001] This invention relates to the field of communication network simulation technology, and specifically to a large-scale virtual-real network simulation architecture based on lightweight virtualization technology. Background Technology

[0002] Network simulation technology can provide important support for the performance evaluation of complex flight networks. Currently, common network simulation tools include NS-3, OMNeT++, OPNET, and MATLAB.

[0003] NS-3 is one of the most widely used network simulation tools. Its modular design and open-source nature allow researchers to extend it to meet the specific needs of FANETs. NS-3 supports detailed modeling of wireless communication protocols, simulating protocols at the physical, MAC, and network layers, making it suitable for studying multi-hop routing, link maintenance, and dynamic topology changes in FANETs. OMNeT++ provides powerful network communication modeling capabilities, suitable for simulating large-scale network scenarios. This tool performs particularly well in multi-UAV network scenarios, supporting multi-protocol stack simulation and effectively evaluating the performance of various routing protocols in FANETs. OPNET and MATLAB excel in in-depth protocol stack analysis and algorithm-level performance evaluation, respectively. MATLAB's simulation flexibility allows researchers to easily build customized FANET models and evaluate communication link reliability and node energy consumption.

[0004] However, existing simulations do not incorporate real hardware; existing simulation tools still have limitations in accurately modeling complex motion models and channel environments of flight ad hoc networks in three-dimensional space; and existing simulation tools face bottlenecks in computing and storage resources when handling simulations of flight ad hoc networks with thousands of nodes.

[0005] Therefore, combining intelligent algorithms and large-scale hardware-in-the-loop simulation, virtual-real network simulation verification will become a trend. Further research is needed to solve the problems of virtual-real synchronization and protocol stack consistency. Summary of the Invention

[0006] To address the aforementioned issues, the present invention aims to provide a large-scale virtual-real integrated network simulation architecture based on lightweight virtualization technology. This architecture solves challenges in cluster network simulation such as virtual-real time synchronization, low-latency interaction of virtual-real data, and consistency of virtual-real protocol stacks. It enables simulation scale expansion, simulation accuracy improvement, and hardware-in-the-loop simulation capabilities, providing an environment for the testing and verification of protocol algorithms, promoting algorithm iteration, and accelerating technology maturity.

[0007] This invention provides a large-scale virtual-physical network simulation architecture based on lightweight virtualization technology, including: a clock module, a communication node equivalent, physical nodes, a simulation host, and a radio frequency switching matrix; The communication node equivalent, physical node, and simulation host are respectively connected to the clock module; the communication node equivalent and physical node are respectively connected to the radio frequency switching matrix. The clock module is used to provide time synchronization signals to the physical nodes, communication node equivalents, and simulation host through one-time or periodic time synchronization; The simulation host is used to calculate link quality and control the link connection and disconnection of the radio frequency switching matrix according to the link quality, as well as to run the digital node simulator. The radio frequency switching matrix is ​​used to construct radio frequency channels between the physical nodes and between the physical nodes and the communication node equivalents; The physical node includes a real communication terminal device under test, used to transmit radio frequency signals through the radio frequency channel; The communication node equivalent is used to parse the radio frequency signal into a data frame at the link layer, and to convert the data frame sent by the digital node simulator in the simulation host into a radio frequency signal, so as to realize the data interaction between the physical node and the digital node on the simulation host.

[0008] In one possible implementation, the simulation host includes: a time synchronization module, a digital node simulator, a virtual-real mapping module, a channel simulation module, and a simulation master controller; The digital node emulator is built on lightweight virtualization technology and is used to run protocol stack programs by isolating resources through Linux namespaces. The virtual-to-real mapping module includes an interface proxy and a mapping node, used for data transmission between the digital nodes in the digital node simulator and the physical nodes. The simulation master controller is used to generate pose information, clock information and service data, and monitor network status; The channel simulation module is used to dynamically calculate link quality and control link status based on link configuration and node location information.

[0009] In one possible implementation, the mapping node of the virtual-physical mapping module runs within a container, implements only the physical layer protocol stack, and completes data interaction with the communication node equivalent through the interface proxy. At the same time, it uses virtual Ethernet to complete data interaction with the digital node via the channel simulation module.

[0010] In one possible implementation, the channel simulation module is also used to dynamically control the virtual links between the digital nodes.

[0011] In one possible implementation, the simulation master controller supports dynamically injecting node motion parameters and communication configurations to generate pose information and business data streams for both virtual and real nodes.

[0012] In one possible implementation, the digital node emulator ensures consistency with the physical node protocol stack by refactoring the Linux system interface functions of the embedded protocol stack.

[0013] In one possible implementation, the communication node equivalent includes an information processing unit for bidirectional conversion between radio frequency signals and data frames at the link layer, and interconnection with the emulation host via an Ethernet interface.

[0014] In one possible implementation, the emulation host creates a containerized environment using lightweight virtualization technology, and uses Linux PID, Mount, Network, and User Namespace to achieve resource isolation, providing an independent network protocol stack environment for each digital node.

[0015] In one possible implementation, the interface proxy encapsulates and decapsulates IP data packets via a UDP port to enable cross-network domain communication between the digital node and the physical node.

[0016] In one possible implementation, the large-scale virtual-physical hybrid network simulation architecture enables cross-operating system network connectivity between digital nodes and equivalent nodes through Linux-bridge and veth virtual network cards, and completes data forwarding based on a unified IP routing strategy.

[0017] This invention provides a large-scale virtual-physical hybrid network simulation architecture based on lightweight virtualization technology, aiming to achieve efficient simulation and interaction between digital nodes and physical nodes. The hardware component includes a clock module, a communication node equivalent, a simulation host, physical nodes, and an RF switching matrix, providing functions such as time synchronization, signal conversion, simulation operation, and RF channel construction. The software component encompasses a digital node simulator, a virtual-physical mapping module, simulation control software, and a channel simulation module. Lightweight virtualization technology achieves resource isolation and protocol stack operation, supporting virtual-physical data pass-through, network status monitoring, and dynamic link quality adjustment. This invention constructs an efficient, flexible, and accurate network simulation environment suitable for simulation and testing of various complex network scenarios. Attached Figure Description

[0018] Figure 1 A schematic diagram of a large-scale virtual-real hybrid network simulation architecture provided for embodiments of the present invention; Figure 2 A schematic diagram of a digital node simulator provided in an embodiment of the present invention; Figure 3 A schematic diagram of the equivalent node forwarding mode provided in an embodiment of the present invention; Figure 4 This is a connection diagram from the equivalent node to the digital node provided in an embodiment of the present invention; Figure 5 A schematic diagram illustrating the communication between a digital node and an equivalent node provided in an embodiment of the present invention; Figure 6 This is a schematic diagram of the communication between the equivalent node and the digital node provided in an embodiment of the present invention. Detailed Implementation

[0019] The embodiments of the present invention will be further described in detail below with reference to the accompanying drawings and examples. The following detailed description of the embodiments and the accompanying drawings are used to illustrate the principles of the present invention by way of example, but should not be used to limit the scope of the present invention. That is, the present invention is not limited to the described preferred embodiments, and the scope of the present invention is defined by the claims.

[0020] In the description of this invention, it should be noted that, unless otherwise stated, "a plurality of" means two or more; the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance; those skilled in the art can understand the specific meaning of the above terms in this invention as appropriate.

[0021] This invention employs lightweight virtualization technology to construct a virtual-physical integrated network simulation platform, overcoming the shortcomings and limitations of traditional network simulation platforms and physical testing methods. It expands advantages in areas such as network simulation scale expansion, fine-grained network traffic management, and virtual-physical consistency. It can be used for designing and verifying protocol stacks for large-scale aircraft cluster ad hoc networks, supporting distributed network simulation combining virtual and physical components. This invention constructs an efficient, flexible, and accurate network simulation environment suitable for simulation and testing of various complex network scenarios.

[0022] Figure 1 A schematic diagram of a large-scale virtual-real hybrid network simulation architecture provided for embodiments of the present invention is shown below. Figure 1 As shown, the present invention provides a large-scale virtual-physical hybrid network simulation architecture based on lightweight virtualization technology, including: a clock module, a communication node equivalent, physical nodes, a simulation host, and a radio frequency switching matrix; The communication node equivalent, physical node, and simulation host are connected to the clock module; the communication node equivalent and physical node are connected to the radio frequency switching matrix. The clock module is used to provide time synchronization signals for physical nodes, communication node equivalents, and the simulation host through one-time or periodic time synchronization, depending on the time accuracy requirements of the verification scenario. The simulation host is used to calculate link quality and control the link connection and disconnection of the RF switching matrix based on the link quality, as well as to run the digital node simulator; Radio frequency switching matrices are used to construct radio frequency channels between physical nodes and between physical nodes and communication node equivalents; Physical nodes include actual communication terminals used to transmit radio frequency signals through radio frequency channels; In one example, a physical node consists of two or more physical products with communication terminals.

[0023] The communication node equivalent is an integral part of the virtual-physical simulation platform. It contains an information processing unit and can be interconnected with the physical communication terminal via radio frequency. It is used to parse radio frequency signals into data frames at the link layer and convert data frames sent by the digital node simulator in the simulation host into radio frequency signals to realize data interaction between the physical node and the digital node on the simulation host, as well as interconnect with the simulation host via an Ethernet interface.

[0024] In one possible implementation, a large-scale virtual-physical hybrid network simulation architecture enables cross-operating system network connectivity between digital nodes and equivalent nodes through Linux-bridge and veth virtual network interface cards, and completes data forwarding based on a unified IP routing policy.

[0025] In one possible implementation, the simulation host is the key hardware of this invention. In addition to the core computing unit, it is additionally configured with a time synchronization module, capable of receiving and processing external time synchronization signals, and used to run digital node simulators, virtual-to-real mapping modules, simulation control software, and channel simulation software. The simulation host creates a containerized environment using lightweight virtualization technology, utilizing Linux PID, Mount, Network, and User Namespace to achieve resource isolation, providing each digital node with an independent network protocol stack environment.

[0026] The simulation host includes: a time synchronization module, a digital node simulator, a virtual-real mapping module, a channel simulation module, and a simulation controller; A digital node emulator, built on lightweight virtualization technology, is used to run protocol stack programs by isolating resources through Linux namespaces; The virtual-to-physical mapping module, including interface proxies and mapping nodes, is used for data transmission between digital nodes and physical nodes within the digital node emulator. The digital node emulator ensures consistency with the physical node protocol stack by reconstructing the Linux system interface functions of the embedded protocol stack.

[0027] The simulation master controller is used to generate pose information, clock information and service data, and monitor network status; The channel simulation module is used to dynamically calculate link quality and control link status based on link configuration and node location information.

[0028] The following provides a further explanation of the components in the simulation host.

[0029] Traditional physical network simulation methods face challenges such as high hardware costs, poor scalability, and difficulty in simulating dynamic changes. While virtualization-based simulation improves flexibility and scalability, traditional virtualization technologies still have shortcomings in terms of resource consumption, startup time, and performance overhead, especially in large-scale digital-combined network simulation scenarios, where they face problems such as limited node numbers, increased latency, and resource waste. Therefore, this invention employs lightweight virtualization technology to construct a digital node simulator, specifically composed of the following components: Figure 2 As shown.

[0030] The essence of a digital node simulator environment is to use CPU resources to simulate the operation of a large number of communication nodes. The main idea is to use Linux namespaces to create a kernel-level resource isolation mechanism, turning a physical host into multiple independent environment namespaces. This provides a real-time distributed digital node network simulation environment that can be connected to physical networks and routers through a unified Ethernet interface.

[0031] Linux Namespaces, provided by the Linux kernel, are a technology for inter-process resource isolation. They encapsulate global system resources in an abstract, isolated environment, allowing processes to have independent instances of these global resources. Changing system resources in one Namespace only affects processes within that Namespace, having no impact on processes in other Namespaces. Linux provides several Namespaces by default for isolating various resources, and these Namespaces can be manipulated via API calls. This invention primarily employs PID, Mount, Network, and UserNamespace isolation types to create isolated, realistic network environments and isolated process environments for simulation nodes. Through integration with container management tools, the following resource isolation can be achieved: PID and Mount Namespace: Ensure that processes and file systems within each container are independent, avoiding resource conflicts.

[0032] Network Namespace: Each container has its own independent network stack, which can be configured with its own IP address, routing table, etc.

[0033] User Namespace: Enhances security by allowing different user identities to be used in different containers, preventing permission abuse.

[0034] A container that integrates the above isolation types can act as a digital node, running network protocols independently and interacting with external applications and other digital nodes through a virtual Ethernet port.

[0035] After completing the construction of the digital node isolation environment, a specified protocol stack program can be run on the container, mainly including the application layer, network layer, link layer, and physical layer. Meanwhile, to ensure consistency between the virtual and physical environments, the communication protocol stack software on the embedded board needs to be adapted for the Linux system. For interface function definitions not available in Linux, the functions should be refactored to achieve the same functionality as the function interfaces in the original operating system while preserving the original protocol stack program architecture and implementation.

[0036] In one possible implementation, the mapping node of the virtual-physical mapping module runs inside a container, implements only the physical layer protocol stack, and completes data interaction with the communication node equivalent through an interface proxy. At the same time, it uses virtual Ethernet to complete data interaction with the digital node through the channel simulation module.

[0037] To connect to real hardware, the virtual-to-physical mapping module requires real hardware to request mappings from the simulation network during simulation, establishing connections and correspondences between real and simulated resources to connect multiple real nodes to the simulation network. It also tests and verifies the interactions between digital nodes and physical nodes, and between physical nodes and digital nodes. The physical mapping module provides interaction between digital and simulated nodes. Furthermore, to improve system adaptability and support testing of different types of physical terminals, the virtual-to-physical mapping module is not directly connected to the physical terminal, but rather through a communication node equivalent. Physical communication terminals can be directly connected to the system for testing without any modifications, and the communication node equivalent, acting as a relay unit, can adaptively adjust the information processing unit according to the physical communication terminal.

[0038] The virtual-real mapping module includes the implementation of interface proxy and multiple mapping nodes.

[0039] Mapping nodes act as "agents" for real nodes in the simulated network. To represent the behavior of real nodes in the simulated network, whether sending and receiving message packets, exchanging data packets, or changing node configuration parameters, mapping nodes maintain consistency with real nodes. Mapping nodes act as data forwarding proxies to achieve transparent message transmission. Mapping nodes also run on resource-isolated containers. Unlike digital nodes, they only need to implement the protocol stack below the link layer. They complete real-time data interaction with equivalent nodes through interface proxies. After processing, the data to be exchanged by the equivalent node is forwarded to the destination digital node through a virtual Ethernet port; similarly, data from digital nodes can be forwarded to the destination equivalent node, thus realizing a virtual-real integrated network simulation system. After receiving data, the equivalent node forwards it to the physical node via radio frequency signals, thereby achieving a closed loop of virtual-real integration.

[0040] In one possible implementation, the interface proxy encapsulates and decapsulates IP packets via UDP ports to enable cross-network domain communication between digital nodes and physical nodes.

[0041] The interface proxy supports data interaction between the virtual-physical mapping module and the communication equivalent node, primarily involving link-layer data packets to be sent and received by the equivalent node. It provides the following two interfaces: The interface module for receiving data from digital nodes to physical nodes (forwarded via equivalent nodes) listens on a UDP port to receive data sent from digital nodes. The specific processing flow is as follows: Upon receiving a data packet, it queries the IP address of the physical node corresponding to the mapped node, encapsulates the data packet into a UDP packet, and sends it to the physical node.

[0042] The interface module for receiving data sent from physical nodes to digital nodes (forwarded via equivalent nodes) listens on a UDP port (the UDP port needs to be mapped to the host machine) to receive data sent from physical nodes. The specific processing flow is as follows: upon receiving a data packet (IP over UDP encapsulation), the UDP header is removed, the IP packet is sent to the digital node emulator via the interface proxy, and then passed to the corresponding digital node.

[0043] In one possible implementation, the simulation master controller supports dynamically injecting node motion parameters and communication configurations to generate pose information and business data streams for both virtual and real nodes.

[0044] The simulation master controller is used to generate or forward the aircraft's pose information, clock information, and service data, and periodically monitors the network status, link quality, and interactive service data reported by digital nodes and virtual-physical mapping nodes.

[0045] The channel simulation module can calculate the link transmission status based on the input link configuration information and link position and attitude information, and implement control over the links between digital nodes and between digital nodes and mapping nodes based on the calculation results. Simultaneously, the corresponding calculation results are transmitted to the radio frequency interaction matrix to realize link simulation between physical nodes and equivalent nodes.

[0046] The communication method of the present invention will be described below with reference to embodiments.

[0047] Example 1: Communication between digital nodes Communication between digital nodes is equivalent to communication between multiple containers. The self-organizing network protocol of digital nodes running on containers, according to the network layer configuration, calls virtual network interfaces to send messages to other digital nodes, and can also receive service data from other digital nodes through virtual network interfaces.

[0048] Example 2: Communication between physical nodes Communication between physical nodes is achieved through a radio frequency (RF) switching matrix. The ad hoc network protocol running on the physical nodes transmits messages destined for other physical nodes via the RF switching matrix according to the network layer configuration. The RF switching matrix controls link connectivity based on channel simulation software calculations.

[0049] Example 3: Communication between digital nodes and physical nodes Communication between physical and digital nodes is achieved through a two-level forwarding process using a communication equivalent node and a virtual-physical mapping module, which can accommodate access from different communication terminals.

[0050] 1) Equivalent node forwarding mode.

[0051] When a physical node sends data to a digital node, the physical node, according to the running ad hoc network protocol stack and corresponding rules, transmits the message to the equivalent node via radio frequency signals. The equivalent node then processes the information, extracts the corresponding link-layer data, and forwards it to the virtual-physical mapping node. The virtual-physical mapping node then transmits the data to the corresponding digital node through channel simulation software. The process of a digital node sending data to a physical node is the reverse. Specifically, as follows... Figure 3 As shown.

[0052] 2) Communication between the communication equivalent node and the digital node The equivalent node runs an embedded real-time operating system, while the digital node runs a Linux operating system, involving network connectivity between different systems. For the Linux operating system, the system abstracts a device file that corresponds to a real physical device, and all operations are performed through this abstract device file. Whether it's a physical or virtual device, the operating system maps it to a corresponding device instance. Linux-bridge is a virtual network device used for TCP / IP Layer 2 protocol switching in a common Linux operating system network kernel module. It possesses the characteristics of a physical network device, such as setting IP addresses and MAC addresses. Furthermore, Linux-bridge is a virtual switch, operating on the same principle as a physical switch; connecting to a terminal is a device connection process. It also possesses the isolation functions of a switch, allowing for any number of ports. Data can flow from any port, and the MAC address table determines the outgoing port. The virtual-physical mapping node and the digital node utilize Linux-bridge functionality to achieve data exchange between multiple virtual network ports; the virtual-physical mapping node then uses an interface proxy to exchange data with the equivalent node. Specifically... Figure 4 As shown.

[0053] Both digital nodes and equivalent nodes have their own unique IP addresses, and they exchange data via these IP addresses. Communication between digital nodes and equivalent nodes involves three network segments: (1) The physical network between the equivalent node and the simulation host running the digital node is connected by a switch, and the corresponding network segment is 192.XX.XX.XX; (2) The communication network segment connecting all digital nodes in the simulation network running on the simulation host is 172.XX.XX.XX; (3) Connect all virtual and physical mapping node networks. The corresponding network segment is 10.XX.XX.XX. This network segment is used to export the simulated traffic to the external host network through the proxy.

[0054] Traffic sent from digital nodes to equivalent nodes needs to be routed through veth pairs or bridge-type virtual network interface cards to the host network protocol stack since the digital nodes run on the simulation host. 1) Since the digital node and the semi-physical mapping node are connected on the same IP network segment, the traffic of the digital node is directly sent to the data conversion process running on the semi-physical mapping node; 2) Virtual-to-real mapping nodes need to be configured with the IP address of their corresponding equivalent node. When the data conversion process running on the virtual-to-real node receives a data packet, it queries the IP address of its corresponding equivalent node, encapsulates the data packet, and sends it to the br bridge through the veth network card. 3) The br bridge resides in the host's protocol stack. After querying the host's routing table and finding that the destination IP is in the same network segment, it sends the data packet to the corresponding equivalent node via the host's eth0 network interface. The specific data flow is as follows: Figure 5 As shown.

[0055] The traffic sent from the equivalent node to the digital node is limited by the fact that the equivalent node is a semi-embedded real-time operating system with limited support for the Ethernet protocol stack and virtual networks. Therefore, methods such as VXLAN and IP forwarding as used in Linux cannot be used. An interface agent needs to be run on the Linux host to forward packets. The equivalent node sends packets to the corresponding port on the UDP port of the listening window running on the Linux host. The specific steps are as follows.

[0056] 1) The equivalent node uses the first 8 bytes of the UDP packet to store the destination IP address and the source IP address. The destination IP address is the IP address corresponding to the destination digital node, and the source IP address is the IP address of the semi-physical mapping node corresponding to the equivalent node. The packet is then sent to the interface proxy of the Linux host. 2) The proxy program obtains UDP data, parses the first 8 bytes of the UDP packet to obtain the destination IP address and source IP address, and encapsulates the packet into UDP according to the source IP address and sends it to the corresponding physical mapping node. 3) The data conversion process running on the virtual-to-real mapping node parses the first 4 bytes of the UDP packet to obtain the destination IP address (the IP address of the digital node). The data forwarding process then sends the packet to the digital node via channel simulation. The specific data flow is as follows: Figure 6 As shown.

[0057] The large-scale virtual-real hybrid network simulation architecture based on lightweight virtualization technology provided by this invention has the following beneficial effects: Architectural innovation: A multi-dimensional fusion network simulation architecture based on lightweight virtualization technology is constructed, which effectively solves three core technical challenges in large-scale cluster network simulation: cross-domain spatiotemporal synchronization, virtual-real interaction latency suppression, and consistent mapping of virtual-real protocol stacks.

[0058] Multi-scenario applicability: Real-time injection of node motion information and communication parameters according to different application scenarios, fully characterizing complex network states such as topology changes and link quality fluctuations caused by aircraft maneuvering characteristics.

[0059] High-precision time synchronization: An innovative design of a virtual and real clock joint driving mechanism achieves microsecond-level time accuracy synchronization; A full-process verification environment for network protocols: A full-element network simulation verification environment is built, which can support multi-dimensional performance evaluation of core protocols such as routing algorithms and topology control, and accelerate the research and verification process of flight ad hoc network protocol algorithms.

[0060] Scalability: This invention has formed a modular technology system that can be adapted to diverse dynamic networking scenarios such as spaceborne and vehicle-mounted networks, and promotes the verification of related intelligent networking technologies.

[0061] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A large-scale virtual-real hybrid network simulation architecture based on lightweight virtualization technology, characterized in that, include: Clock module, communication node equivalent, physical node, simulation host and RF switching matrix; The communication node equivalent, physical node, and simulation host are respectively connected to the clock module; the communication node equivalent and physical node are respectively connected to the radio frequency switching matrix. The clock module is used to provide time synchronization signals to the physical nodes, communication node equivalents, and simulation host through one-time or periodic time synchronization; The simulation host is used to calculate link quality and control the link connection and disconnection of the radio frequency switching matrix according to the link quality, as well as to run the digital node simulator. The radio frequency switching matrix is ​​used to construct radio frequency channels between the physical nodes and between the physical nodes and the communication node equivalents; The physical node includes a real communication terminal device under test, used to transmit radio frequency signals through the radio frequency channel; The communication node equivalent is used to parse the radio frequency signal into a data frame at the link layer, and to convert the data frame sent by the digital node simulator in the simulation host into a radio frequency signal, so as to realize the data interaction between the physical node and the digital node on the simulation host.

2. The large-scale virtual-real hybrid network simulation architecture according to claim 1, characterized in that, The simulation host includes: a time synchronization module, a digital node simulator, a virtual-real mapping module, a channel simulation module, and a simulation master controller; The digital node emulator is built on lightweight virtualization technology and is used to run protocol stack programs by isolating resources through Linux namespaces. The virtual-to-real mapping module includes an interface proxy and a mapping node, used for data transmission between the digital nodes in the digital node simulator and the physical nodes. The simulation master controller is used to generate pose information, clock information and service data, and monitor network status; The channel simulation module is used to dynamically calculate link quality and control link status based on link configuration and node location information.

3. The large-scale virtual-real hybrid network simulation architecture according to claim 2, characterized in that, The mapping node of the virtual-real mapping module runs inside a container, implements only the physical layer protocol stack, and completes data interaction with the communication node equivalent through the interface proxy. At the same time, it completes data interaction with the digital node through the channel simulation module using virtual Ethernet.

4. The large-scale virtual-real hybrid network simulation architecture according to claim 2, characterized in that, The channel simulation module is also used to dynamically control the virtual links between the digital nodes.

5. The large-scale virtual-real hybrid network simulation architecture according to claim 2, characterized in that, The simulation master controller supports dynamically injecting node motion parameters and communication configurations to generate pose information and business data streams for virtual and real nodes.

6. The large-scale virtual-real hybrid network simulation architecture according to claim 2, characterized in that, The digital node emulator ensures consistency with the physical node protocol stack by reconstructing the Linux system interface functions of the embedded protocol stack.

7. The large-scale virtual-real hybrid network simulation architecture according to claim 1, characterized in that, The communication node equivalent includes an information processing unit for bidirectional conversion between radio frequency signals and data frames at the link layer, and interconnection with the simulation host via an Ethernet interface.

8. The large-scale virtual-real hybrid network simulation architecture according to claim 1, characterized in that, The simulation host creates a containerized environment using lightweight virtualization technology, and uses Linux PID, Mount, Network and UserNamespace to achieve resource isolation, providing an independent network protocol stack environment for each digital node.

9. The large-scale virtual-real hybrid network simulation architecture according to claim 2, characterized in that, The interface proxy encapsulates and decapsulates IP data packets via UDP ports to enable cross-network domain communication between the digital node and the physical node.

10. The large-scale virtual-real hybrid network simulation architecture according to claim 1, characterized in that, The large-scale virtual-real hybrid network simulation architecture enables cross-operating system network connectivity between digital nodes and equivalent nodes through Linux-bridge and veth virtual network cards, and completes data forwarding based on a unified IP routing strategy.

Images