A routing partitioning method and apparatus for large-scale low-Earth orbit satellite networks

Abstract

This invention discloses a routing partitioning method and apparatus for large-scale low-Earth orbit satellite networks, primarily addressing the problems of high link overhead and slow convergence speed in existing satellite networks. The scheme includes: generating a spatial network topology based on constellation size, number of orbital planes, orbital altitude, and orbital inclination; dividing the constellation into multiple routing regions based on the relative positions of the satellites, establishing link state information exchange rules, a routing table storage and update mechanism, and a data packet forwarding mechanism; establishing a quantitative relationship between the number of partitions and link overhead based on the network topology, routing region partitioning results, and link state information exchange rules; establishing a quantitative relationship between the number of partitions and routing convergence time based on the dynamic topology change characteristics of the satellites, routing region partitioning results, and routing table storage and update mechanism; calculating link overhead and routing convergence time under different numbers of partitions, constructing routing performance indicators, and outputting the optimal routing partitioning result. This invention can reduce link overhead and accelerate routing convergence speed, and can be applied to large-scale satellite network communication systems.

Description

Technical Field

[0001] This invention belongs to the field of satellite network technology, and specifically relates to a routing partitioning method and apparatus that can be used in large-scale satellite network communication systems. Background Technology

[0002] With the rapid development of aerospace technology and the decline in satellite manufacturing costs, large-scale satellite networks have become a critical infrastructure for ubiquitous global communication, remote sensing, and navigation enhancement. These networks consist of hundreds or thousands of satellites operating collaboratively at high speed in orbit, forming a complex system with rapidly changing topology and relatively limited resource distribution. In this context, accurate and efficient data transmission directly depends on the timely updates and robust maintenance of routing protocols. Traditional routing mechanisms, especially global link-state update strategies, require each node in the network to frequently collect, exchange, and calculate link-state information across the entire network. However, in large-scale satellite networks, this global update model faces severe challenges: First, the high-speed time-varying nature of satellite network topology requires nodes to exchange routing information at extremely high frequencies to maintain the accuracy of routing tables, inevitably consuming scarce inter-satellite bandwidth and onboard computing resources; second, the flood propagation of global information generates exponentially increasing signaling overhead as the network scales up, easily leading to network congestion and significantly reducing overall network performance and service quality; finally, the high latency characteristics of long-distance transmission cause a lag in the synchronization of global information, making it difficult to support services with extremely high timeliness requirements. To address these challenges, researchers have begun exploring partitioned routing mechanisms. These mechanisms aim to divide the vast satellite network into multiple relatively independent logical regions, restricting the updating and maintenance of link-state information to a local scope, thereby suppressing signaling flooding and reducing computational complexity. This approach provides a new solution for balancing routing performance and resource overhead, and has become an important direction in current research on routing technologies for large-scale dynamic satellite networks.

[0003] To ensure efficient utilization of link bandwidth resources and rapid route convergence in large-scale satellite networks, partitioned routing technology has received widespread attention in recent years. S. Li et al., in their paper "A multi-region division routing algorithm based on fuzzy-shortest-path-first for LEO satellite networks," in Proc. Inter. Wireless Commun. and Mobile Comput. (IWCMC), Marrakesh, Morocco, Jun. 2023, pp. 800–805, proposed a region-based routing scheme that utilizes constellation symmetry for graph modeling and hop count constraints, using these as a basis for dividing routing regions. J. Jiao et al., in their paper "Clustered multi-criteria routing algorithm for mega low earth orbit satellite constellations," IEEE Trans. Veh. Technol., vol. 73, no. 9, pp. 13790-13803, Sep. 2024, employed a dynamic partitioning mechanism based on geographic location, orbital velocity vector, and line-of-sight (LOS) connectivity. L. Wang et al., in their paper "Adaptive loadbalancing routing algorithm for low earth orbit satellite cluster networks," in Proc. Inter. Conf. on Comput. and Commun. Systems (ICCCS), Xi'an, China, Apr. 2024, pp. 666-671, performed adaptive partitioning based on real-time network conditions (such as traffic load and link quality) to balance signaling overhead and routing efficiency in large-scale networks.

[0004] Patent document CN202311370318.4 discloses a functional domain routing method for giant constellations based on a non-dominated sorting genetic algorithm. This method divides constellations into multiple functional domains based on functions such as communication, navigation, and remote sensing. When the number of nodes in a functional domain exceeds 200, it further splits it into scale domains. Boundary satellites establish inter-domain neighbor relationships and synchronize link states. Inter-domain topology is constructed using links with minimum cost, and Dijkstra's algorithm is used to calculate the shortest cross-domain path without loops. The total routing cost and the maximum number of hops across the entire network are set as dual optimization objectives to obtain the optimal domain scale.

[0005] However, the partitioning analysis models described above all have the following limitations:

[0006] First, the relationship between the number of partitions and routing overhead and routing convergence time has not been established, making it difficult to explore the impact of the number of partitions on network performance, thus limiting its theoretical guidance.

[0007] Secondly, determining the optimal number of partitions lacks adaptability to network size and constellation configuration, and cannot dynamically adjust under different total number of nodes or orbital geometry to maintain optimal performance.

[0008] Third, in terms of experimental design, none of them involved large-scale multi-layer satellite networks, making it difficult to truly reflect the topological changes and routing calculations in large-scale satellite networks, and the layout of their experimental conclusions lacks universality. Summary of the Invention

[0009] The purpose of this invention is to address the shortcomings of the prior art by proposing a routing partitioning method and apparatus for large-scale low-Earth orbit satellite networks. This method establishes the relationship between the number of partitions and routing overhead and routing convergence time, clarifies the impact of the number of partitions on network performance, enables dynamic adaptive adjustment of the number of partitions under different total number of nodes and orbital geometry, and completes experimental verification of performance improvement in a large-scale multi-layer satellite network scenario.

[0010] The technical approach to achieving the objective of this invention is as follows: Based on constellation size and orbital parameters, a satellite network topology and partition analysis model is constructed; an intra-domain and inter-domain collaborative routing mechanism is established; by deriving the quantitative relationship between the number of partitions and link overhead and routing convergence time, a clear law governing the impact of the number of partitions on network performance is obtained; by designing an optimal partition routing decision algorithm, the number of partitions is dynamically adjusted according to the network node size and orbital geometry; and by running the optimal partition routing decision algorithm in a large-scale, multi-layer satellite network scenario, performance simulation and verification are completed.

[0011] Based on the above approach, the technical solutions for achieving the objectives of this invention include the following:

[0012] 1. A routing partitioning method for large-scale low-Earth orbit satellite networks, characterized by comprising:

[0013] (1) Based on the constellation size N, the number of orbital planes P, the orbital altitude H, and the orbital inclination A, the spatial arrangement of satellite nodes and the orbital geometric constraints can be determined, the relative positions between satellites can be calculated, the visual relationships and inter-satellite link connectivity can be determined, and then a complete constellation spatial network topology can be generated to obtain the satellite network topology model.

[0014] (2) Based on the relative positions of satellites in the satellite network topology model, the constellation is divided into routing areas including the x-direction. Number of routing areas in the y direction Multiple routing areas Z, and satisfying ;

[0015] (3) Based on the relative positions between satellites, the visibility relationship, the inter-satellite link connectivity and multiple routing areas Z in the satellite network topology model, establish a satellite network routing model that includes three parts: link status information interaction rules, routing table storage and update mechanism and data packet forwarding.

[0016] (4) Based on the network topology, routing area division results, and link status information exchange rules in the satellite network routing model, establish a quantitative relationship between the number of partitions and link overhead.

[0017] (5) Based on the dynamic topology change characteristics of the satellite network topology model, the routing area division results, and the routing table storage and update method of the satellite network routing model, establish a quantitative relationship between the number of partitions and the routing convergence time.

[0018] (6) Based on the quantitative relationship between the number of partitions, link overhead and route convergence time, calculate the link overhead and route convergence time under different numbers of partitions, calculate the comprehensive route performance index based on the route convergence time and link overhead, and minimize it to obtain the optimal route partitioning result of the satellite network.

[0019] Furthermore, establishing the quantitative relationship between the number of partitions and link overhead in step (4) includes:

[0020] (4a) Number of partitions and intra-domain link overhead:

[0021] (4a1) Calculate the size of each routing area based on four parameters: number of orbits, number of satellites per orbit, number of routing areas in the x-direction, and number of routing areas in the y-direction. :

[0022] (4a2) Based on the size of each routing area, establish a quantitative relationship between the number of partitions and the intra-domain link cost. :

[0023] (4b) Establishing link overhead between adjacent domains:

[0024] (4b1) Calculate the size of the inter-domain subnet based on four parameters: number of orbits, number of satellites per orbit, number of routing areas in the x-direction, and number of routing areas in the y-direction. :

[0025] (4b2) Based on the size of the subnet formed between each adjacent routing area, establish a quantitative relationship between the number of partitions and the link cost between adjacent domains. :

[0026] (4c) Establishing the cost of links between non-adjacent domains:

[0027] (4c1) Calculate the size of the non-adjacent inter-domain subnet based on four parameters: number of orbits, number of satellites per orbit, number of routing areas in the x-direction, and number of routing areas in the y-direction. ;

[0028] (4c2) Based on the size of the subnets between non-adjacent domains, establish a quantitative relationship between the number of partitions and the link overhead between non-adjacent domains. ;

[0029] (4d) Based on the above results, the quantitative relationship between the number of partitions and the link cost is obtained. ;

[0030] Furthermore, establishing the quantitative relationship between the number of partitions and the routing convergence time in step (5) includes:

[0031] (5a) Based on message transmission interval and decision threshold Determine the time for satellite to detect local link status ;

[0032] (5b) Based on the number of orbits P, the number of satellites S in each orbit, and the number of routing areas in the x-direction. Number of routing areas in the y direction The length of the routing area Width of the routing area Propagation delay of the same track link and the propagation delay of the heterogeneous link Determine the propagation delay required for satellites within the routing area to exchange link state information. Propagation delay required for link state information exchange with regional boundary satellites :

[0033] (5c) Based on the computational efficiency constant of the path selection algorithm within the routing area The computational efficiency constant of the inter-regional path selection algorithm Determine the time required to recalculate the routing table. :

[0034] (5d) Based on the above results, the quantitative relationship between the number of partitions and the routing convergence time is obtained. .

[0035] 2. A routing partitioning device for a large-scale low-Earth orbit satellite network, characterized in that it comprises:

[0036] The topology modeling module is used to determine the spatial arrangement of satellite nodes and their orbital geometric constraints based on constellation size, number of orbital planes, orbital altitude, and orbital inclination. It also calculates the relative positions, visibility relationships, and inter-satellite link connectivity between satellites, generates a complete constellation spatial network topology, and obtains a satellite network topology model.

[0037] The region division module is used to divide the constellation into multiple routing regions Z, including the number of routing regions in the x-direction and the number of routing regions in the y-direction, based on the relative positions between satellites in the satellite network topology model.

[0038] The routing model construction module is used to establish a satellite network routing model based on the relative positions between satellites, visibility relationships, inter-satellite link connectivity, and multiple routing areas Z in the satellite network topology model. This model includes three parts: link state information exchange rules, routing table storage and update mechanism, and data packet forwarding mechanism.

[0039] The link cost quantization module is used to establish a quantitative relationship between the number of partitions and the link cost based on the network topology, routing area division results, and link state information interaction rules in the satellite network routing model.

[0040] The route convergence time quantization module is used to establish a quantitative relationship between the number of partitions and the route convergence time based on the dynamic topology change characteristics of the satellite network topology model, the routing area division results, and the routing table storage update method of the satellite network routing model.

[0041] The optimal partition number decision module is used to calculate the link cost and routing convergence time under different partition numbers based on the quantitative relationship between the number of partitions, link cost, and routing convergence time, minimize the overall routing performance index, and obtain the optimal routing partitioning result for the satellite network.

[0042] Compared with existing methods, the present invention has the following advantages:

[0043] First, by constructing a satellite network topology model, dividing routing areas, and establishing a satellite network routing model, this invention achieves precise quantification of link overhead and routing convergence time under different numbers of partitions, based on link state information interaction rules, routing table storage and update mechanisms, and data packet forwarding mechanisms.

[0044] Second, by establishing a quantitative relationship between the number of partitions and link overhead and route convergence time, this invention achieves significant optimization of the two performance indicators of link overhead and route convergence time, resulting in high resource utilization efficiency and stronger routing response capability.

[0045] Third, by outputting the optimal routing partitioning results under different network sizes and constellation configurations, and by continuously maintaining stable performance gains under high service load scenarios, this invention achieves strong robustness and scalability of the routing partitioning method for large-scale satellite networks.

[0046] Simulation results show that the link overhead and routing convergence time of the present invention are significantly reduced compared with existing methods, while the throughput is significantly improved compared with existing methods. Attached Figure Description

[0047] Figure 1 This is a flowchart illustrating the overall implementation of the large-scale low-Earth orbit satellite network routing partitioning method of the present invention.

[0048] Figure 2 These are diagrams illustrating two typical partitioning scenarios in the method of this invention;

[0049] Figure 3 This is a schematic diagram of the routing table storage and packet forwarding mechanism in the method of this invention;

[0050] Figure 4 This is a flowchart of the optimal partition routing scheme decision sub-process in the method of this invention;

[0051] Figure 5 This is a schematic diagram of the routing partitioning device for a large-scale low-orbit satellite network of the present invention;

[0052] Figure 6 This is a performance simulation diagram using the present invention under different numbers of partitions in the same network size;

[0053] Figure 7 This is a simulation diagram of the optimal number of regions using the present invention under different network sizes and constellation configurations;

[0054] Figure 8 These are performance simulation graphs of different algorithms under different network sizes and different service loads. Detailed Implementation

[0055] To enable those skilled in the art to better understand the present invention, the technical solutions and effects of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention and not all embodiments. Based on the embodiments of the present invention, other embodiments obtained by those skilled in the art without creative effort should all fall within the protection scope of the present invention.

[0056] Example 1: Routing partitioning method for large-scale low-Earth orbit satellite networks.

[0057] Reference Figure 1 The implementation steps of this example include the following:

[0058] Step 1: Construct a satellite network topology model.

[0059] 1.1) Based on the known constellation size N, number of orbital planes P, orbital altitude H, and orbital inclination A, determine the spatial arrangement of satellite nodes and their orbital geometric constraints, calculate the relative positions between satellites, determine their visibility relationships, and establish inter-satellite link connectivity:

[0060] 1.1.1) Calculate the satellite's three-dimensional position coordinates in the geocentric inertial coordinate system using the orbital motion equations, and obtain the satellite's position coordinates in the geocentric Earth-fixed coordinate system through coordinate transformation. Based on the position coordinates of the two satellites... The relative distance between satellites was calculated. Relative azimuth angle between the satellite and :

[0061] ,

[0062] ;

[0063] The relative positions of satellites can be determined based on their relative distances and relative azimuths.

[0064] 1.1.2) Based on Earth's radius Given the positions A and B of the two satellites, calculate the shortest distance between the line connecting the satellites and the Earth's surface. Assuming the geocentric coordinates are O, let the position vector of satellite A be... The position vector of satellite B Calculate the shortest distance between the satellite alignment and the Earth's surface. :

[0065] ;

[0066] 1.1.3) Let A safety distance is reserved for the Earth's radius and the effects of atmospheric refraction, and this distance is compared with the aforementioned shortest distance. By comparing the data, the visual relationship between satellite A and satellite B can be determined:

[0067] like If so, then satellite A and satellite B are determined to be in a visible relationship;

[0068] like Satellite A and Satellite B are determined to be in a non-visual relationship;

[0069] 1.1.4) Let the maximum communication distance threshold for inter-satellite links be... The maximum load limit for a single link is The relative distance between it and the satellite Real-time load on the link Compare and determine the connectivity between satellite A and satellite B:

[0070] like and If both conditions are met, it is determined that satellite A and satellite B are in a connectable state and an inter-satellite link can be established.

[0071] like or If any one of the conditions is met, it is determined that satellite A and satellite B are in a state of no communication and an inter-satellite link cannot be established between them.

[0072] 1.2) Based on the known constellation size N and the number of orbital planes P, calculate the number of satellites in each orbital plane. ;

[0073] 1.3) Based on the number of orbital planes P and the number of satellites S in each orbital plane, let... Represents a set of orbits. Represents the set of satellites within each orbit. This represents the nth satellite in the m-th orbit of the constellation, where and :

[0074] 1.4) Based on the parameters in 1.3) And in section 1.1.4), the inter-satellite link connectivity is determined, and four inter-satellite links are established for each satellite. Each inter-satellite link is represented as follows: ,in Indicates the source satellite of the inter-satellite link. Representing the peer satellite of the inter-satellite link, the satellite network topology model is completed after the four inter-satellite links of all satellites are established.

[0075] Step 2: Based on the satellite network topology model, divide the constellation into multiple routing areas.

[0076] 2.1) Using Define satellite networks: This represents the total number of routing areas. This represents the number of routing areas in the x-direction. Let be the number of routing areas in the y-direction, and let the three satisfy the following relationship: ;

[0077] 2.2) Perform routing area division in both the x and y directions:

[0078] Reference Figure 2 The implementation of this step includes the following:

[0079] 2.2.1) Using the orbital plane distribution of the satellite constellation as the basis for x-direction partitioning, the orbital planes are divided into several consecutive groups according to the total number of orbital planes and the distribution interval. All satellites on the orbital planes in each group are assigned to the same x-direction routing area.

[0080] 2.2.2) Based on the phase distribution of satellites in the same orbit as the basis for dividing in the y direction, according to the relative position of satellites in the same orbit, satellites in the same orbit are divided into several consecutive groups according to their orbital numbers, and satellites at different positions in the same orbital plane are assigned to the same y-direction routing area.

[0081] 2.2.3) After the routing area is divided, the size of each routing area is denoted as... ,in, For the length of the routing area, The width of the routing area is denoted as the size of the subnet between adjacent domains. ,in, The length of the subnet between adjacent domains, This represents the width of the subnet between adjacent domains.

[0082] Step 3: Construct a satellite network routing model that includes three parts: link state information exchange rules, routing table storage and update mechanism, and packet forwarding mechanism.

[0083] Reference Figure 3 The implementation of this step includes the following:

[0084] 3.1) Establish link state information exchange rules, including link state information exchange within the routing area, link state information exchange between adjacent domains, and link state information exchange between non-adjacent domains:

[0085] The link status information exchange within the routing area refers to the satellite sending link status information to neighboring satellite nodes within the area. The link status information is only propagated within the current routing area and is not forwarded outside the area to reduce link overhead. Through the link status information exchange within the routing area, all satellite nodes within the area can obtain the link status information of the area, providing basic data for the calculation of the routing table within the area.

[0086] The link status information exchange between adjacent domains is achieved by forming a subnet with satellites at the boundary of the routing area. All satellites within this subnet send link status information to neighboring satellite nodes without forwarding it outside the subnet, thereby reducing link overhead. Through the exchange of link status information between adjacent domains, satellites at the boundary of the routing area can grasp the link status of inter-regional links, providing basic data for calculating inter-regional routing tables.

[0087] The link status information exchange between non-adjacent domains involves selecting one satellite as a gateway satellite in each routing area, and forming a subnet with all the gateway satellites in all routing areas. Within this subnet, only reachability information between routing areas is exchanged to reduce cross-regional link overhead. Through the link status information exchange between non-adjacent domains, each gateway satellite can grasp the reachability status of non-adjacent areas, providing basic data for inter-regional routing table calculation.

[0088] 3.2) Establish a routing table storage update mechanism that includes both a routing table storage mechanism and a routing table storage mechanism:

[0089] The routing table storage mechanism refers to the fact that each satellite stores a routing table within its region and a routing table between regions: the routing table within the region stores the next-hop satellite of the relay path from the source satellite to any destination satellite within the region; the routing table between regions stores relay satellite information from the source satellite to any other routing region.

[0090] The routing table update mechanism refers to each satellite recalculating the intra-regional routing table and the inter-regional routing table: when an inter-satellite link within a region is interrupted or its quality deteriorates, the satellites exchange link status information within the routing region. After the link status information exchange is completed, the satellites recalculate the intra-regional routing table to ensure normal data forwarding within the routing region; when the link status of adjacent regions changes, the satellites at the region boundary exchange link status information between adjacent regions. After the link status information exchange is completed, the satellites recalculate the inter-regional routing table to ensure normal data forwarding across routing regions.

[0091] 3.3) Configure a packet forwarding mechanism that includes intra-area forwarding and inter-area forwarding:

[0092] The data packet forwarding within the region involves the source satellite, located in the same routing region as the destination satellite, querying the routing table within the region and forwarding the data packet to the relay satellite within the region. The relay satellite within the region then queries the routing table within the region and finally forwards the data packet to the destination satellite.

[0093] The inter-regional data packet forwarding process involves the source satellite, located in a different routing region from the destination satellite, querying the inter-regional routing table and forwarding the data packet to the inter-regional relay satellite. The inter-regional relay satellite then queries its intra-regional routing table and finally forwards the data packet to the destination satellite.

[0094] Step 4: Establish a quantitative relationship between the number of partitions and link overhead.

[0095] The link cost comprises three parts: intra-domain link cost, inter-adjacent domain link cost, and non-adjacent domain link cost. This step involves establishing a quantitative relationship between the number of partitions and these three components, which is implemented as follows:

[0096] 4.1) Establish a quantitative relationship between the number of partitions and the intra-domain link cost:

[0097] 4.1.1) Calculate the size of each routing area based on four parameters: number of orbits, number of satellites per orbit, number of routing areas in the x-direction, and number of routing areas in the y-direction. :

[0098] ,

[0099] Where P represents the number of orbits, and S represents the number of satellites in each orbit. This represents the number of routing areas in the x-direction. This represents the number of routing areas in the y-direction. For the length of the routing area, The width of the routing area;

[0100] 4.1.2) Based on the size of each routing area, obtain the link cost within that routing area:

[0101] ,

[0102] Where B is the size of the link state information, Case 1 is the case where the number of satellites is the same between different routing areas, and Case 2 is the case where the number of satellites is different between different routing areas;

[0103] 4.2) Establish a quantitative relationship between the number of partitions and the link cost between adjacent domains:

[0104] 4.2.1) Calculate the size of the inter-domain subnet based on four parameters: number of orbits, number of satellites per orbit, number of routing areas in the x-direction, and number of routing areas in the y-direction. :

[0105] ,

[0106] in, The length of the subnet between adjacent domains, The width of the subnet between adjacent domains;

[0107] 4.2.2) Based on the size of the subnet formed between each adjacent routing area, the link cost between adjacent domains is obtained:

[0108] ,

[0109] in, The scale of satellites for the regional boundary is Total number of hops for link-state interactions within the subnet;

[0110] 4.3) Establish a quantitative relationship between the number of partitions and the link cost between non-adjacent domains:

[0111] 4.3.1) Calculate the size of the non-adjacent inter-domain subnet based on four parameters: number of orbits, number of satellites per orbit, number of routing areas in the x-direction, and number of routing areas in the y-direction. :

[0112] ,

[0113] in, The length of the subnet between non-adjacent domains, The width of the subnet between non-adjacent domains;

[0114] 4.3.2) Based on the size of the subnets between non-adjacent domains, obtain the link cost between non-adjacent regions:

[0115] ;

[0116] 4.4) Based on the results of 4.1), 4.2), and 4.3), establish a quantitative relationship between the number of partitions and link cost. :

[0117] .

[0118] Step 5: Establish a quantitative relationship between the number of partitions and the routing convergence time.

[0119] The route convergence time comprises three parts: satellite detection time for local link states, link state update message flooding time, and time required to recalculate the routing table. This step involves establishing a quantitative relationship between the number of partitions and these three factors, which includes:

[0120] 5.1) Based on message transmission interval and decision threshold Determine the time for satellite to detect local link status: ;

[0121] 5.2) Establish a quantitative relationship between the number of partitions and the flooding time of link state update messages:

[0122] Number of partitions and flooding time of link state update messages The quantification relationship includes: the number of partitions and the flooding time of link-state information within the routing area. The quantization relationship and number of partitions are related to the flooding time of link state information between adjacent domains. The quantitative relationship between the two is established through the following processes:

[0123] Based on the number of orbits P, the number of satellites S in each orbit, and the number of routing areas in the x-direction. Number of routing areas in the y direction The length of the routing area Width of the routing area Propagation delay of the same track link Propagation delay of heterogeneous links Establish the number of partitions and the flooding time of link state information within the routing area. Quantification relationship:

[0124] ;

[0125] Based on the number of orbits P, the number of satellites per orbit S, and the number of routing areas in the x-direction. Number of routing areas in the y direction Propagation delay of the same track link Propagation delay of heterogeneous links Length of inter-domain subnets Width of inter-domain subnets Establish the number of partitions and the flooding time of link status information between adjacent domains. Quantification relationship:

[0126] ;

[0127] Establish a quantitative relationship between the number of partitions and the flooding time of link state update messages. :

[0128] ,

[0129] 5.3) Calculate the efficiency constant of the path selection algorithm within the routing area. The computational efficiency constant of the inter-regional path selection algorithm Establish a quantitative relationship between the number of partitions and the time required to recalculate the routing table. :

[0130] ;

[0131] 5.4) Based on the results of steps 5.1), 5.2), and 5.3), establish a quantitative relationship between the number of partitions and the routing convergence time. :

[0132] .

[0133] Step 6: Calculate the optimal route partitioning results for the satellite network:

[0134] Reference Figure 4 This step involves determining the optimal routing partitioning result for the satellite network based on the quantized relationship between the number of partitions, link cost, and routing convergence time. Its implementation includes:

[0135] 6.1) Establishing an adjacency list based on the satellite network topology model ;

[0136] 6.1.1) Based on the satellite node set SAT, for each satellite Maintain a separate adjacency list record It is used to store information about all the neighboring nodes that the node can connect to;

[0137] 6.1.2) For any two satellites Based on the inter-satellite link connectivity determination result in step 1.1.3), determine whether a valid inter-satellite link exists between the two satellite nodes:

[0138] like and If inter-satellite links exist, then... join in adjacency list In the middle, join in adjacency list In the middle, construct a complete adjacency list ;

[0139] ,

[0140] in, Indicates satellite All neighboring satellite nodes;

[0141] like and If there is no inter-satellite link, an adjacency table is not established;

[0142] 6.2) Calculate the propagation delay of the inter-satellite link. :

[0143] According to the adjacency list in step 6.1) The position coordinates of satellite i in the geocentric coordinate system The position coordinates of satellite j in the geocentric coordinate system Calculate the propagation delay of the inter-satellite link between satellite i and satellite j. :

[0144] ,

[0145] Where c is the speed of electromagnetic wave propagation in a vacuum. ;

[0146] 6.3) Calculate the link cost for different numbers of partitions:

[0147] Based on the quantitative relationship between the number of partitions and link cost established in step 4.3), calculate the intra-domain link cost, inter-adjacent domain link cost, and inter-non-adjacent domain link cost for different numbers of partitions, and sum these three to obtain the link cost for different numbers of partitions.

[0148] 6.4) Calculate the routing convergence time for different numbers of partitions:

[0149] Based on the quantitative relationship between the number of partitions and the route convergence time established in step 5.2), calculate the local link state time, the satellite detection time for local link state, and the time required to recalculate the routing table for different numbers of partitions. Summate these three to obtain the route convergence time for different numbers of partitions.

[0150] 6.5) Normalize the link cost and route convergence time for different numbers of partitions:

[0151] Based on the link cost calculated in step 6.3) for different numbers of partitions, the maximum link cost is obtained. and minimum value The extreme value normalization method is used to normalize the link cost values ​​under different numbers of partitions. Mapping to the interval [0,1]:

[0152] ;

[0153] Based on the route convergence time calculated in step 6.4) for different numbers of partitions, the maximum value of the route convergence time is obtained. and minimum value The extreme value normalization method is used to normalize the numerical value of routing convergence time under different numbers of partitions. Mapping to the interval [0,1]:

[0154] ;

[0155] 6.6) Based on weighting coefficients Calculate the overall routing performance index under different numbers of partitions. :

[0156] ;

[0157] 6.7) Calculate the overall routing performance index for all partition numbers based on step 6.6), and take the partition number with the lowest overall routing performance index as the optimal routing partition result. ,in, This represents the number of routing areas in the x-direction of the optimal routing partitioning results. This represents the number of regions in the y-direction of the optimal routing partitioning result.

[0158] It should be noted that the step numbers in this example and the claims are only for the purpose of clearly describing the embodiments of the present invention and facilitating understanding, and their order is not limited.

[0159] Example 2: Routing partitioning device for large-scale low-Earth orbit satellite networks.

[0160] Reference Figure 5 This example includes: Topology modeling module 1, Region partitioning module 2, Routing model construction module 3, Link cost quantization module 4, Routing convergence time quantization module 5, and Optimal partition number decision module 6. Link cost quantization module 4 includes: Intra-domain cost quantization submodule 41, Inter-adjacent domain cost quantization submodule 42, and Non-adjacent domain cost quantization submodule 43; Routing convergence time quantization module 5 includes: Interaction delay quantization submodule 51 and Computation time quantization submodule 52.

[0161] The working principle of the entire system is as follows:

[0162] The topology modeling module 1 is used to provide basic data support for the system. Based on the constellation size, number of orbital planes, orbital altitude, and orbital inclination, it determines the spatial arrangement of satellite nodes and the relationship between orbital geometric constraints, calculates the relative positions, visibility relationships, and inter-satellite link connectivity between satellites, generates a complete constellation spatial network topology, obtains a satellite network topology model, and transmits the relative positions and inter-satellite link connectivity data between satellites to the region division module 2, and transmits the network topology structure to the routing model construction module 3, the link cost quantization module 4, and the routing convergence time quantization module 5, respectively.

[0163] The region division module 2 is used to divide the constellation into multiple routing regions Z, including the number of routing regions in the x-direction and the number of routing regions in the y-direction, based on the inter-satellite relative position and inter-satellite link connectivity data provided by the topology modeling module 1 and in combination with the partitioning constraints. It also calls the inter-satellite relative position and inter-satellite link connectivity data transmitted by the topology modeling module 1 to transmit the routing region division results to the routing model construction module 3, the link cost quantization module 4, and the routing convergence time quantization module 5.

[0164] The routing model construction module 3 is used to establish a satellite network routing model that includes three parts: link state information interaction rules, routing table storage and update mechanism, and data packet forwarding mechanism, based on the network topology structure transmitted by the topology modeling module 1 and the routing area division results of the area division module 2, combined with the dynamic characteristics of the satellite network. It also calls the network topology structure of the topology modeling module 1 and the routing area division results of the area division module 2 to transmit the link state information interaction rules to the link cost quantization module 4 and the routing table storage and update mechanism to the routing convergence time quantization module 5, respectively.

[0165] The link cost quantization module 4 is used to establish a quantification relationship between the number of partitions and the link cost. Specifically: the intra-domain cost quantization submodule 41 is used to calculate the cost generated by link state information interaction within a routing area based on the network topology of the topology modeling module 1, the routing area division results of the area division module 2, and the link state information interaction rules of the routing model construction module 3, thus establishing a quantification relationship between the number of partitions and the link cost within the routing area; the inter-domain cost quantization submodule 42 is used to calculate the cost generated by link state information interaction between adjacent domains based on the network topology of the topology modeling module 1, the routing area division results of the area division module 2, and the link state information interaction rules of the routing model construction module 3. The first module establishes a quantitative relationship between the number of partitions and the link cost between adjacent routing areas. The second module, 43, is used to calculate the cost generated by the link state information interaction between non-adjacent domains based on the network topology of the topology modeling module 1, the routing area division results of the area division module 2, and the link state information interaction rules of the routing model construction module 3, and to establish a quantitative relationship between the number of partitions and the link cost between non-adjacent routing areas. All three sub-modules call the network topology of the topology modeling module 1, the routing area division results of the area division module 2, and the link state information interaction rules of the routing model construction module 3 to transmit the quantitative relationship between the number of partitions and the link cost to the optimal number of partitions decision module 6.

[0166] The routing convergence time quantization module 5 is used to establish a quantization relationship between the number of partitions and the routing convergence time. Specifically: the interaction delay quantization submodule 51 is used to calculate the time required for link state information interaction within the routing area and between adjacent domains based on the network topology of the topology modeling module 1, the routing area division results of the area division module 2, and the routing table storage update mechanism of the routing model construction module 3, thus establishing a quantization relationship between the number of partitions and the link state information interaction delay; the calculation time quantization submodule 52 is used to determine the time required to recalculate the routing table within the area and the routing table between areas based on the network topology of the topology modeling module 1, the routing area division results of the area division module 2, and the routing table storage update mechanism of the routing model construction module 3, thus establishing a quantization relationship between the number of partitions and the routing table recalculation time. Both of these submodules call the network topology of the topology modeling module 1, the routing area division results of the area division module 2, and the routing table storage update mechanism of the routing model construction module 3 to transmit the quantization relationship between the number of partitions and the routing convergence time to the optimal partition number decision module 6.

[0167] The optimal partition number decision module 6 is used to construct a routing performance index by introducing weighted coefficients based on the quantization relationship between the number of partitions and link cost transmitted by the link cost quantization module 4 and the quantization relationship between the number of partitions and routing convergence time transmitted by the routing convergence time quantization module 5. The optimal partition number with the smallest comprehensive routing performance index is output as the optimal routing partition result, providing a reference for the deployment of routing partitions in satellite networks.

[0168] It should be noted that the above functional modules can be implemented, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, they can be implemented, in whole or in part, as program instruction products. A program instruction product includes one or a set of program instructions. When the program instructions are loaded and executed on a computer, the described process or function is generated, in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The program instructions can be stored in a computer-readable and writable storage medium, or transferred from one computer's readable and writable storage medium to another.

[0169] In this embodiment, the direct coupling or communication connection between the modules can be achieved through indirect coupling or communication connection via interfaces, devices, or modules. The functional modules and sub-modules in this embodiment can dynamically reside within a single processing unit, or each module can exist physically independently, or two or more modules can dynamically reside within a single processing unit. When the aforementioned dynamic components are implemented as software functional modules and sold or used as independent products, they can also be stored in a computer-readable and writable storage medium. This storage medium can be a memory, disk, or optical disk, etc. The flowchart representations or method representations of the above embodiments can be understood as representing modules, segments, or portions of code comprising one or more executable instructions configured to implement specific logical functions or processes. This invention is not limited to the disclosed preferred embodiments; its implementation may not follow the order shown or discussed, including performing functions substantially simultaneously according to the functions involved.

[0170] The effects of this invention can be further illustrated by the following simulation experiments:

[0171] I. Experimental Scenario and Parameter Settings

[0172] A large-scale multi-layer satellite network was used as the experimental scenario, consisting of one medium Earth orbit (MEO) constellation and seven low Earth orbit (LEO) constellations, with a total of 10,334 satellite nodes, which conforms to a real multi-layer satellite communication environment.

[0173] Assume the constellation parameters of a large-scale, multi-layered satellite network are shown in Table 1, and the routing performance simulation parameters are shown in Table 2.

[0174] Table 1. Constellation Parameters for Large-Scale Multi-Layer Satellite Networks

[0175]

[0176] Table 2 Routing performance simulation parameters

[0177]

[0178] II. Experimental Content and Results

[0179] Simulation 1, under the experimental scenario setup and parameter settings described above, simulates the throughput, end-to-end latency, and routing performance indicators of the present invention under different numbers of partitions at the same network size. The results are as follows: Figure 6 As shown. Wherein:

[0180] Figure 6 (a) shows the throughput performance simulation results under different numbers of partitions for the same network size;

[0181] Figure 6 (b) shows the simulation results of end-to-end latency performance under different numbers of partitions for the same network size;

[0182] Figure 6 (c) shows the simulation results of routing performance metrics under different numbers of partitions for the same network size.

[0183] from Figure 6 (a) It can be seen that throughput performance decreases when the number of partitions is too small or too large; throughput performance reaches its optimal level when the number of partitions is moderate.

[0184] from Figure 6 (b) It can be seen that the end-to-end delay increases with the increase of the number of partitions;

[0185] from Figure 6 (c) It can be seen that when the number of partitions is too small or too large, the routing performance index will be too high; when the number of partitions is moderate, the routing performance index will be the lowest.

[0186] Simulation 2, under the above experimental scenario setup and parameter settings, simulates the optimal number of regions for this invention under different network sizes and constellation configurations. The results are as follows: Figure 7 As shown. Wherein:

[0187] Figure 7 (a) shows the simulation results of the optimal number of regions under different network sizes;

[0188] Figure 7 (b) shows the simulation results of the optimal number of regions under different constellation configurations;

[0189] from Figure 7(a) It can be seen that the number of optimal partitions varies significantly under different network sizes, and the number of optimal partitions gradually increases as the network size increases.

[0190] from Figure 7 (b) It can be seen that the difference in the number of optimal partitions is small under different constellation configurations;

[0191] Simulation 3, under the experimental setup and parameter settings described above, simulates the link cost, routing convergence time, routing performance metrics, and throughput performance of different algorithms under different network sizes and service loads. The results are as follows: Figure 8 As shown. Wherein:

[0192] Figure 8 (a) shows the simulation results of link overhead performance of different algorithms under different network scales;

[0193] Figure 8 (b) shows the simulation results of routing convergence time performance of different algorithms under different network scales;

[0194] Figure 8 (c) shows the simulation results of routing performance metrics for different algorithms under different network scales;

[0195] Figure 8 (d) shows the simulation results of throughput performance indicators for different algorithms under different service loads;

[0196] from Figure 8 (a) It can be seen that the present invention (OPT-ZRSD with ZARUA in the figure) has the best performance in terms of link overhead, and its performance advantage becomes more and more significant as the network scale increases. When the network scale reaches 10,000 satellites, the link overhead of the present invention is reduced by 11.5%, 22.9%, 5.1% and 6.5% compared with FSPF-MDR, NDS-CDPA, IDLBA and GUSA, respectively.

[0197] from Figure 8 (b) It can be seen that the present invention has the best performance in terms of routing convergence time, and its performance advantage becomes more and more significant as the network scale increases. When the network scale reaches 10,000 satellites, the routing convergence time of the present invention is shortened by 8.0%, 10.9%, 10.3% and 0.3% compared with FSPF-MDR, NDS-CDPA, IDLBA and GUSA, respectively.

[0198] from Figure 8 (c) It can be seen that the present invention achieves the lowest routing performance index value compared with other algorithms under all network scales;

[0199] from Figure 8(d) It can be seen that the present invention has the best throughput performance. When the service rate reaches 80% of the network capacity, the throughput of the present invention is improved by 6.8%, 5.1%, 10.7% and 3.5% respectively compared with FSPF-MDR, NDS-CDPA, IDLBA and GUSA.

[0200] The simulation results show that the present invention can determine the optimal number of partitions under a given network size, thereby achieving the best balance between throughput, end-to-end latency and routing performance; the optimal number of partitions increases significantly with the expansion of network size and is less affected by constellation configuration; compared with existing algorithms, the present invention has the lowest link overhead, the shortest routing convergence time and the highest throughput under different network sizes and different service loads.

[0201] In summary, the routing partitioning method proposed in this invention has the characteristics of high scalability, strong adaptability, and superior overall performance, and is suitable for large-scale low-Earth orbit satellite networks.

Claims

1. A routing partitioning method for large-scale low-Earth orbit satellite networks, characterized in that, include: (1) Based on the constellation size N, the number of orbital planes P, the orbital altitude H, and the orbital inclination A, the spatial arrangement of satellite nodes and the orbital geometric constraints can be determined, the relative positions between satellites can be calculated, the visual relationships and inter-satellite link connectivity can be determined, and then a complete constellation spatial network topology can be generated to obtain the satellite network topology model. (2) Based on the relative positions of satellites in the satellite network topology model, the constellation is divided into routing areas including the x-direction. Number of routing areas in the y direction Multiple routing areas Z, and satisfying : (3) Based on the relative positions between satellites, the visibility relationship, the inter-satellite link connectivity and multiple routing areas Z in the satellite network topology model, establish a satellite network routing model that includes three parts: link status information interaction rules, routing table storage and update mechanism and data packet forwarding mechanism; (4) Based on the network topology, routing area division results, and link status information exchange rules in the satellite network routing model, establish a quantitative relationship between the number of partitions and link overhead. (5) Based on the dynamic topology change characteristics of the satellite network topology model, the routing area division results, and the routing table storage and update method of the satellite network routing model, establish a quantitative relationship between the number of partitions and the routing convergence time. (6) Based on the quantitative relationship between the number of partitions, link cost, and route convergence time, calculate the link cost and route convergence time under different numbers of partitions. Calculate the route performance index based on the route convergence time and link cost, and minimize it to obtain the optimal route partitioning result for the satellite network.

2. The method according to claim 1, characterized in that, The calculation of relative positions between satellites, determination of visual relationships, and inter-satellite link connectivity in (1) includes: (1a) Based on four parameters, namely constellation size, number of orbital planes, orbital altitude and orbital inclination, the three-dimensional position coordinates of each satellite in the geocentric inertial coordinate system are calculated using the orbital motion equation. The position coordinates of the satellite in the geocentric geofixed coordinate system are obtained through coordinate transformation. The relative distance and relative azimuth between the satellites are calculated based on the position coordinate difference between the two satellites, and the relative position between the satellites is determined. (1b) Based on the satellite orbital altitude, Earth radius and the position coordinates of the two satellites, calculate the shortest distance between the line connecting the satellites and the Earth's surface: if the distance is greater than the safe distance reserved by the Earth radius and atmospheric refraction, it is determined that there is an unobstructed line-of-sight propagation condition between the two satellites and that they are in a visible relationship; otherwise, they are determined to be in a non-visual relationship. (1c) Based on the established line-of-sight relationship between the two satellites, determine whether the link meets the conditions for establishment according to the maximum communication distance threshold of the inter-satellite link and the maximum load limit of a single link: If the link distance does not exceed the communication distance threshold and the link load does not exceed the maximum limit, then the inter-satellite link can be established and maintained. Otherwise, it is determined to be disconnected.

3. The method according to claim 1, characterized in that, In step (2), the constellation is divided into routing areas in the x-direction based on the relative positions of the satellites. Number of routing areas in the y direction Its implementation includes: (2a) Using the orbital plane distribution of the satellite constellation as the basis for x-direction partitioning, satellites in different orbital planes are assigned to the corresponding x-direction routing areas based on the total number of orbital planes and the distribution interval; (2b) Based on the phase distribution of satellites in the same orbit, satellites at different positions in the same orbital plane are assigned to the corresponding y-direction routing area according to their relative positions.

4. The method according to claim 1, characterized in that, The three parts of the satellite network routing model in (3) are implemented as follows: The link state information exchange rule refers to the link state information exchange between satellites within each routing area in a flooding mode. Satellites in adjacent routing areas form a subnet, and link state information is flooded within this subnet. In each routing area, a satellite is selected as a gateway to form a subnet for link state information flooding, so that each satellite can calculate the next-hop routing area required to reach non-adjacent areas, ensuring the routing reachability of all satellites. The routing table storage update mechanism refers to the satellite updating the two types of routing tables stored in each satellite: the intra-regional routing table and the inter-regional routing table, using the Dijkstra algorithm. Specifically, the intra-regional routing table is updated using the link-state information within the routing area, and the inter-regional routing table is updated using the link-state information within the area and between adjacent areas. The packet forwarding mechanism is divided into two categories: intra-regional packet forwarding and inter-regional packet forwarding. Intra-regional packet forwarding involves the source satellite, which is located in the same routing area as the destination satellite, querying the intra-regional routing table to forward the packet to the intra-regional relay satellite, and finally forwarding it to the destination satellite. Inter-regional packet forwarding involves the source satellite, located in a different routing region from the destination satellite, querying the inter-regional routing table and forwarding the packet to the inter-regional relay satellite in the destination region. The inter-regional relay satellite then queries the intra-regional routing table and finally forwards the packet to the destination satellite.

5. The method according to claim 1, characterized in that, The quantitative relationship between the number of partitions and link overhead established in (4) includes: (4a) Number of partitions and intra-domain link overhead: (4a1) Calculate the size of each routing area based on four parameters: number of orbits, number of satellites per orbit, number of routing areas in the x-direction, and number of routing areas in the y-direction. : ， Where P represents the number of orbits, and S represents the number of satellites in each orbit. This represents the number of routing areas in the x-direction. This represents the number of routing areas in the y-direction. For the length of the routing area, The width of the routing area; (4a2) Based on the size of each routing area, establish a quantitative relationship between the number of partitions and the intra-domain link cost. : ， Where B is the size of the link state information, Case 1 is the case where the number of satellites is the same between different routing areas, and Case 2 is the case where the number of satellites is different between different routing areas; (4b) Establishing link overhead between adjacent domains: (4b1) Calculate the size of the inter-domain subnet based on four parameters: number of orbits, number of satellites per orbit, number of routing areas in the x-direction, and number of routing areas in the y-direction. : ， in, The length of the subnet between adjacent domains, The width of the subnet between adjacent domains; (4b2) Based on the size of the subnet formed between each adjacent routing area, establish a quantitative relationship between the number of partitions and the link cost between adjacent domains. : ， in, The scale of satellites for the regional boundary is Total number of hops for link-state interactions within the subnet; (4c) Establishing the cost of links between non-adjacent domains: (4c1) Calculate the size of the non-adjacent inter-domain subnet based on four parameters: number of orbits, number of satellites per orbit, number of routing areas in the x-direction, and number of routing areas in the y-direction. : ， in, The length of the subnet between non-adjacent domains, The width of the subnet between non-adjacent domains; (4c2) Based on the size of the subnets between non-adjacent domains, establish a quantitative relationship between the number of partitions and the link overhead between non-adjacent domains. : ， (4d) Based on the above results, the quantitative relationship between the number of partitions and the link cost is obtained. : 。 6. The method according to claim 1, characterized in that, The quantitative relationship between the number of partitions and the routing convergence time established in (5) includes: (5a) Based on message transmission interval and decision threshold Determine the time for satellite to detect local link status : ， (5b) Based on the number of orbits P, the number of satellites S in each orbit, and the number of routing areas in the x-direction. Number of routing areas in the y direction The length of the routing area Width of the routing area Propagation delay of the same track link and the propagation delay of the heterogeneous link Determine the propagation delay required for satellites within the routing area to exchange link state information. Propagation delay required for link state information exchange with regional boundary satellites : ； ； in The length of the subnet between adjacent domains, The width of the subnet between adjacent domains; (5c) Based on the computational efficiency constant of the path selection algorithm within the routing area The computational efficiency constant of the inter-regional path selection algorithm Determine the time required to recalculate the routing table. : ； (5d) Based on the above results, the quantitative relationship between the number of partitions and the routing convergence time is obtained. : 。 7. The method according to any one of claims 1-6, characterized in that: The calculation of link cost and route convergence time for different numbers of partitions in (6) includes: (6a) Based on the satellite network topology model, routing area division method and satellite network routing model, traverse different combinations of the number of routing partitions, and calculate the intra-domain link cost for each number of partitions according to the quantitative relationship between the number of partitions and link cost. Inter-domain link overhead Link overhead between non-adjacent domains To obtain the total link cost ; (6b) Based on the quantization relationship between the number of partitions and the routing convergence time, calculate the local link state detection time for each number of partitions. The propagation delay required for satellites within the routing area to exchange link state information. The propagation delay required for link state information exchange between satellites at the regional boundary and routing table recalculation time The total route convergence time is obtained. .

8. The method according to claim 1, characterized in that, The calculation of routing performance metrics based on routing convergence time and link cost in (6) includes: (6c) Normalize the total link cost and total route convergence time according to the link cost and route convergence time values ​​under different numbers of partitions. (6d) A weighting coefficient is introduced to assign weights to the normalized total link cost and the normalized total route convergence time. The weighted sum of the normalized total link cost and the normalized total route convergence time is used as the routing performance index. By minimizing this routing performance index, the corresponding optimal number of partitions is selected, and the optimal routing partition result of the satellite network is finally output.

9. A routing partitioning device for a large-scale low-Earth orbit satellite network, characterized in that, include: The topology modeling module is used to determine the spatial arrangement of satellite nodes and their orbital geometric constraints based on constellation size, number of orbital planes, orbital altitude, and orbital inclination. It also calculates the relative positions, visibility relationships, and inter-satellite link connectivity between satellites, generates a complete constellation spatial network topology, and obtains a satellite network topology model. The region division module is used to divide the constellation into multiple routing regions Z, including the number of routing regions in the x-direction and the number of routing regions in the y-direction, based on the relative positions between satellites in the satellite network topology model. The routing model construction module is used to establish a satellite network routing model based on the relative positions between satellites, visibility relationships, inter-satellite link connectivity, and multiple routing areas Z in the satellite network topology model. This model includes three parts: link state information exchange rules, routing table storage and update mechanism, and data packet forwarding mechanism. The link cost quantization module is used to establish a quantitative relationship between the number of partitions and the link cost based on the network topology, routing area division results, and link state information interaction rules in the satellite network routing model. The route convergence time quantization module is used to establish a quantitative relationship between the number of partitions and the route convergence time based on the dynamic topology change characteristics of the satellite network topology model, the routing area division results, and the routing table storage update method of the satellite network routing model. The optimal partition number decision module is used to calculate the link cost and routing convergence time under different partition numbers based on the quantitative relationship between the number of partitions, link cost, and routing convergence time, minimize the routing performance index, and obtain the optimal routing partition result for the satellite network.

10. The apparatus according to claim 1, characterized in that: The link cost quantization module includes: The intra-domain cost quantization submodule is used to establish the quantization relationship between the number of partitions and the intra-domain link cost; The adjacent domain overhead quantization submodule is used to establish the quantization relationship between the number of partitions and the link overhead between adjacent domains; The non-adjacent domain overhead quantization submodule is used to establish the quantization relationship between the number of partitions and the link overhead between non-adjacent domains; The routing convergence time quantization module includes: The interaction latency quantization submodule is used to establish a quantization relationship between the number of partitions and the interaction latency of link status information based on the link status information interaction rules. The computation time quantization submodule is used to establish a quantization relationship between the number of partitions and the routing table recalculation time based on the routing table storage update mechanism.

Images