A method for high-speed processing and forwarding simulation of space networks

Abstract

The application discloses a high-speed processing and forwarding simulation method of a space network and belongs to the technical field of network simulation. The implementation scheme comprises the following steps: for a space network scene to be simulated, a space network simulation node is constructed, a vector data packet processing component is configured and initialized for each simulation node, and the simulation node scale is deployed; according to link communication requirements and scale deployment needs, the vector data packet processing component is configured in simulation nodes at both ends of a simulation link to construct the simulation link; a simulation link attribute control mechanism is constructed, and the dynamic configuration of the simulation link attribute is performed in each time slot; the simulation link is switched based on a smooth switching mechanism and the vector data packet processing component; and an automatic simulation data acquisition method is adopted to capture and analyze target data in the simulation process. The application can realize the space network simulation function, meet the high-speed processing and forwarding requirements of the communication link in a high-throughput scene, and realize a more accurate network simulation effect.

Description

Technical Field

[0001] This invention relates to the field of network simulation technology, and more specifically to a high-speed processing and forwarding simulation method for spatial networks. Background Technology

[0002] With the development of low-Earth orbit satellite constellations, integrated ground-to-air communication systems, and the space internet, space networks are gradually becoming an important form of the next-generation global information infrastructure. Space networks are characterized by strong topological dynamics, long link propagation times, and widespread node distribution, which places stringent demands on their communication protocols, resource scheduling, and network performance. In the theoretical design and deployment verification phases, building an efficient simulation platform is a crucial means to advance space network research and engineering practice. Especially in typical scenarios involving high-volume forwarding and high-frequency path switching, the simulation platform not only needs to support flexible topology modeling but also must be capable of handling high-speed communication flows to realistically reflect the system's operational characteristics in complex space environments.

[0003] However, while current mainstream space network simulation tools (such as CORE and Mininet) and commonly used virtualization-based single-machine network simulation methods are comprehensive in terms of protocol logic modeling and control flow verification, most are based on traditional kernel protocol stacks. Kernel protocol stacks are designed for single-packet processing, typically requiring a system call to be triggered for each packet, transitioning from user space to kernel space and back again. This process involves frequent context switching and system scheduling, significantly increasing processing latency. Furthermore, the data packet undergoes multiple memory copies and interrupt notifications from the network card reception point to the processing logic and then to the user-space application, further wasting memory bandwidth. For example, the link dynamic control method for a large-scale satellite network simulation platform proposed in application number CN202411175141.7 uses a virtual network link connection method provided by a Veth device between virtual satellite nodes, and the virtual network data communication interaction method based on container technology and the overlay network based on VXLAN encapsulation in application number CN201911241576.6 both adopt the above-mentioned forwarding mode. The data packets undergo multiple transformations between user space and kernel space, which negatively affects processing latency and data packet throughput.

[0004] In current space network communications, laser communication links are gradually becoming the core carrier of inter-satellite and space-to-ground communications. Compared with traditional radio frequency links, laser communication shows significant advantages in high-capacity, low-latency transmission. Currently, the transmission rate of mainstream low-Earth orbit satellite laser links has increased from hundreds of Mbps to several Gbps, and some next-generation inter-satellite communication systems are even evolving towards 10 Gbps or even 100 Gbps levels. Taking Starlink, the most representative example, as an example, it currently has more than 9,000 satellites in orbit, and using laser inter-satellite link technology, the transmission rate has reached 100 Gbps. If we simulate such a large-scale, high-speed satellite communication constellation, it will inevitably place higher demands on the data packet processing of space network simulation. Because the current mainstream space network simulation tools rely on traditional kernel protocol stacks, their efficiency in handling small packet streams and high-frequency throughput is extremely low, and their processing capacity is often less than one million packets per second (PPS), which is far from meeting the requirements of laser link simulation. Even using kernel protocol stacks has significant limitations in data packet forwarding performance, link dynamic response capabilities, and large-scale node deployment. Its processing mechanism relies on frequent system calls and memory copies, which can easily create performance bottlenecks, especially in scenarios such as simulating high-bandwidth inter-satellite links or ground data transfer.

[0005] Vector Packet Processing (VPP) enables vectorized batch processing of data packets, significantly improving packet processing capabilities and reducing latency. However, several key issues remain unresolved when applying it to space network simulation. First, the scale of simulation nodes in space network simulations is typically large, requiring a sufficient number of simulation nodes to be deployed on a single server. Currently, most VPP implementations are deployed directly to individual servers, making them unsuitable for direct application in space network simulations and requiring adaptation. Second, a key characteristic of space networks is that the connections between satellite nodes, link characteristics, and network topology change over time. Therefore, space network simulations involve frequent link switching, network topology changes, and link attribute modifications—issues not considered by VPP, which is designed for fixed links.

[0006] Therefore, in order to cope with the high-throughput scenarios in large-scale space network simulation, the space network simulation method based solely on the traditional kernel protocol stack is insufficient. While considering resource consumption and ensuring large-scale deployment, it is necessary to adapt and improve the high-speed packet processing characteristics of vector packet processing, so as to provide a solution for space network simulation that is superior in terms of performance, flexibility and scalability, and establish a highly reliable, high-load, multi-path space network simulation method. Summary of the Invention

[0007] To address the problem that current spatial network simulation methods based on traditional kernel protocol stacks cannot meet the high bandwidth requirements of communication links in high-throughput scenarios, this invention provides a high-speed processing and forwarding simulation method for spatial networks. This method leverages the inherent advantage of vector packet processing in high-speed packet handling to solve the aforementioned problems while considering the characteristics of spatial network simulation, such as link establishment and switching, link control, and custom packet generation and capture, thereby achieving more accurate spatial network simulation results.

[0008] The above objectives are achieved through the following technical solutions:

[0009] S1. Simulation Environment Initialization: For the spatial network scenario to be simulated, multiple spatial network simulation nodes are constructed, and vector data packet processing components are configured and initialized for each simulation node to achieve large-scale deployment of simulation nodes.

[0010] S2. Hybrid Simulation Link Construction: Construct a spatial network simulation topology between simulation nodes; establish a hybrid link architecture based on the service attributes of the spatial links: according to the expected communication load and performance requirements of the simulation links, divide all simulation links into two types: regular simulation links and high-speed simulation links, and implement this by configuring vector data packet processing components with corresponding working modes in the simulation nodes at both ends of the simulation links; manage the logical connection and disconnection of each simulation link through a virtual switch.

[0011] S3. Dynamic control of link attributes driven by discrete time slots: The spatial network simulation process is divided into discrete time slots; within each discrete time slot, the link attributes of each simulation link are updated according to preset scenario parameters; by loading the link attribute configuration plugin on the vector data packet processing component in the simulation node, the dynamic configuration and update of the link attributes of each discrete time slot are executed.

[0012] S4. Simulation Link Switching: Perform simulation link switching based on a smooth switching mechanism. According to the preset link switching time slot plan and simulation link type, the simulation link switching is achieved in conjunction with the vector data packet processing components in the simulation nodes at both ends of the simulation link. The smooth switching mechanism is as follows: at the planned switching time, a new simulation link is first established, then the data stream is migrated from the old simulation link to the new simulation link, and finally the old simulation link is dismantled.

[0013] S5. Automatic Data Acquisition: An automatic data acquisition method is adopted, in which a data capture plugin is loaded into the vector data packet processing component of the specified simulation node to capture and analyze the target data; the automatic data acquisition method refers to automatically exporting the captured data for analysis by following preset data filtering rules and acquisition termination conditions.

[0014] The S1 part is implemented through the following configuration:

[0015] S1.1 Based on the configuration information of the space network scenario to be simulated, construct a simulation scenario including a space-based transmission network, small satellite links, and ground stations, and confirm the correspondence between the nodes in the space network scenario and the network simulation nodes.

[0016] S1.2 For the spatial network scenario to be simulated, record the connection topology and time slots between the simulation nodes.

[0017] S1.3 For the connection topology to be simulated, record the type of each simulation link as a regular simulation link or a high-speed simulation link, including the corresponding link attributes.

[0018] S1.4 Based on the types of simulation nodes in the space network scenario, different types of simulation nodes are created, corresponding simulation containers are established for the numerous satellite nodes, ground station nodes, and ground access user nodes in the space network scenario, thereby achieving large-scale deployment of simulation nodes. The simulation containers are deployed using containerization technology (Docker), and vector packet processing tools are installed and enabled within each simulation node.

[0019] The S2 part is implemented through the following configuration:

[0020] Hybrid Simulation Link Construction: A spatial network simulation topology is built between simulation nodes. Based on the service attributes of the spatial links, a hybrid link architecture is established: two types of simulation links, namely regular simulation links and high-speed simulation links, are configured by configuring vector packet processing components in the simulation nodes at both ends of the simulation links. This satisfies the preset communication requirements while reducing resource overhead and improving large-scale deployment capabilities. The regular simulation link refers to a simulation link in the spatial network simulation that does not impose quantitative performance constraints on changes in transmission bandwidth and latency jitter. The simulation scenario uses this type of link by default to form the simulation network topology. For regular simulation links, the vector packet processing components in the simulation nodes at both ends are configured in host-interface mode (i.e., host-side interface mode) to create the link, and then connected to a virtual switch for topology management. The high-speed simulation link refers to a simulation link whose preset bandwidth requirement is higher than the preset bandwidth differentiation standard, and / or whose preset latency jitter upper limit is not higher than the preset latency jitter differentiation standard. For high-speed simulation links, the vector packet processing components in the simulation nodes at both ends are configured in memif mode (i.e., shared memory mode) based on shared memory to create the link.

[0021] For the aforementioned conventional simulation link, this method uses a host-interface mode combined with a vector packet processing component to create the simulation link, and manages the link topology through a multi-layer virtual switch. First, based on the connection relationships between nodes in the configuration information of the spatial network scenario to be simulated, simulation nodes at both ends of a simulation link are selected. Next, the host uses Ethernet pairs to connect these two simulation nodes to the multi-layer virtual switch. Then, in the simulation nodes at both ends of the simulation link, the host-interface is used to transfer the network interface card of the simulation node to the vector packet processing process. Finally, the host-interface interface is configured and started in the simulation node.

[0022] For the high-speed simulation link, this method employs shared memory, also known as the MEMIF mode, combined with a vector packet processing component to implement this simulation link. First, based on the connection relationships between nodes in the configuration information of the spatial network scenario to be simulated, simulation nodes at both ends of a simulation link are selected. Next, in the simulation node at one end of the simulation link, a MEMIF interface is created and a socket file is generated for negotiation and control channel establishment. Then, in the simulation link at the other end, a shared memory region is negotiated based on the socket file created by the previous simulation node, and a MEMIF interface is created accordingly. Finally, the MEMIF interface is activated in the simulation nodes at both ends of the simulation link to enable the interface.

[0023] The S3 part is implemented through the following configuration:

[0024] S3.1 Delay control of the simulated link: Within each time slot, the delay is calculated based on the actual distance of the simulated link in the spatial network scenario. The delay of this simulated link is then calculated using the following formula. ,

[0025] ,

[0026] in This represents the distance between the nodes at both ends of the simulated space network communication link, in meters. The unit representing signal propagation speed is meters per second. Link attribute configuration plugins are loaded into the vector data packet processing components at both ends of the simulation link to dynamically configure and update latency.

[0027] S3.2 Packet loss control of the simulation link: In each time slot, according to the pre-set packet loss rate plan of the simulation, the link attribute configuration plugin is loaded for the vector data packet processing component in the nodes at both ends of the simulation link to dynamically configure and update the packet loss rate.

[0028] The S4 part is implemented through the following configuration:

[0029] A designed smooth switching mechanism is adopted to perform simulation link switching according to the link switching time slot plan and the simulation link type. The smooth switching mechanism is characterized by establishing a new simulation link first, then guiding data from the old simulation link to the new one, and finally disconnecting the old simulation link. In space network simulation, a significant and crucial characteristic is the frequent switching of simulation links, a phenomenon widely observed in low-Earth orbit satellite networks, inter-satellite communication systems, and integrated ground-space networks. Therefore, this method completes the simulation link switching according to the simulation link switching time slot plan. Furthermore, the switching method varies depending on the simulation link type.

[0030] The implementation process of conventional simulation link switching described in S4.1 includes the following steps:

[0031] (1) Before the planned switchover time, the logical configuration of the new link is completed in the simulation nodes at both ends of the new simulation link through the vector data packet processing component, but it is not connected to the virtual switch for the time being.

[0032] (2) When the simulation time reaches the planned switching time, enable the logical ports of the simulation nodes at both ends of the new link on the virtual switch to realize the access of the new link;

[0033] (3) Disable the logical ports of the emulated nodes at both ends of the old link on the virtual switch to remove the old link and complete the smooth switch.

[0034] The implementation process of high-speed simulation link switching described in S4.2 includes the following steps:

[0035] (1) Before the planned switchover time, the new link is created and configured in advance in the simulation nodes at both ends of the new simulation link using the vector data packet processing component in memif mode;

[0036] (2) When the simulation time reaches the planned switching time, the corresponding memif interface is activated in the simulation nodes at both ends of the new simulation link through the vector data packet processing component to enable the new link;

[0037] (3) In the simulation nodes at both ends of the old simulation link, the corresponding memif interface is deleted by the vector data packet processing component, the old link is dismantled, and a smooth switch is completed.

[0038] The S5 part is implemented through the following configuration:

[0039] An automatic simulation data acquisition method is adopted. In simulation nodes where simulation data needs to be collected, a data capture plugin is loaded onto the vector data packet processing component to capture and analyze target data during the simulation process. This automatic data acquisition method automates the export of simulation data files by setting filtering conditions and acquisition limits during the simulation process. The implementation process includes: loading the data capture plugin onto the vector data packet processing component and selecting a predetermined data capture interface; setting filtering conditions for the captured data; setting an acquisition limit for data packets; and automatically exporting the captured data packets to a file once the predetermined target is reached. This file is then read to obtain the filtered data packet information from this interface, enabling the capture and analysis of data packets. Finally, the results are evaluated, and network simulation and analysis are performed.

[0040] The advantages of this invention compared to the prior art are:

[0041] This invention provides a high-speed processing and forwarding simulation method for space networks. While addressing the challenge of meeting the high-speed processing and forwarding requirements of communication links in high-throughput scenarios during space network simulation, this invention enables large-scale establishment and configuration of simulation nodes, large-scale establishment of simulation links, control of simulation link attributes, and flexible switching of simulation link topologies. Simultaneously, it utilizes automatic simulation data acquisition methods to capture and analyze target data during the simulation process. Therefore, this method can ensure the normal operation of large-scale space network simulation services while prioritizing high-speed data processing and forwarding, thus guaranteeing the realism and accuracy of the entire simulation process and possessing significant potential for wider application. Attached Figure Description

[0042] Figure 1 This is a structural diagram of a simulation method for high-speed processing and forwarding in space networks;

[0043] Figure 2 This is a connection diagram of a space communication network simulation scenario according to an embodiment of the present invention;

[0044] Figure 3 A schematic diagram of the conventional simulation link establishment structure in an embodiment of the present invention;

[0045] Figure 4 A schematic diagram of the high-speed simulation link establishment structure according to an embodiment of the present invention;

[0046] Figure 5 A comparison chart of bandwidth performance between the two simulated links of this invention and existing simulated links based on traditional kernel protocol stacks;

[0047] Figure 6 A schematic diagram of a conventional simulation link switching strategy according to an embodiment of the present invention;

[0048] Figure 7 A schematic diagram of the high-speed simulation link switching strategy according to an embodiment of the present invention;

[0049] Figure 8 Flowchart of data packet acquisition according to an embodiment of the present invention;

[0050] Figure 9 A diagram showing the packet capture and analysis results of an embodiment of the present invention. Detailed Implementation

[0051] The embodiments of the present invention will be further described below with reference to the accompanying drawings and specific implementation details.

[0052] Reference Appendix Figure 1 This paper introduces a simulation method for high-speed processing and forwarding in spatial networks. The entire process can be divided into the following five parts:

[0053] S1. Simulation Environment Initialization: For the spatial network scenario to be simulated, multiple spatial network simulation nodes are constructed, and vector data packet processing components are configured and initialized for each simulation node to achieve large-scale deployment of simulation nodes.

[0054] S2. Hybrid Simulation Link Construction: Construct a spatial network simulation topology between simulation nodes; establish a hybrid link architecture based on the service attributes of the spatial links: according to the expected communication load and performance requirements of the simulation links, divide all simulation links into two types: regular simulation links and high-speed simulation links, and implement this by configuring vector data packet processing components with corresponding working modes in the simulation nodes at both ends of the simulation links; manage the logical connection and disconnection of each simulation link through a virtual switch.

[0055] S3. Dynamic control of link attributes driven by discrete time slots: The spatial network simulation process is divided into discrete time slots; within each discrete time slot, the link attributes of each simulation link are updated according to preset scenario parameters; by loading the link attribute configuration plugin on the vector data packet processing component in the simulation node, the dynamic configuration and update of the link attributes of each discrete time slot are executed.

[0056] S4. Simulation Link Switching: Perform simulation link switching based on a smooth switching mechanism. According to the preset link switching time slot plan and simulation link type, the simulation link switching is achieved in conjunction with the vector data packet processing components in the simulation nodes at both ends of the simulation link. The smooth switching mechanism is as follows: at the planned switching time, a new simulation link is first established, then the data stream is migrated from the old simulation link to the new simulation link, and finally the old simulation link is dismantled.

[0057] S5. Automatic Data Acquisition: An automatic data acquisition method is adopted, in which a data capture plugin is loaded into the vector data packet processing component of the specified simulation node to capture and analyze the target data; the automatic data acquisition method refers to automatically exporting the captured data for analysis by following preset data filtering rules and acquisition termination conditions.

[0058] Furthermore, the specific configuration for step S1 is achieved through the following methods:

[0059] S1.1 Based on the configuration information of the space network scenario to be simulated, a simulation scenario including a space-based transmission network, small satellite links, and ground stations is constructed, and the correspondence between nodes in the space network scenario and network simulation nodes is confirmed. In this embodiment, refer to... Figure 2 The connection diagram of the space communication network simulation scenario is as follows: The space network simulation scenario in this embodiment is a typical communication constellation scenario. This scenario contains 35 simulation nodes, including 33 satellite simulation nodes and 2 ground receiving stations. Therefore, in the simulation, it is necessary to map the satellites and ground stations to the corresponding simulation nodes.

[0060] S1.2 For the spatial network scenario to be simulated, record the connection topology and time slots between the simulated nodes. In this embodiment, refer to... Figure 2 The diagram illustrates the connection relationships in a space communication network simulation scenario. There are 33 simulated satellite nodes belonging to 3 satellite orbits. As shown, the 3 orbits are separated by dashed lines. Each orbit hosts 11 satellites, with adjacent satellites connected end-to-end within each orbit. Satellites on different orbits are connected left-right. It is important to note that due to the presence of reverse seams, satellites on the leftmost and rightmost orbits (i.e., orbits 1 and 3) do not establish links with each other. Link establishment between the ground station and satellites follows this rule: at any given time, the ground station only connects to the nearest satellite within its line of sight, and does not establish connections with other satellites within its line of sight.

[0061] S1.3 For the connection topology to be simulated, record the type of each simulation link as either a conventional simulation link or a high-speed simulation link, including the corresponding link attributes. In this embodiment, the link connecting the ground station and the satellite, as well as some inter-satellite links, are considered high-speed simulation links, while other simulation links are considered conventional simulation links.

[0062] S1.4 Based on the types of simulation nodes in the space network scenario, different types of simulation nodes are created, corresponding simulation containers are established for the numerous satellite nodes, ground station nodes, and ground access user nodes in the space network scenario, thereby achieving large-scale deployment of simulation nodes. The simulation containers are deployed using containerization technology (Docker). Within each simulation node, a vector data packet processing tool is installed and enabled. In this embodiment, there are two different types of simulation nodes: satellite and ground station. Therefore, different container images are used to create the simulation containers. After creation, the vector data packet processing tool is started in each simulation container.

[0063] Step S2 is implemented through the following configuration:

[0064] Hybrid Simulation Link Construction: A spatial network simulation topology is built between simulation nodes. Based on the service attributes of the spatial links, a hybrid link architecture is established: two types of simulation links, namely regular simulation links and high-speed simulation links, are configured by configuring vector packet processing components in the simulation nodes at both ends of the simulation links. This satisfies the preset communication requirements while reducing resource overhead and improving large-scale deployment capabilities. The regular simulation link refers to a simulation link in the spatial network simulation that does not impose quantitative performance constraints on changes in transmission bandwidth and latency jitter. The simulation scenario uses this type of link by default to form the simulation network topology. For regular simulation links, the vector packet processing components in the simulation nodes at both ends are configured using the host-interface mode to create the link and connected to a virtual switch for topology management. The high-speed simulation link refers to a simulation link whose preset bandwidth requirement is higher than the preset bandwidth differentiation standard and / or whose preset latency jitter upper limit is not higher than the preset latency jitter differentiation standard. For high-speed simulation links, the vector packet processing components in the simulation nodes at both ends are configured using the shared memory-based memif mode to create the link.

[0065] In this embodiment, the preset bandwidth differentiation standard is 10Gbps, and the preset latency jitter differentiation standard is 10ms. Since the inter-satellite links between satellites are not at traffic aggregation points, they will not generate a large amount of data in a short period of time, and therefore there will be no high-traffic scenarios. Their bandwidth will not exceed 10Gbps, and the requirements for latency jitter are relatively broad, exceeding the 10ms latency jitter differentiation standard. Therefore, this type of simulated link is set as a conventional simulated link. On the other hand, the communication link between the two ground stations and the satellite carries all traffic interaction between the satellite and ground users. Its bandwidth requirement exceeds 10Gbps, or the latency jitter requirement is within 10ms. Therefore, it is set as a high-speed simulated link.

[0066] For the aforementioned conventional simulation link, this method uses a host-interface mode combined with a vector packet processing component to create the simulation link, and manages the link topology through a multi-layer virtual switch. First, based on the connection relationships between nodes in the configuration information of the required simulated spatial network scenario, simulation nodes at both ends of a simulation link are selected. Next, the host uses Ethernet pairs to connect these two simulation nodes to the multi-layer virtual switch. Then, in the simulation nodes at both ends of the simulation link, the host-interface is used to transfer the network interface card of the simulation node to the vector packet processing process. Finally, the host-interface interface is configured and started in the simulation node. In this embodiment, refer to... Figure 3 For inter-satellite links in scenarios where high traffic volumes are not expected, a host-interface model combined with a vector packet processing component is used to create this conventional simulation link. The specific structure is as follows: Figure 3 As shown.

[0067] For the high-speed simulation link, this method uses shared memory, also known as the MEMIF mode, combined with a vector packet processing component to implement this simulation link. First, based on the connection relationships between nodes in the configuration information of the spatial network scenario to be simulated, simulation nodes at both ends of a simulation link are selected. Next, in the simulation node at one end of the simulation link, a MEMIF interface is created and a socket file is generated for negotiation and control channel establishment. Then, in the simulation link at the other end, a shared memory region is negotiated based on the socket file created by the previous simulation node, and a MEMIF interface is created accordingly. Finally, the MEMIF interface is activated in the simulation nodes at both ends of the simulation link to enable the interface. In this embodiment, reference... Figure 4 For inter-satellite links with a stronger demand for high bandwidth and low latency, a shared memory mode combined with vector packet processing components is used to create this high-speed simulated link. The specific structure is as follows: Figure 4 As shown. Reference Figure 5 The performance of the two types of simulation links proposed in this scheme is compared with that of existing simulation links in current spatial network simulation methods based on traditional kernel protocol stacks. The results show that the forwarding bandwidth of both conventional and high-speed simulation links is significantly improved compared with existing simulation links. Specifically, the bandwidth of conventional simulation links is increased by about 68.8%, while the bandwidth of high-speed simulation links is increased by about 15 times.

[0068] The S3 part is implemented through the following configuration:

[0069] Compared to traditional terrestrial networks, space communication links face a more complex and variable physical environment. Their transmission characteristics are affected not only by distance but also by multiple factors, including link visibility, antenna pointing errors, electromagnetic interference, weather conditions, and even celestial motion. Therefore, idealized link models struggle to accurately reflect the operational state of space networks, necessitating the configuration of appropriate link characteristics during space network simulation. In space network communication, link distances typically range from hundreds to thousands of kilometers. For example, the one-way propagation delay between a low-Earth orbit satellite and the ground is approximately 5–20 milliseconds, and inter-satellite links (ISLs) may experience even higher delays. Inaccurate modeling of these delays can lead to distortions in network behavior simulations. Furthermore, space network communication introduces random errors due to factors such as the signal-to-noise ratio of the physical channel, link alignment errors, and atmospheric disturbances. Therefore, for the simulated links in the simulation system, a link configuration plugin is loaded into the vector data packet processing component to configure link attributes and meet the requirements of the simulated space network scenario. In this embodiment, there are three low-Earth orbit satellite orbits, with 11 satellites in each orbit. Therefore, the distance between two adjacent satellites is calculated to be approximately 3864 km, and the distance between the satellite and the connected ground station is approximately 800 km. Meanwhile, the packet loss rate in satellite network communication scenarios is typically around 1%.

[0070] S3.1 Delay control of the simulated link: Within each time slot, the delay is calculated based on the actual distance of the simulated link in the spatial network scenario. The delay of this simulated link is then calculated using the following formula. ,

[0071] ,

[0072] in This represents the distance between the nodes at both ends of the simulated network communication link in the simulated space, in meters. The unit representing signal propagation speed is meters per second. Link attribute configuration plugins are loaded into the vector data packet processing components at both ends of the simulation link to dynamically configure and update latency. In this embodiment, within a specific time slot, the calculated distance between two adjacent satellites is approximately 3864 km, and the distance between the satellite and the connected ground station is approximately 600 km. Therefore, the latency between the satellite simulation nodes is approximately 12.88 ms, and the latency between the satellite node and the connected ground station node is approximately 2 ms. In the next time slot, the distance between two adjacent satellites is approximately 4108 km, and the distance between the satellite and the connected ground station is approximately 800 km. Therefore, the latency between the satellite simulation nodes is approximately 13.69 ms, and the latency between the satellite node and the connected ground station node is approximately 2.67 ms, thus achieving dynamic configuration and updating of latency.

[0073] S3.2 Packet loss control of the simulation link: Within each time slot, according to the pre-set packet loss rate plan, link attribute configuration plugins are loaded for the vector data packet processing components at both ends of the simulation link to dynamically configure and update the packet loss rate. In this embodiment, within a specific time slot, a simulated link between two satellite simulation nodes is set with a packet loss rate of 1%. In the next time slot, due to the influence of the propagation medium, the packet loss rate spikes to 10%. Through this approach, dynamic configuration and updating of the packet loss rate are achieved.

[0074] The S4 part is implemented through the following configuration:

[0075] A designed smooth switching mechanism is adopted to perform simulated link switching according to the link switching time slot plan and the simulated link type. The smooth switching mechanism is characterized by establishing a new simulated link first, then guiding data from the old simulated link to the new simulated link, and finally disconnecting the old simulated link. In space network simulation, a significant and crucial characteristic is the frequent switching of links, a phenomenon widely observed in low-Earth orbit satellite networks, inter-satellite communication systems, and integrated ground-space networks. Therefore, this method completes the simulated link switching according to the simulated link switching time slot plan. However, the switching method varies depending on the type of simulated link. Compared to the relatively stable link status and slow topology changes in terrestrial networks, the link connections in space networks are highly dependent on the spatial motion state and physical visibility conditions of nodes, exhibiting significant time-varying, dynamic, and short-term effectiveness. Taking a low-Earth orbit (LEO) constellation system as an example, satellites orbit the Earth at high speeds, with orbital periods typically between 90 and 120 minutes. Due to the constantly changing line-of-sight distances between satellites and between satellites and ground stations, the available time for a link may only be a few minutes. After this time window, the link may become unusable due to exceeding the viewing angle, obstruction, or signal attenuation, necessitating a switch to a new link to maintain communication. Since the satellite trajectories are fixed, this method completes the communication link switchover according to a simulated link switching time slot plan. However, because different simulated link types exist, the switching methods also differ. In this embodiment, reference... Figure 2Because the satellite's position is constantly changing, the distance between the satellite node and the ground station node is also constantly changing. When the satellite's position moves to a certain extent, the satellite closest to the ground station node has changed. Therefore, a simulated link switch should be performed at this time. The link switch time slots in this embodiment are shown in Table 1, where P1-S7 represent the seventh satellite in the first operating orbit. For example, in this embodiment, during the time slot of 0-60 minutes, ground station 1 is connected to the seventh numbered satellite in the first satellite orbit, while ground station 2 is connected to the fourth numbered satellite in the third satellite orbit.

[0076] Table 1. Ground station communication link switching time slots:

[0077] Time (minutes) Satellite connected to ground station 1 Satellite connected to ground station 2 0 - 60 P1-S7 P3-S4 60 – 120 P1-S8 P3-S5 120 – 180 P1-S9 P3-S6 180 – 240 P1-S10 P3-S7 240 – 300 P1-S11 P3-S8 300 – 360 P1-S1 P3-S9 360 – 420 P1-S2 P3-S10 420 – 480 P1-S3 P3-S11 480 – 540 P1-S4 P3-S1 540 – 600 P1-S5 P3-S2

[0078] The implementation process of conventional simulation link switching described in S4.1 includes the following steps:

[0079] (1) Before the planned switchover time, the logical configuration of the new link is completed in the simulation nodes at both ends of the new simulation link through the vector data packet processing component, but it is not connected to the virtual switch for the time being.

[0080] (2) When the simulation time reaches the planned switching time, enable the logical ports of the simulation nodes at both ends of the new link on the virtual switch to realize the access of the new link;

[0081] (3) Disable the logical ports of the emulated nodes at both ends of the old link on the virtual switch to remove the old link and complete the smooth switch.

[0082] In this embodiment, the link switching between the ground station simulation node and the satellite simulation node can also use this switching strategy, as referenced. Figure 6 During the first 60 minutes of the simulation, ground station 1 connects to satellite P1-S7. Just before the 60-minute mark, the host computer uses Ethernet to connect the two simulation nodes (ground station 1 and satellite P1-S8) to the multilayer virtual switch. Then, in the simulation nodes of ground station 1 and satellite P1-S8, the host-interface is used to transfer the network interface card (NIC) to the vector packet processing process. The host-interface is configured accordingly in the simulation nodes to prepare for the simulation link switchover. Finally, when the simulation time reaches 60 minutes, the multilayer virtual switch opens all connections between the simulation nodes of ground station 1 and satellite P1-S8 and simultaneously closes all connections between the simulation nodes on both sides of the simulation link between ground station 1 and satellite P1-S7 and the multilayer virtual switch, thus achieving a smooth switching process for the simulation link.

[0083] This switching scheme also applies to the scenario of switching from a high-speed simulation link to a conventional simulation link. In this embodiment, if ground station 1 is connected to satellite P1-S7 during minutes 0-60, and the connected link is a high-speed simulation link for fast data exchange and forwarding, and during minutes 60-120, ground station 1 is connected to satellite P1-S8, without the need for high-speed data exchange, this scenario represents the switching from a high-speed simulation link to a conventional simulation link. In this case, firstly, just before the simulation reaches 60 minutes, the two simulation nodes connecting ground station 1 and satellite P1-S8 are connected to a multi-layer virtual switch via Ethernet on the host. Then, in the simulation nodes of ground station 1 and satellite P1-S8, the network interface card of the simulation node is transferred to the vector packet processing process using the host-interface. The host-interface interface is configured accordingly in the simulation node to prepare for the simulation link switching. Finally, when the simulation time reaches 60 minutes, in the multi-layer virtual switch, first open all connections between ground station 1 and satellite P1-S8 simulation nodes and the multi-layer virtual switch. Then, delete the corresponding memif interfaces in ground station 1 and satellite P1-S7 simulation nodes to simulate the disconnection of this simulation link at this moment, thereby realizing the simulation link switching process of this type.

[0084] The implementation process of high-speed simulation link switching described in S4.2 includes the following steps:

[0085] (1) Before the planned switchover time, the new link is created and configured in advance in the simulation nodes at both ends of the new simulation link using the vector data packet processing component in memif mode;

[0086] (2) When the simulation time reaches the planned switching time, the corresponding memif interface is activated in the simulation nodes at both ends of the new simulation link through the vector data packet processing component to enable the new link;

[0087] (3) In the simulation nodes at both ends of the old simulation link, the corresponding memif interface is deleted by the vector data packet processing component, the old link is dismantled, and a smooth switch is completed.

[0088] In this embodiment, the link switching between the ground station simulation node and the satellite simulation node can use this switching strategy, as referred to... Figure 7During the first 60 minutes of the simulation, ground station 2 connects to satellite P2-S2. Just before the 60-minute mark, ground station 2 and satellite P2-S3 negotiate a shared memory region based on the socket file created by the simulation nodes, and thus create a memif interface. The newly established memif interface is configured accordingly on the simulation nodes at both ends of the simulation link to prepare for link switching. When the simulation time reaches 60 minutes, the memif interface is activated on both ground station 2 and satellite P2-S3, simulating the establishment of a new simulation link. Then, the corresponding memif interface is deleted on both ground station 2 and satellite P2-S2, simulating the disconnection of this simulation link at that moment, thus simulating this type of simulation link switching process.

[0089] This switching scheme also applies to the scenario of switching from a conventional simulation link to a high-speed simulation link. In this embodiment, if ground station 2 is connected to satellite P2-S2 via a conventional simulation link during the first 60 minutes, and then connects to satellite P2-S3 via a high-speed data exchange during the second 60-120 minutes, this scenario represents a switch from a conventional simulation link to a high-speed simulation link. In this case, just before the simulation reaches 60 minutes, a shared memory region is negotiated between ground station 2 and satellite P2-S3 simulation nodes based on the socket file created by the simulation nodes. A memif interface is then created from this shared memory region. The newly established memif interface is then configured accordingly on the simulation nodes at both ends of the simulation link to prepare for the link switch. When the simulation time reaches 60 minutes, the memif interface is activated in both the ground station 2 and satellite P2-S3 simulation nodes to enable the interface and simulate the establishment of a new simulation link. At the same time, the connections between the simulation nodes on both sides of the simulation link between the ground station 2 and satellite P2-S2 and the multi-layer virtual switch are all closed, thereby achieving a smooth switching process of the simulation link.

[0090] Step S5 is implemented through the following specific configuration:

[0091] An automatic simulation data acquisition method is adopted, which involves loading a data capture plugin onto the vector data packet processing component in the simulation node to capture and analyze target data during the simulation process. This automatic data acquisition method automatically exports simulation data files by setting filtering conditions and acquisition limits during the simulation process. The implementation process includes: selecting a predetermined data capture interface after loading the data capture plugin onto the vector data packet processing component; setting filtering conditions for the captured data; setting an acquisition limit for data packets; and automatically exporting the captured data packets to a file once the predetermined target is reached. This file is then read to obtain the filtered data packet information from this interface, enabling the capture and analysis of data packets. Finally, the results are evaluated, and network simulation and analysis are performed.

[0092] In this embodiment, reference Figure 8 The capture and analysis of data packets adopts Figure 8 The procedures specified in [the document / document]. (Refer to [the document / document]). Figure 9 After configuration, it can capture data packets from specific network ports and save them as relevant files, such as analyze in the figure. Then, it can perform traffic analysis through network analysis software, thereby achieving the purpose of analyzing the network simulation effect.

Claims

1. A high-speed processing and forwarding simulation method for spatial networks, characterized in that, include: S1. Simulation Environment Initialization: For the spatial network scenario to be simulated, multiple spatial network simulation nodes are constructed, and vector data packet processing components are configured and initialized for each simulation node to achieve large-scale deployment of simulation nodes; S2. Hybrid Simulation Link Construction: Construct a spatial network simulation topology between simulation nodes; establish a hybrid link architecture based on the service attributes of the spatial links: according to the expected communication load and performance requirements of the simulation links, divide all simulation links into two types: regular simulation links and high-speed simulation links, and implement this by configuring vector data packet processing components with corresponding working modes in the simulation nodes at both ends of the simulation links; manage the logical connection and disconnection of each simulation link through a virtual switch; S3. Dynamic control of link attributes driven by discrete time slots: The spatial network simulation process is divided into discrete time slots; within each discrete time slot, the link attributes of each simulation link are updated according to preset scenario parameters; by loading a link attribute configuration plugin on the vector data packet processing component in the simulation node, the dynamic configuration and update of the link attributes of each discrete time slot are executed. S4. Simulation Link Switching: Perform simulation link switching based on a smooth switching mechanism. According to the preset link switching time slot plan and simulation link type, the simulation link switching is realized in conjunction with the vector data packet processing components in the simulation nodes at both ends of the simulation link. The smooth switching mechanism is as follows: at the planned switching time, a new simulation link is first established, then the data stream is migrated from the old simulation link to the new simulation link, and finally the old simulation link is dismantled. S5. Automatic Data Acquisition: An automatic data acquisition method is adopted, in which a data capture plugin is loaded into the vector data packet processing component of the specified simulation node to capture and analyze the target data; the automatic data acquisition method refers to automatically exporting the captured data for analysis by following preset data filtering rules and acquisition termination conditions.

2. The high-speed processing and forwarding simulation method for space networks according to claim 1, characterized in that, The conventional simulation link mentioned in step S2 refers to a simulation link in which no quantitative performance constraints are imposed on the changes in transmission bandwidth and latency jitter in spatial network simulation. The simulation scenario uses this type of link for networking by default to form a simulation network topology. For the conventional simulation link, the vector data packet processing components in the simulation nodes at both ends are configured in host-interface mode to create the link and connect to the virtual switch for topology management.

3. The high-speed processing and forwarding simulation method for space networks according to claim 1, characterized in that, The high-speed simulation link mentioned in step S2 refers to a simulation link where the preset bandwidth requirement is higher than the preset bandwidth differentiation standard, and / or the preset latency jitter limit is not higher than the preset latency jitter differentiation standard. For the high-speed simulation link, the vector packet processing components in the simulation nodes at both ends are configured in a shared memory-based memif mode to create the link.

4. The high-speed processing and forwarding simulation method for space networks according to claim 1, characterized in that, The dynamic control of link attributes driven by discrete time slots in step S3 specifically includes dynamic control of the following attributes: S3.

1. Propagation delay control of simulation link: In each simulation time slot, the propagation delay of the link is calculated based on the dynamic distance of the simulation link in the spatial scene, and the delay parameters of the vector data packet processing components in the simulation nodes at both ends of the simulation link are dynamically configured through the link attribute configuration plugin. S3.

2. Packet loss rate control of the simulation link: In each simulation time slot, according to the packet loss rate plan determined by the simulation script or dynamic model, the packet loss rate parameters of the vector data packet processing components in the simulation nodes at both ends of the simulation link are dynamically configured through the link attribute configuration plugin.

5. The high-speed processing and forwarding simulation method for space networks according to claim 1, characterized in that, In step S4, when implementing simulated link switching based on the aforementioned smooth switching mechanism, different operation procedures will be adopted depending on the type of simulated link: S4.

1. The switching procedure for a conventional simulation link is as follows: (1) Before the planned switchover time, the logical configuration of the new link is completed in the simulation nodes at both ends of the new simulation link through the vector data packet processing component, but it is not connected to the virtual switch for the time being. (2) When the simulation time reaches the planned switching time, enable the logical ports of the simulation nodes at both ends of the new link on the virtual switch to realize the access of the new link; (3) Disable the logical ports of the emulated nodes at both ends of the old link on the virtual switch to remove the old link and complete the smooth switch. S4.

2. The switching process for the high-speed simulation link is as follows: (1) Before the planned switchover time, the new link is created and configured in advance in the simulation nodes at both ends of the new simulation link using the vector data packet processing component in memif mode; (2) When the simulation time reaches the planned switching time, the corresponding memif interface is activated in the simulation nodes at both ends of the new simulation link through the vector data packet processing component to enable the new link; (3) In the simulation nodes at both ends of the old simulation link, the corresponding memif interface is deleted by the vector data packet processing component, the old link is dismantled, and a smooth switch is completed.

6. The high-speed processing and forwarding simulation method for space networks according to claim 1, characterized in that, The automatic data acquisition method described in step S5 includes the following steps: in the target simulation node, loading a data capture plugin for the vector data packet processing component and specifying the capture interface; setting data packet filtering rules to filter target data; setting acquisition termination conditions; when the captured data reaches the termination conditions, the plugin automatically exports the data to a specified file, and obtains network simulation results by analyzing the file.

Images