A multiple parallel networking method based on large-scale constellation satellites

Abstract

The present invention discloses a multiple parallel networking method based on large-scale constellation satellites, comprising the following steps: obtaining the IP addresses of each satellite and the laser connection information table; according to the laser connection information table, each satellite node initializes the connection with adjacent satellites and plans the routing according to the traditional path algorithm; setting a threshold value to screen important links; using the strongly connected algorithm to obtain the strongly connected components of the inter-satellite connections, and each strongly connected component is used as a parallel subnet to reduce the graph of the directed graph; re-planning the routing according to the traffic, using span connections between parallel subnets and strong connections within parallel subnets, repeating the above steps to convert the connections between subnets into long-distance connections. The multiple parallel networking method based on large-scale constellation satellites of the present invention, by obtaining the inter-satellite link traffic information, constructing a directed graph with directivity, planning parallel subnets, and adjusting the laser connections between subnets and within subnets, realizes a variable-structure networking and ensures efficient inter-regional transmission.

Description

Technical Field

[0001] The present invention relates to a networking method, and in particular to a multiple parallel networking method based on large-scale constellation satellites. Background Art

[0002] Currently, the research on satellite networking methods mainly still focuses on small-scale networking and satellite-ground communication based on existing microwave communication transmission technologies. The link scheduling scheme is not applicable to low-earth orbit giant constellations in the context of current inter-satellite laser interconnection. Most of its research targets Iridium satellites or simple inclined orbit constellations. Most of the schemes adopt adjacent satellite communication schemes, focusing on the shortest path and multi-hop transmission, and there is little research on networking for ultra-large-scale constellations. Nowadays, the number of satellites has increased sharply, the networking scale has expanded, and laser communication schemes are expected to be adopted between satellites. In the context of tens of thousands of satellites, after dense satellite networking, the number of routing hops increases and the delay increases, and the advantages of laser communication are exhausted. Currently, traditional satellite networking schemes cannot fully utilize the ultra-low delay advantages brought by networking tens of thousands of satellites and inter-satellite laser communication. Summary of the Invention

[0003] In view of the above-mentioned defects of the prior art, the technical problem to be solved by the present invention is that the existing satellite networking methods are not suitable for low-earth orbit giant constellations in the context of inter-satellite laser interconnection, nor are they suitable for the networking research of ultra-large-scale constellations. In the context of tens of thousands of satellites, after dense satellite networking, the number of routing hops increases and the delay increases. Therefore, the present invention provides a multiple parallel networking method based on large-scale constellation satellites. By obtaining inter-satellite link traffic information, constructing a directed graph through traffic directivity, planning parallel subnets through a strongly connected algorithm, and adjusting laser connections between and within subnets, a variable-structure networking is realized to ensure efficient inter-region transmission.

[0004] To achieve the above object, the present invention provides a multiple parallel networking method based on large-scale constellation satellites, including the following steps:

[0005] Obtain the IP addresses of each satellite and the laser connection information table;

[0006] According to the laser connection information table, each satellite node initializes the connection with adjacent satellites and plans the routing according to the traditional path algorithm;

[0007] Calculate the traffic weight of the inter-satellite link according to the planned routing, set a threshold, screen important links and create a traffic directed graph based on the important links;

[0008] According to the traffic directed graph, obtain the strongly connected components of the inter-satellite connection by using the strongly connected algorithm;

[0009] According to the traffic directed graph and the strongly connected components, each strongly connected component is used as a parallel subnet to reduce the graph of the directed graph;

[0010] Re - plan the routing according to the traffic. Use span connections between parallel sub - networks and strong connections within parallel sub - networks. Repeat the above steps to convert the connections between sub - networks into long - distance connections to ensure specific services.

[0011] Furthermore, each satellite has a laser connection information table. The laser connection information table records the currently connected satellites of each satellite, the satellite IPs within the laser reach at the current moment, and their location logical addresses. Among them, the satellite IPs within the laser reach are the satellite IPs to which each satellite can adjust the laser for connection, and the location logical addresses are the current location information of all satellites that each satellite can connect to. The laser connection information table is continuously updated as the satellite position changes.

[0012] Furthermore, use the satellite logical addresses in the laser connection information table to find the satellites before and after the same orbit and the satellites on the adjacent orbits on the left and right of each satellite. Each satellite node is connected to the above 4 adjacent satellites to complete the initialization, thus completing the connection initialization of each satellite, and the initial overhead of each link is equal.

[0013] Furthermore, based on the connection status of each link after initialization, use traditional algorithms to plan the routing for current various communication requests, obtain the traffic conditions of each inter - satellite link, and calculate the overhead of each link.

[0014] Furthermore, screen important links and create a traffic directed graph based on important links. Specifically, set a link overhead threshold, compare the overhead of each link with the threshold. If it is greater than the threshold, it is judged as an important link; if it is less than the threshold, it is judged as a small - traffic service. Then create a traffic directed graph, add all important links to the directed graph, and small - traffic services do not need to be included in the directed graph.

[0015] Furthermore, for the traffic directed graph created by important links, use the strongly connected algorithm to calculate the strongly connected components of the directed graph. Any pair of satellite nodes in the entire satellite network existing in the strongly connected component can be connected.

[0016] Furthermore, plan each obtained strongly connected component as a parallel sub - network and store the information of each node within the sub - network. Condense each sub - network into a node of the directed graph, and a traffic directed graph between sub - networks can be obtained. The connections in this graph are the connections between sub - networks.

[0017] Furthermore, perform routing re - planning. For the services within each sub - network, use the original routing method to calculate the overhead of each link within the sub - network. For cross - sub - network services, use long - distance links, and the satellite nodes in the two sub - networks are connected across long distances.

[0018] Further, re-plan the route according to the traffic. The parallel subnets are connected by spans, and strong connections are used within the parallel subnets. Repeat the above steps to convert the inter-subnet connections into long-distance connections. Specifically, for cross-subnet services, set a link cost threshold, and filter out the satellite node sets A and B with small link costs in the source and destination subnets. For the filtered source subnet satellite node set A, query the laser connection information table, filter out the satellite node set A' that can be connected to B, and obtain B' connected to the nodes within A'. If there are multiple optional nodes in A', use the shortest path algorithm to obtain the final result, adjust the lasers, and perform long-distance connections between subnets. If there are no optional nodes in A', select the subnet that is adjacent to the target subnet and has the smallest total link cost as the target subnet.

[0019] Further, repeat the node screening step until connected to the destination satellite node.

[0020] Technical effects

[0021] Most current routing schemes are based on adjacent satellite communications, focusing on the cases of shortest paths and multi-hop transmissions. Now, with the networking of ultra-large-scale constellations, the scale has expanded, and a laser communication scheme is expected to be used between satellites. After the dense satellite networking, the number of routing hops increases and the latency increases, and the advantages of laser communication are exhausted. The present invention can make up for the ultra-low latency advantage that cannot be maintained by traditional satellite networking schemes.

[0022] A multiple parallel networking method based on large-scale constellation satellites of the present invention plans an initial route through a traditional path algorithm, and then obtains the inter-satellite link traffic weights. Then, using a strongly connected algorithm, it plans the large-scale constellation satellites to obtain multiple parallel subnets. Subsequently, it converts the inter-subnet connections into long-distance connections, optimizes the transmission hops and link connections in the network communication, realizes the adjustment of laser connections between subnets and within subnets, realizes a variable structure networking, maintains the ultra-low latency advantage, and ensures efficient inter-regional transmission.

[0023] The following will further illustrate the concept, specific structure and technical effects generated by the present invention in conjunction with the drawings, so as to fully understand the purpose, features and effects of the present invention. Brief description of the drawings

[0024] Figure 1 is a schematic diagram of a multiple parallel networking method based on large-scale constellation satellites according to a preferred embodiment of the present invention;

[0025] Figure 2 is a schematic diagram of the change in the satellite inter-satellite link connection process of a multiple parallel networking method based on large-scale constellation satellites according to a preferred embodiment of the present invention. Detailed implementation manners

[0026] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0027] As Figure 1 shown, the embodiments of the present invention provide a multiple parallel networking method based on large-scale constellation satellites, including the following steps:

[0028] Step 100: Obtain the IP addresses of each satellite and the laser connection information table; specifically, each satellite has a laser connection information table, which records the currently connected satellites of each satellite, the satellite IPs within the laser reach range at the current moment and their position logical addresses; among them, the satellite IPs within the laser reach range are the satellite IPs to which each satellite can adjust the laser for connection, and the position logical address is the current position information of all satellites that each satellite can connect to; the laser connection information table is continuously updated following the change of the satellite position.

[0029] Step 200: According to the laser connection information table, each satellite node initializes the connection with adjacent satellites and plans the routing according to the traditional path algorithm; specifically, use the satellite position logical address to find the satellites before and after the same orbit and the satellites on the left and right of the adjacent orbit of the satellite, and this satellite node connects with these four adjacent satellites to complete the initialization. After completing this operation for each satellite, the satellite connection initialization is completed, and the initial overhead of each link is equal. Based on the current connection situation, use the Dijkstra shortest path algorithm or other traditional algorithms to plan the routing for the current various communication requests, obtain the traffic situation of each inter-satellite link, and calculate the overhead of each link.

[0030] Step 300: Set a threshold according to the weight of the inter-satellite link traffic to screen important links; specifically, set a link overhead threshold, compare the overhead of each link with the threshold, if it is greater than the threshold, it is judged as an important link, and if it is less than the threshold, it is judged as a small traffic service; then create a traffic directed graph, add all important links to the directed graph, and small traffic services do not need to be included in the directed graph.

[0031] Step 400: According to the traffic directed graph, use the strongly connected algorithm to obtain the strongly connected components of the inter-satellite connection; specifically, according to the selected traffic directed graph, use the strongly connected algorithm to calculate the strongly connected components of the directed graph, and any pair of satellite nodes in the entire satellite network existing in the strongly connected components can be connected.

[0032] Step 500: According to the traffic directed graph and strongly connected components, each strongly connected component is regarded as a parallel subnet, and the directed graph is reduced. Specifically, each obtained strongly connected component is planned as a parallel subnet, and the information of each node in the subnet is stored. Each subnet is reduced to a node of the directed graph, and the traffic directed graph between subnets can be obtained. The connections in this graph are the connections between subnets.

[0033] Step 600: According to the traffic, re-plan the route. Use span connections between parallel subnets and strong connections within parallel subnets. Repeat the above steps to convert the connections between subnets into long-distance connections to ensure specific services. Specifically, perform route re-planning. For the services within each subnet, use the original routing method to calculate the link costs within each subnet. For cross-subnet services, use long-distance links, and the satellite nodes in the two subnets are connected across long distances. Set a link cost threshold to screen the sets A and B of satellite nodes with small link costs in the source and destination subnets. For the set A of satellite nodes in the source subnet, query the laser connection information table to screen the set A' of satellite nodes that can connect to B, and obtain B' connected to the nodes in A'. If there are multiple optional nodes in A', use the shortest path algorithm to obtain the final result and adjust the lasers for long-distance connections between subnets. If there are no optional nodes in A', select the subnet that is adjacent to the target subnet and has the smallest total link cost as the target subnet.

[0034] Repeat the node screening step until connected to the destination satellite node.

[0035] The following will use a specific example to illustrate a multiple parallel networking method based on large-scale constellation satellites proposed by the present invention.

[0036] Suppose there is a communication request among 12 satellites currently, and the 14 satellites are numbered A - L. According to Step 100, each satellite has a laser connection information table, and a partial example is shown in Table 1. For satellite A, assume that the laser of satellite A can connect to satellites B - K and cannot connect to satellites L - N. Then the current satellite A is connected to satellite B, and the satellite IP and location logical addresses within the laser reach range at the current moment record satellites B - K.

[0037] Table 1 Laser connection information table

[0038]

[0039] This connection information table will be continuously updated as the satellite moves.

[0040] According to Step 200, based on the information table, each satellite completes the initialization connection with the satellites before and after it in the same orbit and the satellites on the left and right in the adjacent orbits. The connections are as Figure 2As shown in (a). Based on the current connection situation, the Dijkstra shortest path algorithm or other traditional algorithms are used to plan routes for current various communication requests and obtain the traffic conditions of each link. For example, when using the shortest path algorithm, input the current satellite connection graph (bidirectional). The algorithm first makes a vertical (same orbit) determination and then a horizontal (adjacent orbit) determination. For instance, if there is one communication task from satellite A to satellite F, input A / F, and the output path is ABEGH. Then, record the task ratio between satellites as 1. As a result, the traffic of links A→B, B→E, E→G, and G→H each increases by 1. After all communication requests are input and output, the traffic conditions of each inter-satellite link of the 14 satellites after obtaining the planned routes are as follows Figure 2 As shown in (b).

[0041] According to step 300, the average value of the link overhead threshold is calculated to be 2. Important links with a value greater than this threshold are retained and included in the directed graph. For example, in this embodiment, if the traffic of link A→B is 3, then this vector is retained; if the traffic of link A←B is 1, then this vector is removed. The obtained directed graph of inter-satellite traffic is as follows Figure 2 As shown in (c).

[0042] According to steps 400 and 500, the strongly connected components of the directed graph can be calculated using the strongly connected algorithm. Input the total number of satellite nodes as 14, the number of links as 19, and the nodes connected by each link, such as 1(A)→2(B), 2(B)→4(D), 2(B)→3(C), 2(B)→5(E), etc. Define DFN[U] as the order number (timestamp) of the depth-first search of node U, and LOW[U] as the smallest sequence number that can be found with U or its subtrees. The algorithm starts a depth-first search in the order of node input, and the derivation conditions are as follows: ① For each newly found node X, DFN[X] = LOW[X]; ② When there is a connection between node Y and node Z, if Z is not in the stack at the time of DFN[Y], LOW[Y] = min(LOW[Y], LOW[Z]); ③ When node Y and node Z are not in the stack, LOW[Y] = min(LOW[Y], DFN[Z]). When DFN[X] = LOW[X], it is determined that all nodes in the search subtree rooted at X are a strongly connected component. Output strongly connected component 1: A; strongly connected component 2: D;...; strongly connected component 7: IJKNML. Each strongly connected component is planned as a parallel subnet. For example, satellites B and E form a parallel subnet, and satellites I - N form a parallel subnet. The reduced graph (i.e., regarding all nodes within the same network as a new node, and there are 7 nodes in total from ① to ⑦ currently) is as follows Figure 2 As shown in (d).

[0043] Step 600: Perform route re-planning. For services within each subnet, the original routing method is adopted. For example, if there is one communication task from satellite I to satellite K within subnet ⑦, the path is still IJK. For cross-subnet services, long-distance links are adopted (i.e., reconstruct the links from node ① to node ⑦ after reducing the graph). For the communication between satellite B in subnet ③ and satellite M in subnet ⑦, the original path should be BEGJM. Assume that the satellite nodes with low overhead in subnets ② and ⑦ are satellite B, E and satellite I. Query the laser connection information table for satellite B and E. It can be seen that both satellite B and E can adjust the laser and be directly connected to satellite I. Then, using the shortest path algorithm, if satellite B is selected for the cross-span connection in subnet ③, the path is BIJM; if satellite E is selected, the path is BEIJM, with one less hop in the connection. Output the selection of satellite B for cross-distance connection to satellite I within subnet ⑦, and then route from satellite I to satellite M through satellite I, thus realizing inter-satellite long-distance connection. The final path BIJM is shorter than BEGJM, reducing the multi-hop delay of data. This delay will increase with the increase in the number of network nodes and connection complexity. Therefore, this method can ensure efficient regional transmission.

[0044] The preferred specific embodiments of the present invention have been described in detail above. It should be understood that those of ordinary skill in the art can make many modifications and variations based on the concept of the present invention without creative labor. Therefore, all technical solutions that can be obtained by those skilled in the art in the technical field of the present invention through logical analysis, reasoning, or limited experiments based on the concept of the present invention on the basis of the prior art shall fall within the protection scope determined by the claims.

Claims

1. A multiple parallel networking method based on large-scale constellation satellites, characterized in that, it includes the following steps: Step 100, obtain the IP addresses of each satellite node and the laser connection information table; Step 200, according to the laser connection information table, each satellite node initializes the connection with adjacent satellite nodes and plans the route according to the traditional path algorithm; Step 300, calculate the inter-satellite link traffic weight according to the planned route, set a threshold, screen important links and create a traffic directed graph based on the important links; Step 400, according to the traffic directed graph, use the strongly connected algorithm to obtain the strongly connected components of the inter-satellite connection; Step 500, according to the traffic directed graph and the strongly connected components, each strongly connected component is used as a parallel subnet to reduce the graph of the directed graph; Step 600, re-plan the route according to the traffic, use span connection between the parallel subnets, and use strong connection within the parallel subnets. Repeat the above steps to convert the connection between subnets into long-distance connection to ensure specific services.

2. A multiple parallel networking method based on large-scale constellation satellites according to claim 1, characterized in that, each satellite node has a laser connection information table, and the laser connection information table records the currently connected satellite nodes of each satellite node, the IP addresses of the satellite nodes within the laser reachable range at the current moment and their position logical addresses; wherein, the IP addresses of the satellite nodes within the laser reachable range are the IP addresses of the satellite nodes that each satellite node can adjust the laser to connect to, and the position logical address is the current position information of all the satellite nodes that each satellite node can connect to; the laser connection information table is continuously updated following the change of the satellite node position.

3. A multiple parallel networking method based on large-scale constellation satellites according to claim 2, characterized in that, use the satellite node logical addresses in the laser connection information table to find the satellite nodes before and after the same orbit and the satellite nodes on the adjacent orbits on the left and right of each satellite node, and each satellite node is connected to the above 4 adjacent satellite nodes to complete the initialization, so as to complete the connection initialization of each satellite node, and the initial cost of each link is equal.

4. A multiple parallel networking method based on large-scale constellation satellites according to claim 3, characterized in that, Based on the connection conditions of each link after initialization, use the traditional algorithm to plan the route for current various communication requests, obtain the traffic conditions of each inter-satellite link, and calculate the cost of each link.

5. A multiple parallel networking method based on large-scale constellation satellites according to claim 4, characterized in that, Screen important links and create a traffic directed graph based on the important links, specifically including, setting a link cost threshold, comparing the cost of each link with the threshold, if it is greater than the threshold, it is judged as an important link, and if it is less than the threshold, it is judged as a small traffic service; then create a traffic directed graph, add all the important links to the directed graph, and the small traffic services do not need to be included in the directed graph.

6. A multiple parallel networking method based on large-scale constellation satellites according to claim 5, characterized in that, Re - plan the route according to the traffic. A span connection is adopted between the parallel sub - networks, and a strong connection is adopted within the parallel sub - networks. Repeat step 100 - step 500 to convert the inter - subnet connection into a long - distance connection. Specifically, for cross - subnet services, set a link cost threshold, and screen out the satellite node sets A and B with small link costs in the source subnet and the target subnet; for the screened source - subnet satellite node set A, query the laser connection information table, screen out the satellite node set A' that can be connected to B, and obtain B' connected to the nodes in A'; if there are multiple optional nodes in A', obtain the final result using the shortest - path algorithm, adjust the laser, and perform long - distance connection between sub - networks; If there are no optional nodes in A', select the subnet that is adjacent to the target subnet and has the smallest total link cost as the target subnet.

7. A multiple - parallel networking method based on large - scale constellation satellites as described in claim 6, characterized in that, Repeat the node screening step until connected to the destination satellite node.

Images