Giant Constellation Network Simulation Method, System and Storage Medium Based on Topological Partitioning

Abstract

The present invention provides a simulation method, system and storage medium for a giant constellation network based on topological partitioning. The method includes the following steps: obtaining an original topology in the scenario of a giant satellite constellation, and obtaining a preset target number of sub-topologies; statistically analyzing the global distribution and density of satellites according to the original topology, and dividing the original topology into the target number of sub-topologies according to the global distribution and density of satellites, wherein each sub-topology contains an equal number of satellite sampling points, and an event of a satellite updating its position once within a preset time length is recorded as a satellite sampling point; obtaining the target number of containers corresponding to the target number of sub-topologies, and respectively inputting each sub-topology into the corresponding container to perform a simulation operation. The present invention can achieve more efficient simulation of a giant constellation network based on a load-balanced topological partitioning strategy.

Description

Technical Field

[0001] The present invention relates to the technical field of giant constellation network simulation, and in particular to a method, system and storage medium for giant constellation network simulation based on topology partitioning. Background Art

[0002] Satellite Internet provides Internet access services globally through satellites. As an important component of the space-air-ground integrated network, satellite Internet is a current research hotspot. Many scholars have carried out research in the fields of constellation networking design, network protocol development, communication performance evaluation, etc., and their main research method is software simulation. Traditional serial single-machine network simulation simulators based on OPNET or NS3 cannot provide efficient simulation performance for giant satellite networks with a large number of nodes and wide coverage due to hardware conditions and software performance limitations. Even on a simulation system with a large amount of physical memory, the maximum number of networks that can be represented cannot carry the order of thousands of nodes of the simulator. The basic idea to solve this problem is to divide the network model into multiple sub-models, instantiate a separate simulator for each sub-model on different processors, and all sub-models are combined to form the original network model. However, when the partitioning strategy for sub-models is more unbalanced, the simulation acceleration effect is closer to serial single-machine simulation. NS3 is a discrete-event-based network simulator that can simulate various types and scales of networks in the real world on a single computer.

[0003] There are three partitioning strategies adopted in the prior art: (1) Adopt the graph partitioning strategy, but this strategy is limited to static network models, is prone to falling into local optimal solutions, and has a high computational complexity. (2) Adopt the uniform topology partitioning approach based on geographical location (UTPA), but this strategy has the problem of load imbalance because the number of satellites in each sub-topology is uneven due to the time-varying nature of satellites, and the simulation acceleration performance is poor. (3) Adopt the greedy algorithm to obtain a load-balanced topology partitioning strategy, but the process of iterative solution according to the optimization target in this strategy is prone to falling into local optimal solutions, with a complex calculation amount and general performance.

[0004] Therefore, in the simulation of giant constellation networks, how to provide a topology partitioning strategy that can reasonably allocate simulation events, ensure load balance of all simulation nodes, shorten the overall simulation running time, reduce computational resource consumption, and improve parallel network simulation performance is a technical problem to be solved urgently. Summary of the Invention

[0005] In view of this, embodiments of the present invention provide a method, system and storage medium for giant constellation network simulation based on topology partitioning to eliminate or improve one or more defects existing in the prior art.

[0006] One aspect of the present invention provides a method for simulating a giant constellation network based on topological partitioning, the method comprising the following steps:

[0007] Obtain the original topology in the giant satellite constellation scenario and obtain the preset number of sub-topology targets.

[0008] Statistically analyze the distribution and density of satellites globally based on the original topology, and divide the original topology into the target number of sub-topologies according to the distribution and density of satellites globally. Among them, each sub-topology contains an equal number of satellite sampling points internally. An event where a satellite updates its position once within a preset time length is denoted as a satellite sampling point.

[0009] Obtain the target number of containers corresponding to the target number of sub-topologies allocated, and input each sub-topology into the corresponding container to perform simulation operations.

[0010] In some embodiments of the present invention, dividing the original topology into the target number of sub-topologies according to the distribution and density of satellites globally includes: statistically analyzing the number of times each satellite in the original topology updates its position within a preset time length; calculating the range of the number of satellite sampling points that should be contained within each sub-topology; and dividing the original topology into the target number of sub-topologies according to the distribution and density of satellites globally and the range of the number of satellite sampling points that should be contained within each sub-topology.

[0011] In some embodiments of the present invention, dividing the original topology into the target number of sub-topologies according to the distribution and density of satellites globally and the range of the number of satellite sampling points that should be contained within each sub-topology includes: inputting the distribution and density of satellites globally and the requirements for the range of the number of satellite sampling points that should be contained within each sub-topology into a sub-topology partitioning model based on the greedy algorithm, so as to calculate and obtain the geographical range of longitude and latitude divisions for each sub-topology.

[0012] In some embodiments of the present invention, the target number is a positive integer that can be factorized; dividing the original topology into the target number of sub-topologies according to the distribution and density of satellites globally further includes: factorizing the target number into two positive integers, taking the maximum value of the two positive integers as the number of latitude intervals, and taking the minimum value of the two positive integers as the number of longitude intervals.

[0013] In some embodiments of the present invention, the step of dividing the original topology into the target number of sub-topologies according to the distribution and density of satellites globally and the range of the number of satellite sampling points that should be contained within each sub-topology includes: evenly dividing the original topology into the number of longitude intervals in the longitude direction according to the distribution and density of satellites globally and the range of the number of satellite sampling points that should be contained within each sub-topology, and dividing the original topology into the number of latitude intervals proportional to the satellite's orbital period between the negative orbital inclination and the positive orbital inclination in the latitude direction, to obtain the target number of sub-topologies.

[0014] In some embodiments of the present invention, after obtaining the target number of containers corresponding to the target number of sub-topologies and inputting each sub-topology into the corresponding container to perform the simulation operation steps, the method further includes: updating the sub-topology in each container according to a preset time slicing period, and updating the satellites in each container at the initial moment recorded.

[0015] In some embodiments of the present invention, after obtaining the target number of containers corresponding to the target number of sub-topologies and inputting each sub-topology into the corresponding container to perform the simulation operation steps, the method further includes: in the initialization mapping stage, combining the range of each sub-topology and the updated position of the satellites to record the satellites in each container at different moments; for each container, comparing the satellites at the current moment recorded with the satellites in each container at the initial moment, and counting the departure rate of the satellites leaving each container; when the departure rate of the satellites in the container reaches a preset threshold, updating the sub-topology in each container and updating the satellites in each container at the initial moment recorded.

[0016] In some embodiments of the present invention, after obtaining the target number of containers corresponding to the target number of sub-topologies and inputting each sub-topology into the corresponding container to perform the simulation operation steps, the method further includes: in the initialization mapping stage, combining the range of each sub-topology and the updated position of the satellites to record the satellites in each container at different moments; for each container, comparing the satellites at the current moment recorded with the satellites in each container at the initial moment, and counting the departure rate of the satellites leaving each container; updating the sub-topology in each container according to a preset time slicing period, and updating the satellites in each container at the initial moment recorded; wherein, based on the historical simulation data record, the time length required for the departure rate of the satellites in any one of all the containers to reach a preset percentage is set as the time slicing period.

[0017] On the other hand, the present invention provides a giant constellation network simulation system based on topology partitioning, including a processor and a memory. Computer instructions are stored in the memory, and the processor is used to execute the computer instructions stored in the memory. When the computer instructions are executed by the processor, the system implements the steps of any one of the above embodiments.

[0018] On the other hand, the present invention provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the steps of any one of the above embodiments.

[0019] The simulation method for a giant constellation network based on topology partitioning proposed by the present invention, based on the proposed concept of satellite sampling points, divides the original topology into a plurality of sub-topologies with balanced quantities according to the quantity and distribution of satellite sampling points, and simulates each sub-topology in a separate container, thereby jointly forming an original network model. The topology partitioning strategy with load balancing makes the parallel simulation acceleration efficiency of the method provided by the present invention significantly higher than that of other existing simulation methods for giant constellation networks. The present invention can significantly overcome the problems of poor simulation acceleration performance and high computational complexity.

[0020] Additional advantages, objects, and features of the present invention will be partially described below, and will become partially apparent to those of ordinary skill in the art after studying the following, or can be learned from the practice of the present invention. The objects and other advantages of the present invention can be achieved and obtained by the structures specifically pointed out in the specification and the drawings.

[0021] Those skilled in the art will understand that the objects and advantages that can be achieved by the present invention are not limited to the above specifically described, and the above and other objects that the present invention can achieve will be more clearly understood according to the following detailed description. Brief Description of the Drawings

[0022] The drawings described herein are used to provide a further understanding of the present invention, form a part of this application, and do not constitute a limitation to the present invention. The components in the drawings are not drawn to scale, but only to illustrate the principles of the present invention. For the convenience of illustrating and describing some parts of the present invention, the corresponding parts in the drawings may be enlarged, that is, may become larger relative to other components in the exemplary device actually manufactured according to the present invention. In the drawings:

[0023] Figure 1 It is a flowchart of the simulation method for a giant constellation network based on topology partitioning in an embodiment of the present invention.

[0024] Figure 2 It is a diagram of the giant constellation network architecture.

[0025] Figure 3 It is a communication flowchart under the giant constellation network.

[0026] Figure 4 It is a schematic diagram of satellite orbital parameters under the giant constellation network.

[0027] Figure 5 It is a schematic diagram of the partitioning result of the intelligent topology partitioning algorithm based on load balancing in an embodiment of the present invention.

[0028] Figure 6 It is a schematic diagram of the change in longitude and latitude information of satellite re-mapping in an embodiment of the present invention.

[0029] Figure 7 This is a comparison graph of the effects between an embodiment of the present invention and other existing simulation algorithms.

[0030] Figure 8 This is a comparison graph of the effects of embodiments of the present invention with different numbers of satellites.

[0031] Figure 9 This is an acceleration trend graph of embodiments of the present invention with different numbers of satellites. Detailed implementation manners

[0032] To make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below in conjunction with the implementation manners and the accompanying drawings. Herein, the illustrative implementation manners of the present invention and their descriptions are used to explain the present invention, but do not limit the present invention.

[0033] Herein, it should also be noted that in order to avoid obscuring the present invention due to unnecessary details, only the structures and / or processing steps closely related to the solution of the present invention are shown in the drawings, while other details less related to the present invention are omitted.

[0034] It should be emphasized that the term "comprising / including" when used herein refers to the presence of features, elements, steps or components, but does not exclude the presence or addition of one or more other features, elements, steps or components.

[0035] Herein, it should also be noted that if not otherwise specified, the term "connection" in this article can not only refer to direct connection, but also represent indirect connection with an intermediate.

[0036] In the following, embodiments of the present invention will be described with reference to the accompanying drawings. In the drawings, the same reference numerals represent the same or similar components, or the same or similar steps.

[0037] In order to overcome the problems existing in the existing simulation methods for mega satellite constellation networks, the present invention provides a simulation method, system and storage medium for mega satellite constellation networks based on topology partitioning. The method partitions the mega satellite constellation network based on geographical location, and takes into account the time-varying characteristics and operation rules of low-earth orbit mega satellites. Each sub-topology obtained by topology partitioning (TopologyPartitioning) is scheduled and allocated to a container for network simulation (NetworkSimulation). The method provided by the present invention can solve the problems of poor simulation acceleration performance and high computational complexity. Among them, the low-earth orbit mega satellite constellation has the characteristics of large bandwidth, high speed and low latency. Parallel Discrete Event Simulation (PDES) mainly parallelizes the simulation of discrete events through a multi-core processor.

[0038] The simulation method for mega satellite constellation networks provided by the present invention uses parallel discrete event simulation to accelerate the execution time of the satellite network protocol stack simulation model. Its main idea is to divide the network model into multiple sub-models, and instantiate a separate simulator for each sub-model on different processors. All sub-models jointly form the original network model. However, when the load of the partitioning strategy is more unbalanced, the simulation acceleration effect is closer to serial single-machine simulation. Therefore, the efficiency of parallel simulation is related to the partitioning strategy. Partitioning based on geographical location is an effective way to solve the partitioning strategy in parallel simulation. It divides the network topology to be simulated into several sub-topologies, and each sub-topology runs on different simulation nodes. Different sub-topologies avoid communication according to the loose coupling between different regions of the global satellite network. Among them, the result of topology partitioning directly determines the acceleration efficiency of parallel simulation.

[0039] Figure 1 The flowchart of the simulation method for mega satellite constellation networks based on topology partitioning in an embodiment of the present invention includes the following steps:

[0040] Step S110: Obtain the original topology in the mega satellite constellation scenario and obtain the preset target number of sub-topologies.

[0041] Step S120: Statistically analyze the global distribution and density of satellites according to the original topology, and divide the original topology into the target number of sub-topologies according to the global distribution and density of satellites. Among them, each sub-topology contains an equal number of satellite sampling points. An event in which a satellite updates its position once within a preset time length is recorded as a satellite sampling point.

[0042] Step S130: Obtain the target number of containers corresponding to the target number of sub-topologies, and input each sub-topology into the corresponding container to perform a simulation operation.

[0043] Among them, in step S120, dividing the original topology into the target number of sub-topologies according to the global distribution and density of satellites includes:

[0044] Step S121: Count the number of times each satellite in the original topology updates its position within a preset time period.

[0045] Step S122: Calculate the range of the number of satellite sampling points that should be included within each sub-topology.

[0046] Step S123: Divide the original topology into the target number of sub-topologies according to the global distribution and density of satellites and the range of the number of satellite sampling points that should be included within each sub-topology.

[0047] In an embodiment of the present invention, step S123 dividing the original topology into the target number of sub-topologies according to the global distribution and density of satellites and the range of the number of satellite sampling points that should be included within each sub-topology includes: inputting the global distribution and density of satellites and the requirement for the range of the number of satellite sampling points that should be included within each sub-topology into the sub-topology division model based on the greedy algorithm, so as to calculate and obtain the longitude and latitude division geographical ranges of each sub-topology. The greedy algorithm (Greedy Algorithm, GA) makes an optimal choice within each local range and finally obtains a global result. In the algorithm of ITPA, the greedy algorithm (Greedy Algorithm, GA) can be used to calculate the division result of the two-dimensional geographical location, obtain the geographical range after the division of longitude and latitude, and the computational complexity is relatively high but a similar effect can also be achieved. The parallel acceleration performance gap between the two methods is small.

[0048] In another embodiment of the present invention, the target quantity is a positive integer that can be factorized. In this embodiment of the invention, step S120 of dividing the original topology into the target number of sub-topologies according to the global distribution and density of satellites further includes: factorizing the target quantity into two positive integers, taking the maximum value of the two positive integers as the number of latitude intervals, and taking the minimum value of the two positive integers as the number of longitude intervals. This step is before step S121. Further, step S123 of dividing the original topology into the target number of sub-topologies according to the global distribution and density of satellites and the number range of satellite sampling points that should be included in each sub-topology includes: dividing the original topology evenly into the number of longitude intervals in the longitude direction and into the number of latitude intervals proportional to the satellite operation period between the negative orbital inclination and the positive orbital inclination in the latitude direction according to the global distribution and density of satellites and the number range of satellite sampling points that should be included in each sub-topology, so as to obtain the target number of sub-topologies.

[0049] In an embodiment of the present invention, after step S130 of obtaining the target number of containers corresponding to the target number of sub-topologies and inputting each sub-topology into the corresponding container to perform the simulation operation step, the method further includes: updating the sub-topology in each container according to a preset time sharding period, and updating the satellites in each container at the initial moment recorded.

[0050] In another embodiment of the present invention, after step S130 of obtaining the target number of containers corresponding to the target number of sub-topologies and inputting each sub-topology into the corresponding container to perform the simulation operation step, the method further includes: (1) in the initialization mapping stage, combining the range of each sub-topology and the updated position of the satellites to record the satellites in each container at different moments; (2) for each container, comparing the satellites at the current moment recorded with the satellites in each container at the initial moment, and statistically calculating the departure rate of the satellites leaving each container; (3) when the departure rate of the satellites in the container reaches a preset threshold, updating the sub-topology in each container and updating the satellites in each container at the initial moment recorded.

[0051] In another embodiment of the present invention, after obtaining the target number of containers corresponding to the target number of sub-topologies allocated in step S130 and respectively inputting each sub-topology into the corresponding container to perform the simulation operation steps, the method further includes: (1) In the initialization mapping stage, combine the range of each sub-topology and the updated position of the satellite to record the satellites in each container at different times; (2) For each container, compare the satellites recorded at the current time with the satellites in each container at the initial time, and count the departure rate of the satellites leaving each container; (3) Update the sub-topology in each container according to the preset time slice period, and update the satellites in each container recorded at the initial time; wherein, based on the historical simulation data record, the time length required for the departure rate of the satellites in any one of all containers to reach the preset percentage is set as the time slice period.

[0052] The giant constellation network simulation method of the present invention proposes an Intelligent Topology Partitioning Algorithm (ITPA). The satellite constellation network is partitioned based on geographical location, considering the time-varying characteristics and operating rules of low-earth-orbit giant satellites. Each sub-topology obtained by partitioning is allocated to a Docker container encapsulating OPNET for simulation. OPNET is a network simulation simulator that can accurately analyze the performance and behavior of complex networks. Standard or user-specified probes can be inserted at any position in the network model to collect data and perform statistics. Docker is a type of container technology. Docker packages the program and all its dependencies into a docker container, so that the program can have a consistent performance in any environment. Here, the dependencies for the program to run, that is, the container, is like a container, and the operating system environment where the container is located is like a cargo ship or a port. The performance of the program only relates to the container (container), and has nothing to do with which cargo ship or which port (operating system) the container is placed in. Therefore, Docker can shield environmental differences, that is to say, as long as the program is packaged into Docker, the behavior of the program will be the same no matter what environment it runs in. The topology partitioning algorithm includes the following steps: (1) Given the number of topology partitions, each sub-topology will be simulated by a container; (2) According to the global distribution and density of satellites, calculate the longitude and latitude coordinate ranges of the sub-topologies to ensure that the computing resource consumption in each sub-topology is basically the same; (3) Determine the time slice duration of the simulation time according to the dynamic trajectory of the satellite and the sub-topology range. (4) Start a parallel simulation process in each time slice duration in sequence.

[0053] The following introduction to the topology partitioning algorithm is divided into two parts - longitude and latitude partitioning and topology mapping process.

[0054] Figure 2 It is a network architecture diagram of a giant constellation. In the scenario of a giant satellite constellation, user terminals located on the ground establish communication with low-earth orbit (LEO) satellites located in the near-earth orbit and have data transmission and communication networking functions. The LEO satellites move around the earth. Figure 2 Inside the box is a line schematic diagram of the satellite position. R is the radius of the earth, h is the orbital altitude, and R + h is the orbital radius of the LEO satellite. On the right side of the box is a further line schematic diagram of the satellite position and orbit. The details of this structure can be seen in Figure 4 .

[0055] Figure 3 It is a communication flow chart under the giant constellation network. This system is a satellite constellation topology protocol stack simulation system built in the scenario of the giant constellation network. The node models and communication function modules in this system are as shown in Figure 3 . This system develops and builds a satellite constellation topology protocol stack simulation system according to the 3GPP standard, and realizes functions such as synchronization between satellites and users, system messages, registration, random access, data transmission, paging, measurement, and handover. This system can complete the basic communication processes in different scenarios. Specifically, the communication processes in the initial access stage include system message reception, cell selection, and public land mobile network (PLMN) selection. The communication processes in the registration process stage include random access, RRC establishment, and registration request. The communication processes in the paging process stage include paging reception and RAN paging. The communication processes in the service request stage include random access, RRC establishment, and service request. The communication processes in the measurement process stage include measurement, parameter configuration, and measurement reporting. The communication processes in the handover stage include random access, RRC process, and other processes, etc. The communication protocol layers of 3GPP include UE, NAS, RRC, SDAP, RLC, MAC, and the PHY layer. The entire original topology contains thousands of nodes with the above protocol stack simulation system. The original topology is divided into longitude and latitude based on geographical location, and the position mapping of satellite user equipment is carried out. 3GPP is an international standard organization formed by the cooperation of seven global standard-setting organizations (SSO). Currently, its members include ETSI in Europe, ARIB and TTC in Japan, CCSA in China, TTA in South Korea, ATIS in North America, and the Telecom Standards Development Society of India. Its purpose is very simple, which is to develop and maintain global wireless communication standards, rather than focusing on the needs of a certain local area or region.

[0056] Figure 4It is a schematic diagram of satellite orbital parameters under a giant constellation network. The low-earth-orbit giant constellation adopted in the embodiments of the present invention is composed of regularly distributed LEO (Low Earth Orbit) satellites. Generally, a Walker constellation networking scheme based on inclined orbits is used. The Walker constellation is a uniformly distributed model that covers the entire earth and consists of G orbits with S satellites in each orbit. The Walker constellation is such that the general satellite orbit is a circular orbit, the orbital planes are evenly distributed, and the satellites in the orbital planes are evenly distributed. The main characteristic parameters are shown in the following figure, including the number of orbits G, the phase factor F, the total number of satellites N, and the orbital inclination inc: the angle measured counterclockwise from the equatorial plane to the satellite orbital plane at the ascending node. The orbital height h, the true anomaly θ: the angle measured counterclockwise along the satellite's motion orbit from the perigee, the eccentricity e: a measure of the flatness of the ellipse, the argument of perigee, ω: the angle measured counterclockwise from the ascending node along the satellite's motion orbit to the perigee, the right ascension of the ascending node, Ω: the equatorial longitude of the satellite orbit's ascending node, etc. (In addition, λ0 represents the longitude of the satellite's position at t = 0, and W e represents the geocentric gravitational constant, V represents the satellite's operating speed, and the labels in the subscript position represent compound meanings. For example, d1 represents the first container).

[0057] Based on Figure 4 the structure of the giant constellation network shown, the longitude of satellite n at time t is defined as l λ (n, t), and the latitude of satellite n at simulation time t is The satellite's operating period is expressed as T s . The calculation of the true anomaly and the right ascension of the ascending node of each satellite is shown in the following formula. g ∈ {1, 2,..., G} represents the orbital identifier of the current satellite, and s ∈ {1, 2,..., N / G} represents the satellite identifier of the current satellite in the orbit. The longitude and latitude formulas are expressed as follows:

[0058]

[0059]

[0060] Latitude and longitude division process: According to the number of containers D, the global two-dimensional area is divided to achieve balanced computing resources in each area, so as to achieve the goal of simulation acceleration. Each area corresponds to a Docker container. To more clearly illustrate the resource consumption of satellites in the actual system simulation, the present invention proposes the concept of "satellite sampling". For example, for satellite n, with one second as the satellite simulation granularity, that is, the position of the satellite is updated once per second and the entire simulation process is completed. An instance of satellite n per second is called a satellite sampling point. For example, a satellite should have T satellite sampling points in a period T. Because the number of events processed by the simulator under the same device is the same at the moment corresponding to each sampling point, all satellite sampling points consume the same computing resources. Therefore, this article only needs to ensure that the number of satellite sampling points in each area is the same, that is, to ensure that each container consumes the same resources.

[0061] According to the parameter definition required to divide the original topology into sub-topologies, each divided sub-topology can be regarded as an independent subset, placed in each Docker container for simulation. It is defined that D Docker containers can be used, and each container d ∈ {1...D}. Two positive integers z1 and z2 are defined to represent the number of intervals for latitude and longitude division respectively, and D = z1 × z2. The number of intervals for latitude and longitude division constitutes the total number of containers. Represents the longitude division interval. Represents the latitude division interval.

[0062] Combined with the motion law of the Walker constellation, the present invention obtains the following two theories. Based on these two theories, the division of two-dimensional geographical locations is completed. The two theories are as follows:

[0063] Theory 1: For the Walker constellation scenario, uniform longitude division can always keep the number of satellite sampling points balanced in each divided sub-topology.

[0064] Theory 2: If the number of sampling points of a satellite in two different latitude ranges is equal, then no matter how many satellites there are, the sum of the sampling points in these two latitude ranges is equal.

[0065] Based on the above two conclusions, the division of longitude and latitude is carried out. First, since the number of containers D has been confirmed and D is not a prime number, D must be factorable and can be expressed as two positive integers z1 and z2, and D = z1 × z2. Also, because we hope that the satellite's residence time in a certain area d is as long as possible, according to the Walker constellation, max(z1, z2) needs to be set as the number of latitude intervals, and min(z1, z2) needs to be set as the number of longitude intervals. Then, according to Theory 1, the division of longitude is a uniform division between [-v, π] and is divided into min(z1, z2). Therefore, the boundaries of the longitude division are shown in formula (5), where d1 ∈ (1, 2, …, min(z1, z2) + 1). According to Conclusion 2, the division of latitude is a proportional division to the time of any satellite between [-inc, inc] and is divided into max(z1, z2). The time division result is shown in formula (6), where related to the orbital altitude h, and the boundaries of the latitude division are shown in formula (7). Where d2 ∈ (1, 2, …, max(z1, z2) + 1). According to x(d1) and y(d2), the longitude and latitude ranges of each area d can be obtained. For example, the d-th range is:

[0066]

[0067] Figure 5 This is a schematic diagram of the division result of the intelligent topology division algorithm based on load balancing in an embodiment of the present invention. Taking the number of containers as 20 as an example, this figure shows the distribution identification of each container and the size of the geographical partition. The abscissa corresponds to the value of longitude x(d1), and the division interval is 4. The ordinate corresponds to the value of latitude y(d2), and the division interval is 5. Load balancing is a concept of computer network technology used to distribute loads among multiple computers (computer clusters), network connections, CPUs, disk drives, or other resources to achieve optimal resource utilization, maximize throughput, minimize response time, and avoid overload at the same time.

[0068] In step S130 of the present invention, each sub-topology is placed in the corresponding container for protocol stack parallel simulation. The sub-topology corresponding to each container consists of a satellite and a user terminal. Each satellite and user execute the protocol stack standard network model (based on 3GPP) in sequence. Now, the actual simulation area of each container i is obtained. Next, the satellite and user terminal need to be mapped into these areas. Each satellite and user simulate the complete satellite network protocol stack. For satellite n according to longitude l λ (n,t) and latitude Confirm the container where each region i is located according to its range. Therefore, we define d1 ∈ (1, 2, …, min(z1, z2)+1) and d2 ∈ (1, 2, …, max(z1, z2)+1), satisfying the following:

[0069] x(d1) ≤ l λ (n, t) ≤ x(d1 + 1); (8)

[0070]

[0071] Figure 6 This is a schematic diagram of the change in the longitude and latitude information remapped by the satellite in an embodiment of the present invention. Δt is the mapping period (or time slice period). The three regions in the figure represent two remappings. The abscissa represents the longitude of the satellite's position, and the ordinate represents the latitude of the satellite's position. As shown in the above figure, due to the time-varying nature of the satellite (its meaning is time-selective fading, and the topological time-varying nature of the satellite network will affect the performance of the network, resulting in many existing network technologies being unable to be effectively used in satellite networks), the satellite will leave the position corresponding to the current container after a period of time. Therefore, the mapping needs to be re-completed every period of time Δt to update the satellite and position information. Of course, the time when each satellite enters or leaves the region can be set as the remapping time Δt. However, when the number of satellites is large, since the small number of satellites not in the region will not have too much impact on the global network simulation, the present invention assumes that when ρ% of the satellites change, the mapping operation is re-performed, and the mapping period Δt can be determined by the time-slicing algorithm (TSA, Time-slicing Algorithm). The calculation of the time slice period and the number of parallel simulations based on the TSA algorithm includes the following steps:

[0072] ① Initialize the mapping: This part maps different satellites and users to the corresponding geographical regions. First, determine whether satellite n is in container d according to formulas (1) and (2); then, record the satellites in each container at different times.

[0073] ② Remapping time: This part is used to calculate the time slice period Δt. First, calculate the satellites in each container at the initial moment; then calculate the moment when the satellite departure rate reaches ρ% as Δt; finally, select the largest Δt in different containers as the final Δt.

[0074] ③ Update the mapping: After calculating Δt, calculate the number of parallel simulations T n = t s / Δt, and re-determine the satellites and user nodes in different containers according to formulas (1) and (2), and re-generate the sub-topology information in different containers. Where t s represents the total time for parallel simulation.

[0075] For the load - balanced topology partitioning ITPA technology based on the Walker operation law proposed in the above - mentioned invention embodiments, through the two major principles of a Walker constellation, it overcomes the two major problems of computational complexity and performance, ensuring the advantages of load balance and simple calculation. The ITPA algorithm has a significant effect on parallel simulation acceleration. By comparing with the common uniform - partitioning - based method and the greedy algorithm based on optimization objectives, the present invention can bring better simulation acceleration effects and load - balancing characteristics.

[0076] Figure 7 This is a comparison chart of the effects of an embodiment of the present invention and other existing simulation algorithms. From Figure 7 it can be seen that the uniform topology partitioning does not consider the load - balancing degree between containers and does not consider the regularity and periodicity of satellites, so its performance is the worst; the greedy algorithm finds the optimal solution through iterations according to the optimization objective, but this algorithm is prone to falling into local optimal solutions and depends on the selection of the initial solution; while the proposed intelligent topology - partitioning algorithm fully combines the operation law of the satellite constellation, performs topology segmentation in combination with the satellite motion trajectory, ensuring the load balance and simulation efficiency of the divided sub - topologies in each container.

[0077] Figure 8 This is a comparison chart of the effects of embodiments of the present invention with different numbers of satellites. From Figure 8 it can be seen that the more satellites there are, the better the performance of the partitioning result and the more balanced the load in the container; the fewer the number of containers, the better the performance of the partitioning result and the more balanced the load in the container. However, considering that the higher the number of containers, the higher the hardware cost, so we determine the final number of containers by combining the maximization of resource utilization. Among them, in the scenario of 900 satellites, as the number of containers increases, the load - balancing factor is controlled at about 5%.

[0078] Figure 9 This is an acceleration trend chart of embodiments of the present invention with different numbers of satellites. Figure 9 It is a comparison chart of the performance improvement of parallel simulation and single - machine simulation with different numbers of satellites. First, calculate the number of containers required for different satellites to obtain the theoretical value of the improvement in parallel simulation efficiency; then calculate the satellite values in different divided containers, and calculate the corresponding computing resources in each container according to the satellites and users. Find the maximum computing resource as the overall computing resource value and compare it with the computing resource required for a single machine to obtain the simulation efficiency improvement multiple Speedup; finally, combine the parallel theoretical efficiency and the parallel actual efficiency to obtain Figure 9 the content shown in Figure 9 it can be seen that the parallel simulation performance of the ITPA algorithm used is close to the theoretical value, solving the problem of simulation acceleration of giant constellations.

[0079] In summary, the method for simulating a giant constellation network based on topological partitioning proposed by the present invention proposes a concept of "satellite sampling points". Based on the number and distribution of satellite sampling points, the original topology is divided into multiple sub-topologies with balanced quantities, and each sub-topology is placed in a separate container for simulation, thereby jointly forming the original network model. The load-balanced topological partitioning strategy makes the parallel simulation acceleration efficiency of the method provided by the present invention significantly higher than that of other existing methods for simulating giant constellation networks. The present invention can significantly overcome the problems of poor simulation acceleration performance and high computational complexity.

[0080] The topological partitioning algorithm of the present invention based on the characteristics of the walker constellation configuration applies the idea of parallel discrete event simulation to the giant constellation scenario, assigns the same number of simulation events to each sub-topology, realizes the optimal partitioning of the parallel network topology, maps each partitioned sub-topology to each container, and realizes the scheduling of the computational load balance of a large number of simulation events. The acceleration effect of the parallel simulation performance is close to the theoretical value. Through the optimization of topological partitioning in the parallel simulation environment, the resource balance of each sub-topology is improved, and combined with two conclusions discovered by the inventor, the intelligent topological partitioning with load balance can be conveniently realized, the network simulation performance is improved, and the bottleneck of the inability to quickly simulate large-scale satellite networks is solved.

[0081] Correspondingly, the present invention also provides a system for simulating a giant constellation network based on topological partitioning. The system includes a computer device, the computer device includes a processor and a memory, computer instructions are stored in the memory, and the processor is used to execute the computer instructions stored in the memory. When the computer instructions are executed by the processor, the system implements the steps of the method described above.

[0082] An embodiment of the present invention also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the method described above are implemented. The computer-readable storage medium may be a tangible storage medium, such as a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a floppy disk, a hard disk, a removable storage disk, a CD-ROM, or any other form of storage medium known in the art.

[0083] Those of ordinary skill in the art should understand that the various exemplary components, systems, and methods described in connection with the embodiments disclosed herein can be implemented in hardware, software, or a combination of both. Specifically, whether to implement it in hardware or software depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods for each specific application to implement the described functions, but such implementation should not be considered to exceed the scope of the present invention. When implemented in hardware, it can be, for example, an electronic circuit, an application-specific integrated circuit (ASIC), appropriate firmware, a plug-in, a functional card, and so on. When implemented in software, the elements of the present invention are programs or code segments used to perform the required tasks. The program or code segment can be stored in a machine-readable medium or transmitted through a data signal carried in a carrier wave over a transmission medium or a communication link.

[0084] It should be clear that the present invention is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present invention is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order between steps after understanding the spirit of the present invention.

[0085] In the present invention, the features described and / or illustrated for one embodiment can be used in the same or a similar manner in one or more other embodiments, and / or combined with the features of other embodiments or replace the features of other embodiments.

[0086] The above are only the preferred embodiments of the present invention and are not used to limit the present invention. For those skilled in the art, various changes and variations can be made to the embodiments of the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A simulation method for a giant constellation network based on topological partitioning, characterized in that, The method includes the following steps: Obtain the original topology in the scenario of a giant satellite constellation, and obtain the preset number of sub-topology targets; Statistically analyze the global distribution and density of satellites based on the original topology, and divide the original topology into the target number of sub-topologies according to the global distribution and density of satellites. Among them, each sub-topology contains an equal number of satellite sampling points internally. An event where a satellite updates its position once within a preset time length is recorded as a satellite sampling point; Obtain the target number of containers corresponding to the target number of sub-topologies allocated, and input each sub-topology into the corresponding container to perform simulation operations; The step of dividing the original topology into the target number of sub-topologies according to the global distribution and density of satellites includes: statistically analyzing the number of times each satellite in the original topology updates its position within a preset time length; calculating the range of the number of satellite sampling points that should be included in each sub-topology internally; dividing the original topology into the target number of sub-topologies according to the global distribution and density of satellites and the range of the number of satellite sampling points that should be included in each sub-topology internally; among them, the target number of sub-topologies is a positive integer that can be factorized. Factorize the target number into two positive integers, take the maximum value of the two positive integers as the number of latitude intervals, and take the minimum value of the two positive integers as the number of longitude intervals; The above step of dividing the original topology into the target number of sub-topologies according to the global distribution and density of satellites and the range of the number of satellite sampling points that should be included in each sub-topology internally includes: evenly dividing the original topology into the number of longitude intervals according to the global distribution and density of satellites and the range of the number of satellite sampling points that should be included in each sub-topology internally, and dividing it into the number of latitude intervals proportional to the satellite operation period between the negative orbital inclination and the positive orbital inclination in the latitude direction, to obtain the target number of sub-topologies.

2. The method for simulating a giant constellation network based on topological partitioning according to claim 1, wherein, The step of dividing the original topology into the target number of sub-topologies according to the global distribution and density of satellites and the range of the number of satellite sampling points that should be included in each sub-topology internally includes: Input the global distribution and density of satellites and the requirement for the range of the number of satellite sampling points that should be included in each sub-topology internally into the sub-topology division model based on the greedy algorithm, so as to calculate and obtain the longitude and latitude division geographical ranges of each sub-topology.

3. The method for simulating a giant constellation network based on topological partitioning according to claim 1, wherein After the step of obtaining the target number of containers corresponding to the target number of sub-topologies allocated, and inputting each sub-topology into the corresponding container to perform simulation operations, the method further includes: Update the sub-topology in each container according to the preset time slice period, and update the satellites in each container at the initial moment recorded.

4. The method for simulating a giant constellation network based on topological partitioning according to claim 1, wherein After the step of obtaining the target number of containers corresponding to the target number of sub-topologies allocated, and inputting each sub-topology into the corresponding container to perform simulation operations, the method further includes: In the initialization mapping stage, combine the range of each sub-topology and the updated position of the satellite to record the satellites in each container at different moments; For each container, compare the satellites recorded at the current moment with the satellites in each container at the initial moment, and statistically analyze the departure rate of the satellites leaving each container; When the departure rate of the satellites of a container reaches a preset threshold, update the sub-topology in each container and update the satellites in each container at the recorded initial moment.

5. The method for simulating a giant constellation network based on topological partitioning according to claim 1, wherein After obtaining and allocating the target number of containers corresponding to the target number of sub-topologies and respectively inputting each sub-topology into the corresponding container to perform the simulation operation steps, the method further includes: In the initialization mapping stage, combine the range of each sub-topology and the updated positions of the satellites to record the satellites in each container at different moments; For each container, compare the satellites recorded at the current moment with the satellites in each container at the initial moment, and count the departure rate of the satellites leaving each container; Update the sub-topology in each container according to a preset time slice period, and update the satellites in each container at the recorded initial moment; wherein, based on the historical simulation data record, set the time length required for the departure rate of the satellites in any one of all containers to reach a preset percentage as the time slice period.

6. A giant constellation network simulation system based on topological partitioning, comprising a processor and a memory, characterized in that Computer instructions are stored in the memory, and the processor is configured to execute the computer instructions stored in the memory. When the computer instructions are executed by the processor, the system implements the steps of the method according to any one of claims 1 to 5.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps of the method according to any one of claims 1 to 5.

Images