Satellite communication scheduling method and device based on double links, equipment and medium

Abstract

The invention discloses a satellite communication scheduling method and device based on double links, equipment and a medium. The method comprises the following steps: constructing a first multi-dimensional matrix according to position information of a plurality of satellites; performing feature extraction on the first multi-dimensional matrix to determine a second multi-dimensional matrix; first graph data is constructed according to the second multi-dimensional matrix, nodes of the first graph data correspond to satellites respectively, node attribute information is used for indicating the data transmission quantity and the data receiving quantity of cells corresponding to the satellites, and edges of the first graph data are used for indicating that the two satellites can communicate with each other; the edge attributes are used for indicating the data transmission quantity between the two satellites and enforceable communication strategies; performing message passing on the first graph data by using a pre-trained graph neural network so as to determine corresponding second graph data; and uniquely determining a communication strategy of one-way communication among the satellites according to the edge feature information of one-way edges in the second graph data.

Description

Technical Field

[0001] The present application relates to the technical field of satellite communication, and in particular, to a satellite communication scheduling method, apparatus, device, and medium based on a dual-link. Background Art

[0002] At present, low-earth orbit satellites can be equipped with microwave phased array antennas and laser communication terminals at the same time, so that communication via microwave links and laser links can be achieved simultaneously. Communicating via a laser link has the advantages of ultra-high bandwidth, high transmission rate, and low latency, and can achieve communication with an ultra-long single-hop distance (theoretically, the single-hop distance can reach 5,400 kilometers). However, laser is easily affected by haze and turbulence in the atmosphere, and the complexity of the tracking technology is high when the satellite is moving at high speed. In contrast, communicating via a microwave link has the advantages of high reliability and strong ability to resist bad weather, but its bandwidth and transmission rate are limited, and the latency is relatively high.

[0003] Therefore, in actual low-earth orbit satellite communication services, how to schedule microwave links and laser links so as to take into account the transmission advantages of both microwave links and laser links is a technical problem that urgently needs to be solved in low-earth orbit satellite communication systems. Summary of the Invention

[0004] Embodiments of the present disclosure provide a satellite communication scheduling method, apparatus, device, and medium based on a dual-link to at least solve the technical problems existing in the prior art.

[0005] According to one aspect of the embodiments of the present disclosure, a satellite communication scheduling method based on a dual-link is provided, including: constructing a first multi-dimensional matrix according to the position information of multiple satellites, where the first multi-dimensional matrix is used to indicate the positions between the satellites and the position environment information related to signal communication; performing feature extraction on the first multi-dimensional matrix to determine a second multi-dimensional matrix, where the second multi-dimensional matrix is used to indicate the communication strategies that can be implemented between the satellites, and the communication strategies include: communicating via a microwave link and communicating via a laser link; constructing a first graph data according to the second multi-dimensional matrix, where the nodes of the first graph data respectively correspond to the satellites, the node attribute information is used to indicate the data transmission volume and data reception volume of the cells corresponding to the satellites, and the edges of the first graph data are used to indicate that communication can be performed between two satellites, and the edge attribute is used to indicate the data transmission volume between two satellites and the communication strategies that can be implemented; using a pre-trained graph neural network to perform message passing on the first graph data, so as to determine the corresponding second graph data; and uniquely determining the communication strategies for one-way communication between the satellites according to the edge feature information of the one-way edges in the second graph data.

[0006] According to another aspect of the embodiments of the present disclosure, there is also provided a storage medium, which includes a stored program. When the program runs, the above-described method is executed by a processor.

[0007] According to another aspect of the embodiments of the present disclosure, there is also provided a satellite communication scheduling device based on a dual-link, including: a first matrix construction module, configured to construct a first multi-dimensional matrix according to the position information of multiple satellites, where the first multi-dimensional matrix is used to indicate the positions between the satellites and the position environment information related to signal communication; a second matrix determination module, configured to perform feature extraction on the first multi-dimensional matrix to determine a second multi-dimensional matrix, where the second multi-dimensional matrix is used to indicate the communication strategies that can be implemented between the satellites, and the communication strategies include: communicating through a microwave link and communicating through a laser link; a first graph data construction module, configured to construct a first graph data according to the second multi-dimensional matrix, where the nodes of the first graph data respectively correspond to the satellites, the node attribute information is used to indicate the data transmission amount and data reception amount of the cells corresponding to the satellites, and the edges of the first graph data are used to indicate that communication can be performed between two satellites, and the edge attribute is used to indicate the data transmission amount between two satellites and the communication strategies that can be implemented; a message passing module, configured to perform message passing on the first graph data by using a pre-trained graph neural network to determine the corresponding second graph data; and a communication strategy scheduling module, configured to uniquely determine the communication strategies for one-way communication between the satellites according to the edge feature information of the one-way edges in the second graph data.

[0008] According to another aspect of the embodiments of the present disclosure, there is also provided a satellite communication scheduling device based on a dual-link, including: a processor; and a memory, connected to the processor, configured to provide instructions for the processor to perform the following processing steps: constructing a first multi-dimensional matrix according to the position information of multiple satellites, where the first multi-dimensional matrix is used to indicate the positions between the satellites and the position environment information related to signal communication; performing feature extraction on the first multi-dimensional matrix to determine a second multi-dimensional matrix, where the second multi-dimensional matrix is used to indicate the communication strategies that can be implemented between the satellites, and the communication strategies include: communicating through a microwave link and communicating through a laser link; constructing a first graph data according to the second multi-dimensional matrix, where the nodes of the first graph data respectively correspond to the satellites, the node attribute information is used to indicate the data transmission amount and data reception amount of the cells corresponding to the satellites, and the edges of the first graph data are used to indicate that communication can be performed between two satellites, and the edge attribute is used to indicate the data transmission amount between two satellites and the communication strategies that can be implemented; performing message passing on the first graph data by using a pre-trained graph neural network to determine the corresponding second graph data; and uniquely determining the communication strategies for one-way communication between the satellites according to the edge feature information of the one-way edges in the second graph data.

[0009] In the embodiments of the present disclosure, first, a feature extraction operation based on a multi-dimensional matrix is used to determine a feasible communication strategy between each low-earth orbit satellite according to the distribution of multiple low-earth orbit satellites and the signal transmission environment. Then, a message passing operation based on a graph neural network is used to further determine a unique communication strategy for one-way communication between each satellite according to the data transmission volume of each cell corresponding to each satellite and the one-way data transmission volume between each satellite, in combination with the feasible communication strategy between each low-earth orbit satellite that has been determined. Therefore, through two decisions, the present application combines the distribution of multiple low-earth orbit satellites and the signal transmission environment, and combines the data transmission volume and data reception volume of each cell, so as to take into account the transmission advantages of microwave links and laser links and reasonably schedule microwave links and laser links. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] The drawings described herein are used to provide a further understanding of the present disclosure, form a part of this application, and the schematic embodiments of the present disclosure and their descriptions are used to explain the present disclosure and do not constitute an improper limitation to the present disclosure. In the drawings: Figure 1 is a schematic diagram for implementing the satellite communication scheduling system based on a microwave-laser dual link according to Embodiment 1 of the present disclosure; Figure 2 is a schematic flowchart of the satellite communication scheduling method based on a microwave-laser dual link according to the first aspect of Embodiment 1 of the present disclosure; Figure 3 is a schematic diagram of the first multi-dimensional matrix according to Embodiment 1 of the present disclosure; Figure 4 is a schematic diagram of the second multi-dimensional matrix according to Embodiment 1 of the present disclosure; Figure 5 is a schematic diagram of the graph structure based on multiple low-earth orbit satellites according to Embodiment 1 of the present disclosure; Figure 6 is a schematic diagram of obtaining second graph data by message passing on the first graph data according to the first aspect of Embodiment 1 of the present disclosure; Figure 7 is a schematic diagram of determining the communication strategy information for one-way communication between satellites according to the edge features of the second graph data according to the first aspect of Embodiment 1 of the present disclosure; Figure 8 is a schematic diagram of dividing the atmospheric space into grids according to Embodiment 1 of the present disclosure; Figure 9 is a schematic diagram of determining the inter-satellite grid between satellites according to Embodiment 1 of the present disclosure; Figure 10 is a schematic diagram of generating the second multi-dimensional matrix according to the first multi-dimensional matrix according to Embodiment 1 of the present disclosure; Figure 11 It is a schematic diagram of the matrix conversion model according to Embodiment 1 of the present disclosure; Figure 12 It is a schematic diagram of the satellite communication scheduling device based on the microwave-laser dual link according to Embodiment 2 of the present disclosure; and Figure 13 It is a schematic diagram of the satellite communication scheduling equipment based on the microwave-laser dual link according to Embodiment 3 of the present disclosure. Detailed implementation manners

[0011] In order to enable those skilled in the art to better understand the technical solutions of the present disclosure, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only a part of the embodiments of the present disclosure, rather than all of the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present disclosure.

[0012] It should be noted that the terms "first", "second", etc. in the specification and claims of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily need to be used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of the present disclosure described here can be implemented in an order different from those illustrated or described here. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units does not necessarily need to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products, or devices.

[0013] Embodiment 1 According to this embodiment, a method embodiment of a satellite communication scheduling method based on a microwave-laser dual link is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order from that here.

[0014] Figure 1 A schematic diagram of the satellite communication scheduling system based on the microwave-laser dual link according to this embodiment is shown. The system includes: a plurality of low-earth orbit satellites S1~S m and a ground station 100. Among them, a plurality of low-earth orbit satellites S1~S m can communicate with the ground station 100 respectively.

[0015] Based on the above system, this embodiment proposes a satellite communication scheduling method based on a microwave-laser dual link. Among them Figure 2 shows a schematic flowchart of this method. Refer to Figure 2 as shown, this method includes: S202: Construct a first multi-dimensional matrix according to the position information of multiple satellites. The first multi-dimensional matrix is used to indicate the positions between each satellite and the position environment information related to signal communication; S204: Perform feature extraction on the first multi-dimensional matrix to determine a second multi-dimensional matrix, where the second multi-dimensional matrix is used to indicate the communication strategies that can be implemented between each satellite. The communication strategies include: communicating through a microwave link and communicating through a laser link; S206: Construct a first graph data according to the second multi-dimensional matrix. The nodes of the first graph data respectively correspond to each satellite. The node attribute information is used to indicate the data transmission volume and data reception volume of the cells corresponding to each satellite, and the edges of the first graph data are used to indicate that communication can be carried out between two satellites. The edge attribute is used to indicate the data transmission volume between two satellites and the communication strategies that can be implemented; S208: Use a pre-trained graph neural network to perform message passing on the first graph data, so as to determine the corresponding second graph data; and S210: According to the edge feature information of the unidirectional edges in the second graph data, uniquely determine the communication strategies for unidirectional communication between each satellite.

[0016] Specifically, the ground station 100 can determine the position information of multiple satellites S1~S m according to the ephemeris information of multiple satellites S1~S m . Then, the ground station 100 can obtain the meteorological data of each region from websites such as meteorological data platforms, and thus determine the first multi-dimensional matrix related to the environmental element information characterizing the signal communication between each satellite according to the position information of the satellites S1~S m and the obtained meteorological data, where c represents the number of environmental element types. Among them , where c represents the number of environmental element types. Among them Figure 3 shows a schematic diagram of the first multi-dimensional matrix.

[0017] Among them, in this embodiment, use x i,j,u to represent each element data in the first multi-dimensional matrix, where i, j = 1~m, and u = 1~c. Thus x i,j,uIndicates the position environment information between the \(i\)-th satellite and the \(j\)-th satellite with respect to the \(u\)-th information type. Specifically, the information type can include, for example: straight-line distance, satellite orbital inclination difference, atmospheric visibility, precipitation intensity, refractive index structure constant, etc. The above position environment information not only affects the attenuation of laser link communication, but also affects the attenuation of microwave link communication.

[0018] So that the ground station 100 can target multiple satellites S1~S m For each pair of satellites S i And S j , determine the position environment information \(x\) of each type between this pair of satellites i,j,u , thus forming the first multi-dimensional matrix (S202). In addition, Can be regarded as the position environment information vector between satellites S i And S j (see Figure 3 Shown), each element of this position environment information vector is used to indicate the position environment information of different types between satellites S i And S j , which will not be elaborated here. The specific method for determining the first multi-dimensional matrix \(X\) will be described in detail later.

[0019] Then, the ground station 100 performs feature extraction on the first multi-dimensional matrix \(X\), thereby generating a second multi-dimensional matrix corresponding to the first multi-dimensional matrix . Among them, Figure 4 Shows a schematic diagram of the second multi-dimensional matrix \(Y\).

[0020] In this embodiment, \(y\) i,j,v Represents the probability of implementing the \(v\)-th communication strategy between satellites S i And S j , where \(v = 1\sim2\). The first communication strategy is for satellites S i And S j To communicate via a microwave link; the second communication strategy is for satellites S i And S j To communicate via a microwave link. Thus, when \(y\) i,j,v ≥50%, it means that satellites S i And S j Can communicate via the \(v\)-th communication strategy, otherwise it means that satellites S i And S j Cannot communicate via the \(v\)-th communication strategy.

[0021] For example, when \(y\) i,j,1 ≥50%, it means that satellites S i And S jcan communicate with each other via a microwave link; otherwise, it indicates satellite S i and S j cannot communicate with each other via a microwave link; when y i,j,1 ≥ 50%, it indicates that satellite S i and S j can communicate with each other via a laser link; otherwise, it indicates satellite S i and S j cannot communicate with each other via a laser link; when y i,j,1 and y i,j,2 are both ≥ 50%, it indicates that satellite S i and S j can communicate with each other via both a microwave link and a laser link; when y i,j,1 and y i,j,2 are both < 50%, it indicates that satellite S i and S j cannot communicate with each other via either a microwave link or a laser link.

[0022] Thus, the ground station 100 can determine the probability y of the communication strategy that can be implemented between each pair of satellites S1 to S m for multiple satellites, and thus form a second multi-dimensional matrix i and S j , thereby forming a second multi-dimensional matrix i,j,v (S204). In addition, can be regarded as the communication strategy information vector between satellite S and S i and S j (see Figure 4 ), and the elements of this information vector are used to indicate the probability information of each communication strategy that can be implemented between S i and S j , which will not be elaborated here. The specific method for determining the second multi-dimensional matrix Y will be described in detail later.

[0023] Then, the ground station 100 defines the graph structure according to the second multi-dimensional matrix. Specifically, the ground station 100 constructs nodes N1 to N m corresponding to each satellite S1 to S m . Then, the ground station 100 constructs an adjacency matrix m corresponding to nodes N1 to N according to the following rules: 1) w i,j = 1 when y i,j,1 ≥ 50% or y i,j,2 ≥ 50%; 2) w i,j = 0 when y i,j,1 and yi,j,2 When both are < 50%.

[0024] Then, the ground station 100 constructs the edges of the graph structure according to the adjacency matrix, thereby defining the graph structure. Considering the bidirectionality of data transmission between satellites, this graph structure is a directed graph, and two nodes are connected by two edges with opposite directions. Among them Figure 5 shows a schematic diagram of this graph structure. In Figure 5 the shown graph structure, there are a total of L unidirectional edges.

[0025] Then, with further reference to Figure 6 as shown, the ground station 100 determines the first graph data corresponding to this graph structure. The first graph data includes node attribute information A1~A corresponding to each node N1~N m and edge attribute information B1~B corresponding to each edge E1~E m L L . Among them, the node attribute information A1~A m is used to indicate the data transmission volume and data reception volume of the corresponding cells C1~C of each satellite S1~S m m . The edge attribute information B1~B L is used to indicate the data transmission volume between its corresponding two satellites and the communication strategies that can be implemented. Among them, regarding the node attribute information A1~A m and the edge attribute information B1~B L will be described in detail later. Thus, in this way, the ground station 100 constructs the first graph data (S206).

[0026] Then, the ground station 100 uses the message passing model based on the graph neural network to perform message passing on the first graph data to extract the features of the first graph data, and then generates the second graph data corresponding to the first graph data. The second graph data is the feature information corresponding to the first graph data (S208). With reference to Figure 6 as shown, the second graph data includes node features F1~F corresponding to each graph node m and edge features G1~G corresponding to each unidirectional E1~E L L . Regarding the specific operation of message passing, it will be described in detail later.

[0027] Finally, with reference to Figure 7 as shown, the ground station 100 inputs the edge features G k corresponding to each unidirectional edge E k into the pre-set fully connected layer and binary classifier, so as to uniquely determine the unidirectional edge E k ​​​​The corresponding communication type. For example, the binary classifier outputs a two-dimensional classification vector Q k , which contains two elements, namely the probability of communicating via a microwave link and the probability of communicating via a laser link (S210). Thus, the strategy with a higher probability can be used as the one-way communication strategy corresponding to the one-way edge. For example, in Figure 6 , when in the classification vector Q1 corresponding to the one-way edge E1, the probability corresponding to communicating via a microwave link is higher, it indicates that satellite S2 communicates with satellite S1 via a microwave link. If in the classification vector Q2 corresponding to the one-way edge E2, the probability corresponding to communicating via a laser link is higher, it indicates that satellite S1 communicates with satellite S2 via a laser link. Therefore, the two-dimensional classification vector Q k can also be regarded as the communication strategy information for the one-way edge E k .

[0028] Then, the ground station 100 sends the determined communication strategy information corresponding to the communication types of each one-way edge to each low-earth orbit satellite S1~S m . Thus, the low-earth orbit satellites S1~S m can select microwave link communication or laser link communication for data transmission according to the corresponding communication strategy information.

[0029] Thus, in this way, the present application first uses the feature extraction operation based on a multi-dimensional matrix to determine the feasible communication strategies between each low-earth orbit satellite according to the distribution of multiple low-earth orbit satellites and the signal transmission environment, and then uses the message passing operation based on a graph neural network to further determine the unique communication strategy for one-way communication between each satellite according to the data transmission volume of the cells corresponding to each satellite and the one-way data transmission volume between each satellite, in combination with the feasible communication strategies between each low-earth orbit satellite that have been determined. Thus, through two decisions, the present application combines the distribution of multiple low-earth orbit satellites and the signal transmission environment, and combines the data transmission volume and data reception volume of each cell, so as to be able to take into account the transmission advantages of both microwave links and laser links and reasonably schedule the microwave links and laser links.

[0030] Optionally, the operation of constructing the first multi-dimensional matrix based on the position information of multiple satellites includes: determining the distances between the satellites and the differences in orbital inclination angles according to the position information of the multiple satellites; determining the atmospheric environment information between the satellites according to the position information of the multiple satellites; and constructing the first multi-dimensional matrix according to the distances, differences in orbital inclination angles, and atmospheric environment information between the satellites. Further, the operation of determining the atmospheric environment information between the satellites according to the position information of the multiple satellites includes: dividing the atmospheric space into multiple grids; determining the grids corresponding to the satellites according to the position information of the multiple satellites; determining the atmospheric environment information of the grids between the satellites through which the straight line connections between the satellites pass; and determining the atmospheric environment information between the satellites according to the atmospheric environment information of the grids between the satellites.

[0031] Specifically, in this embodiment, the ground station 100 may use the distances between the satellites as the first channel matrix of the first multi-dimensional matrix X, i.e., x i,j,1 for representing the distance i between satellite S j and S

[0032] In addition, the ground station 100 may use the differences in orbital inclination angles between the satellites as the second channel matrix of the first multi-dimensional matrix X, i.e., x i,j,2 for representing the difference in orbital inclination angle i between satellite S j and S

[0033] Then, as shown in Figure 8 the ground station 100 may divide the atmospheric space into multiple grids Z1~Z 16 , so as to determine the atmospheric environment information within each grid, such as atmospheric visibility, precipitation intensity, refractive index structure constant, etc. Thus, the atmospheric environment information corresponding to each grid Z1~Z 16 can be determined, and this environmental information can be represented in the form of a vector, for example, different elements of each vector represent the numerical values of different types of atmospheric environment information, such as atmospheric visibility, precipitation intensity, refractive index structure constant, etc.

[0034] Then, as shown in Figure 9 the ground station 100 may determine the grids corresponding to each satellite S1~S m according to the position information of each satellite S1~S m .

[0035] Then, further as shown in Figure 9 the ground station 100 determines the grids passed by the straight line connections between the satellites, i.e., the grids between the satellites. For example, in Figure 9Among them, the inter-satellite grids between satellites S1 and S2 are grids Z2 and Z3, the inter-satellite grid between satellites S1 and S3 is grid Z6, and the inter-satellite grids between satellites S2 and S3 are grids Z7 and Z8.

[0036] Then, ground station 100 determines the atmospheric environment information between each pair of satellites according to the atmospheric environment information of the inter-satellite grids. Among them, when there are multiple inter-satellite grids (for example, the inter-satellite grids between satellites S1 and S2 are grids Z2 and Z3, and the inter-satellite grids between satellites S2 and S3 are grids Z7 and Z8), the average value of the atmospheric environment information of multiple grids can be calculated as the atmospheric environment information between the satellites. For example, for satellites S1 and S2, the average value of the atmospheric visibility of grids Z2 and Z3 can be used as the atmospheric visibility between satellites S1 and S2, the average value of the precipitation intensity of grids Z2 and Z3 can be used as the precipitation intensity between satellites S1 and S2, and the average value of the refractive index structure constant of grids Z2 and Z3 can be used as the refractive index structure constant between satellites S1 and S2, and so on. By analogy, the atmospheric environment information between each pair of satellites can be determined.

[0037] Then, ground station 100 takes the different types of atmospheric environment information between each pair of satellites S i and S j as the data of channels 3 to channel c of the first multi-dimensional matrix, that is, x i,j,3 ~x i,j,c respectively correspond to the different types of atmospheric environment information between satellites S i and S j .

[0038] In this way, ground station 100 constructs a first-dimensional matrix for indicating the positions and the atmospheric environment between each pair of satellites S i and S j . .

[0039] Optionally, the operation of performing feature extraction on the first multi-dimensional matrix to determine the second multi-dimensional matrix includes: inputting the first multi-dimensional matrix into a matrix conversion model, and generating a third multi-dimensional matrix corresponding to the first multi-dimensional matrix through the matrix conversion model, where the third multi-dimensional matrix includes four channels, the first channel and the second channel are used to indicate whether microwave link communication can be implemented between each pair of satellites, and the third channel and the fourth channel are used to indicate whether laser link communication can be implemented between each pair of satellites; using a first classifier to classify the elements corresponding to the first channel and the second channel to determine the probability that microwave link communication can be implemented between each pair of satellites; using a second classifier to classify the elements corresponding to the third channel and the fourth channel to determine the probability that laser link communication can be implemented between each pair of satellites; and taking the results output by the first classifier and the second classifier as the second multi-dimensional matrix.

[0040] Specifically, referring to Figure 10 as shown, the ground station 100 inputs the first multi-dimensional matrix X into a pre-trained matrix conversion model, thereby generating a third multi-dimensional matrix with four channels . Among them, the element d in the first channel of the third multi-dimensional matrix i,j,1 is used to indicate the integral value for determining that microwave link communication can be implemented between satellites S i and S j ; the element d in the second channel i,j,2 is used to indicate the integral value for determining that microwave link communication cannot be implemented between satellites S i and S j ; the element d in the third channel i,j,3 is used to indicate the integral value for determining that laser link communication can be implemented between satellites S i and S j ; the element d in the fourth channel i,j,4 is used to indicate the integral value for determining that laser link communication cannot be implemented between satellites S i and S j .

[0041] Then, the first classifier based on binary classification is used to classify the corresponding elements of the first channel and the second channel, and determine the probability that microwave link communication can be implemented between each satellite. For example: (h1, h2) = softmax(d i,j,1 , d i,j,2 ); and y i,j,1 = h1.

[0042] In addition, the second classifier based on binary classification is used to classify the corresponding elements of the third channel and the fourth channel, and determine the probability that laser link communication can be implemented between each satellite. For example: (h3, h4) = softmax(d i,j,3 , d i,j,4 ); and y i,j,1 = h3.

[0043] Thus, in the above manner, the ground station 100 can determine the second multi-dimensional matrix Y.

[0044] Further, the matrix transformation model includes an encoding network and a decoding network. The encoding network is used to encode the first multi-dimensional matrix to generate a corresponding feature map, and the decoding network is used to decode the feature map to generate a second multi-dimensional matrix. Moreover, feature fusion is performed between the encoding network and the decoding network in a skip connection manner. Further, the encoding network includes a plurality of encoding modules arranged in cascade, where each encoding module includes a convolutional layer and a pooling layer; the decoding network includes a plurality of decoding modules arranged in cascade, where each decoding module includes a transposed convolutional layer and an upsampling layer; and the feature maps output by each encoding module are fused with the feature maps to be input to the corresponding decoding module in a skip connection manner.

[0045] Specifically, referring to Figure 11 as shown, the matrix transformation model includes an encoding network and a decoding network. The encoding network is used to encode the first multi-dimensional matrix X to generate a feature map, and the decoding network is used to decode the feature map generated by the encoding network to generate a third multi-dimensional matrix D. Moreover, feature fusion is performed between the encoding network and the decoding network in a skip connection manner.

[0046] In addition, further referring to Figure 11 as shown, the encoding network includes a plurality of encoding modules. Each encoding module includes a plurality of convolutional layers and a pooling layer, so that each encoding module can be regarded as a convolutional neural network unit, and the first multi-dimensional matrix X is feature-extracted by convolution. Thus, each encoding module will output a corresponding feature map. After the encoding network encodes the first multi-dimensional matrix X multiple times using a plurality of encoding modules, the final feature map is output to the decoding network.

[0047] Although the figure shows that the encoding network includes 3 encoding modules and each encoding module includes 3 convolutional layers, the specific number of encoding modules and the number of convolutional layers in each encoding module can be deployed according to the actual situation. Similarly, the downsampling ability of the pooling layer can also be deployed according to the actual situation, which will not be elaborated here.

[0048] Further referring to Figure 11 as shown, the decoding network includes a plurality of decoding modules. Among them, decoding modules 1 to 3 include their respective transposed convolutional layers and upsampling layers, which are used to perform decoding operations on the feature map. In addition, further referring to Figure 11 as shown, the last decoding module includes not only a transposed convolutional layer but also a convolutional layer. The convolutional layer is, for example, a convolutional layer of this structure, that is, the convolutional layer includes 4 convolutional kernels, and each convolutional kernel has only one weight parameter. Thus, the final decoding network can output the third multi-dimensional matrix .

[0049] Thus, in the present application, by using the convolution-based encoding operation and the transposed convolution-based decoding operation, the first multi-dimensional matrix for indicating the position environment information between each satellite can be mapped to the third multi-dimensional matrix for indicating the feasibility of microwave link communication and the feasibility of laser link communication between each satellite.

[0050] In the prior art, an encoding network and a decoding network are often used for image processing, such as region segmentation of an image. However, in the present application, the models of the encoding network and the decoding network are used to implement the mapping between multi-dimensional matrices representing different meanings (for example, mapping from the space indicating position environment information to the space indicating communication strategy information), so that the processing capabilities of the spatial data of the encoding network and the decoding network can be utilized to accurately evaluate and predict the communication strategies between each satellite.

[0051] Further, the operation of constructing the first graph data according to the second multi-dimensional matrix includes: splicing the following information of the cells corresponding to each satellite to obtain the node attribute information corresponding to each satellite: the current data transmission request volume of the cell; the data transmission volumes respectively transmitted by the cell in a plurality of consecutive periods before the current moment; and the data reception volumes respectively received by the cell in a plurality of consecutive periods before the current moment. Further, the operation of constructing the first graph data according to the second multi-dimensional matrix further includes: splicing the following information corresponding to each one-way edge to obtain the edge attribute information corresponding to each one-way edge: the communication strategy information corresponding to the one-way edge and the data transmission volumes respectively transmitted through the one-way edge in a plurality of consecutive periods before the current moment.

[0052] Specifically, referring to Figure 6 shown, in the present application, each piece of node attribute information A1 to A m includes the following information of the cells C1 to C m corresponding to the respective satellites S1 to S m : 1) the current data transmission request volume a0 of the cell; 2) the data transmission volumes a1 to a n respectively transmitted by the cell in n consecutive periods before the current moment; 3) the data reception volumes a n+1 to a 2n+1 respectively received by the cell in n consecutive periods before the current moment.

[0053] For example, A i =[a i,0 , a i,1 , a i,2 , a i,3 ,..., a i,2n+1 T , where a i,0 represents the current data transmission request volume of the cell C i ; a​i,1 ~a i,n represents cell C i the data transmission amounts transmitted in n consecutive cycles before the current moment respectively; and a i,n+1 ~a i,2n+1 represents cell C i the data reception amounts received in n consecutive cycles before the current moment respectively.

[0054] In addition, for each unidirectional edge E k (k = 1 to L), its edge attribute B k includes n + 2 elements b0 to b n+1 . Among them, the elements b0 and b1 correspond to the communication strategy information between the two satellites corresponding to the two nodes connected by the unidirectional edge respectively, and the elements b2 to b n+1 correspond to the data amounts transmitted from the cell corresponding to the source node of the unidirectional edge to the cell corresponding to the destination node in n consecutive cycles before the current moment respectively.

[0055] For example, for Figure 5 the unidirectional edge E1 shown in, its attribute B1 is [b 1,0 , b 1,1 , b 1,2 ,..., b 1,n+1 T . Among them, the elements b 1,0 and b 1,1 correspond to y 1,2,1 and y 1,2,2 respectively, and the elements b 1,2 ~b 1,n+1 correspond to the data amounts transmitted from cell C2 to cell C1 in n consecutive cycles before the current moment. And for the unidirectional edge E2, its attribute B2 is [b 2,0 , b 2,1 , b 2,2 ,..., b 2,n+1 T . Among them, the elements b 2,0 and b 2,1 correspond to y 1,2,1 and y 1,2,2 respectively, and the elements b 2,2 ~b 2,n+1 correspond to the data amounts transmitted from cell C1 to cell C2 in n consecutive cycles before the current moment. And so on, the edge attributes B1 to B Figure 5 of each unidirectional edge E1 to E L shown in can be determined. L .

[0056] Thus, in this way, the ground station 100 determines the ones corresponding to each graph node N1 to N​​m The corresponding node attribute information A1 to A m , and for each unidirectional edge E1 to E L The corresponding edge attribute information B1 to B L , thereby constructing the first graph data.

[0057] Optionally, use a pre-trained graph neural network to perform message passing on the first graph data to determine the corresponding second graph data. The operations include performing multi-layer message passing on the first graph data to determine the corresponding second graph data. And among them, a single message passing includes: according to the node information of the target node to be passed and the edge information of the unidirectional edge with the target node as the source node, fusing to generate a first fusion feature; and according to the first fusion feature, using a first multi-layer perceptron to generate updated node information corresponding to the target node. And among them, a single message passing also includes: according to the edge information of the target unidirectional edge to be passed and the node information of the source node and the destination node associated with the unidirectional edge, fusing to generate a second fusion feature; and according to the second fusion feature, using a second multi-layer perceptron to generate updated edge information corresponding to the target unidirectional edge.

[0058] Specifically, although Figure 6 the message passing process described in 1. Message passing for node data 1) For the target node to be passed, fuse the node information of the target node to be passed and the edge information of the unidirectional edge with the target node as the source node to generate a first fusion feature. For example, for the graph node N1, the node attribute information A1 corresponding to the graph node in the first graph data, and the edge attribute information B2 and B3 of the unidirectional edges E2 and E3 with the graph node N1 as the source node can be feature-fused to generate a first fusion feature.

[0059] Specifically, for example, the average value of the edge attribute information B2 and B3 can be calculated to obtain the edge attribute mean information, and then the edge attribute mean information is concatenated with the node attribute information A1 to obtain the first fusion feature.

[0060] 2) Input the first fusion feature into the first MLP layer to obtain the updated node information corresponding to the graph node N1. The first MLP layer is specifically used for message passing operations on graph nodes.

[0061] Thus, through the above method, the node attribute information A of each graph node i is updated.

[0062] 2. Message passing for edge data 1) For the target unidirectional edges with transmission, according to the edge information of the target unidirectional edges to be transmitted and the node information of the source node and the destination node associated with the unidirectional edges. For example, taking the unidirectional edge E1 as an example, the edge attribute information B1 of the unidirectional edge E1, the node attribute information A2 of the source node N2, and the node attribute information A1 of the destination node N1 are spliced to generate the second fusion feature.

[0063] 2) Input the second fusion feature into the second MLP layer to obtain the updated edge information corresponding to the unidirectional edge E1. The second MLP layer is dedicated to performing message passing operations on edge nodes.

[0064] Thus, in the above way, the edge attribute information B of each unidirectional edge i is updated.

[0065] Thus, one round of message passing is completed through the above operations. Then, based on the graph data output by the first round of message passing above, the next round of message passing is performed in the same way, and so on iteratively until the second graph data is generated.

[0066] Thus, in the above way, the first graph data is constructed according to the cells corresponding to each satellite and the data transmission information of the unidirectional communication between each satellite. Then, the second graph data is generated according to the first graph data by using the message passing operation based on the graph neural network. Thus, the features of the second graph data can not only reflect the feasible unidirectional communication strategies between each satellite, but also reflect the data transmission volume of the cells corresponding to each satellite and the data transmission volume of the unidirectional communication between each satellite, so that the communication of multiple satellites can be scheduled more accurately.

[0067] Optionally, the operation of uniquely determining the communication strategy of the unidirectional communication between each satellite according to the edge feature information of the unidirectional edges in the second graph data includes: inputting the edge features of the unidirectional edges into a fully connected layer and a communication strategy classification unit based on binary classification to determine the communication strategy uniquely corresponding to the unidirectional edges.

[0068] As described above, referring to Figure 7 shown, the ground station 100 inputs the edge features G k corresponding to each unidirectional edge E k into a pre-set fully connected layer and a binary classifier, so as to uniquely determine the communication type corresponding to the unidirectional edge E k . For example, the binary classifier outputs a two-dimensional classification vector Q k , and the two-dimensional classification vector contains two elements, namely the probability of communicating through the microwave link and the probability of communicating through the laser link.

[0069] In addition, referring to Figure 1As shown, according to the second aspect of this embodiment, a storage medium is provided. The storage medium includes a stored program, wherein when the program runs, the method described in any one of the above is executed by a processor.

[0070] Thus, according to this embodiment, first, by using the feature extraction operation based on the multi-dimensional matrix, a feasible communication strategy between each low-earth orbit satellite is determined according to the distribution of multiple low-earth orbit satellites and the signal transmission environment. Then, by using the message passing operation based on the graph neural network, according to the data transmission volume of each cell corresponding to each satellite and the one-way data transmission volume between each satellite, combined with the feasible communication strategy between each low-earth orbit satellite that has been determined, the unique communication strategy for one-way communication between each satellite is further determined. Therefore, through two decisions in this application, combining the distribution of multiple low-earth orbit satellites and the signal transmission environment, and combining the data transmission volume and data reception volume of each cell, the transmission advantages of both the microwave link and the laser link can be taken into account, and the microwave link and the laser link can be reasonably scheduled.

[0071] It should be noted that for the foregoing method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should know that the present invention is not limited by the described action sequence, because according to the present invention, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to the present invention.

[0072] Through the description of the above embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be implemented by means of software plus a necessary general hardware platform. Of course, it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disc), and includes several instructions for causing a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in various embodiments of the present invention.

[0073] Embodiment 2 Figure 12 Shows a satellite communication scheduling device 1200 based on a microwave-laser dual link according to this embodiment. The device 1200 corresponds to the method described in the first aspect of Embodiment 1. Refer to Figure 12As shown, the device 1200 includes: a first matrix construction module 1210, configured to construct a first multi-dimensional matrix based on the position information of multiple satellites, where the first multi-dimensional matrix is used to indicate the positions between the satellites and the position environment information related to signal communication; a second matrix determination module 1220, configured to perform feature extraction on the first multi-dimensional matrix to determine a second multi-dimensional matrix, where the second multi-dimensional matrix is used to indicate the communication strategies that can be implemented between the satellites, and the communication strategies include: communicating via a microwave link and communicating via a laser link; a first graph data construction module 1230, configured to construct first graph data based on the second multi-dimensional matrix, where the nodes of the first graph data respectively correspond to the satellites, the node attribute information is used to indicate the data transmission volume and data reception volume of the cells corresponding to the satellites, and the edges of the first graph data are used to indicate that communication can be performed between two satellites, and the edge attributes are used to indicate the data transmission volume between two satellites and the communication strategies that can be implemented; a message passing module 1240, configured to perform message passing on the first graph data by using a pre-trained graph neural network, so as to determine the corresponding second graph data; and a communication strategy scheduling module 1250, configured to uniquely determine the communication strategies for one-way communication between the satellites according to the edge feature information of the one-way edges in the second graph data.

[0074] Thus, according to this embodiment, first, by using the feature extraction operation based on the multi-dimensional matrix, the feasible communication strategies between multiple low-earth orbit satellites are determined according to the distribution of the low-earth orbit satellites and the signal transmission environment. Then, by using the message passing operation based on the graph neural network, according to the data transmission volume of the cells corresponding to the satellites and the one-way data transmission volume between the satellites, combined with the feasible communication strategies between the low-earth orbit satellites that have been determined, the unique communication strategies for one-way communication between the satellites are further determined. Therefore, through two decisions in this application, by combining the distribution of multiple low-earth orbit satellites and the signal transmission environment, and combining the data transmission volume and data reception volume of each cell, the transmission advantages of both the microwave link and the laser link can be taken into account, and the microwave link and the laser link can be reasonably scheduled.

[0075] Embodiment 3 Figure 13 Shows a satellite communication scheduling device 1300 based on a microwave-laser dual link according to this embodiment, and the device 1300 corresponds to the method described in the first aspect of Embodiment 1. Refer to Figure 13As shown, the device 1300 includes: a processor 1310; and a memory 1320, connected to the processor 1310, for providing instructions for the processor 1310 to process the following steps: constructing a first multi-dimensional matrix based on the position information of multiple satellites, where the first multi-dimensional matrix is used to indicate the positions between the satellites and the position environment information related to signal communication; performing feature extraction on the first multi-dimensional matrix to determine a second multi-dimensional matrix, where the second multi-dimensional matrix is used to indicate the communication strategies that can be implemented between the satellites, and the communication strategies include: communicating through a microwave link and communicating through a laser link; constructing first graph data based on the second multi-dimensional matrix, where the nodes of the first graph data respectively correspond to the satellites, the node attribute information is used to indicate the data transmission volume and data reception volume of the cells corresponding to the satellites, and the edges of the first graph data are used to indicate that communication can be carried out between two satellites, and the edge attributes are used to indicate the data transmission volume between two satellites and the communication strategies that can be implemented; using a pre-trained graph neural network to perform message passing on the first graph data, thereby determining the corresponding second graph data; and uniquely determining the communication strategies for one-way communication between the satellites according to the edge feature information of the one-way edges in the second graph data.

[0076] Thus, according to this embodiment, first, by using the feature extraction operation based on the multi-dimensional matrix, the feasible communication strategies between multiple low-earth orbit satellites are determined according to the distribution of the low-earth orbit satellites and the signal transmission environment. Then, by using the message passing operation based on the graph neural network, according to the data transmission volume of the cells corresponding to each satellite and the one-way data transmission volume between each satellite, combined with the feasible communication strategies between the low-earth orbit satellites that have been determined, the unique communication strategies for one-way communication between each satellite are further determined. Thus, through two decisions in this application, combined with the distribution of multiple low-earth orbit satellites and the signal transmission environment, and combined with the data transmission volume and data reception volume of each cell, the transmission advantages of both the microwave link and the laser link can be taken into account, and the microwave link and the laser link can be reasonably scheduled.

[0077] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0078] In the above embodiments of the present invention, the descriptions of each embodiment have their own emphases. For the parts not detailed in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0079] In several embodiments provided in the present application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections between each other can be through some interfaces. The indirect couplings or communication connections of units or modules can be in electrical or other forms.

[0080] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0081] In addition, in each embodiment of the present invention, the functional units can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.

[0082] If the above-mentioned integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in each embodiment of the present invention. The foregoing storage medium includes: various media such as USB flash drives, read-only memories (ROMs), random access memories (RAMs), mobile hard disks, magnetic disks, or optical discs that can store program codes.

[0083] The above are only the preferred embodiments of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present invention.

Claims

1. A satellite communication scheduling method based on dual links, characterized in that Including: Construct a first multi-dimensional matrix based on the position information of multiple satellites, where the first multi-dimensional matrix is used to indicate the positions between each satellite and the position environment information related to signal communication; Extract features from the first multi-dimensional matrix to determine a second multi-dimensional matrix, where the second multi-dimensional matrix is used to indicate the communication strategies that can be implemented between each satellite, and the communication strategies include: communicating via a microwave link and communicating via a laser link; Construct a first graph data according to the second multi-dimensional matrix, where the nodes of the first graph data respectively correspond to each satellite, the node attribute information is used to indicate the data transmission volume and data reception volume of the cells corresponding to each satellite, and the edges of the first graph data are used to indicate that communication can be carried out between two satellites, and the edge attributes are used to indicate the data transmission volume between two satellites and the communication strategies that can be implemented; Use a pre-trained graph neural network to perform message passing on the first graph data, thereby determining the corresponding second graph data; and According to the edge feature information of the unidirectional edges in the second graph data, uniquely determine the communication strategies for unidirectional communication between each satellite.

2. The method according to claim 1, wherein The operation of constructing a first multi-dimensional matrix based on the position information of multiple satellites includes: Determine the distances and orbital inclination differences between each satellite according to the position information of the multiple satellites; Determine the atmospheric environment information between each satellite according to the position information of the multiple satellites; and Construct the first multi-dimensional matrix according to the distances, orbital inclination differences, and atmospheric environment information between the multiple satellites, and among them, The operation of determining the atmospheric environment information between each satellite according to the position information of the multiple satellites includes: Divide the atmospheric space into multiple grids; Determine the grids corresponding to each satellite according to the position information of the multiple satellites; Determine the atmospheric environment information of the inter-satellite grids passed by the straight line connections between each satellite; and Determine the atmospheric environment information between each satellite according to the atmospheric environment information of the inter-satellite grids.

3. The method according to claim 1, characterized in that The operation of extracting features from the first multi-dimensional matrix to determine a second multi-dimensional matrix includes: Input the first multi-dimensional matrix into a matrix conversion model, and generate a third multi-dimensional matrix corresponding to the first multi-dimensional matrix through the matrix conversion model, where the third multi-dimensional matrix includes four channels, the first channel and the second channel are used to indicate whether microwave link communication can be implemented between each satellite, and the third channel and the fourth channel are used to indicate whether laser link communication can be implemented between each satellite; Use a first classifier to classify the elements corresponding to the first channel and the second channel to determine the probability that microwave link communication can be implemented between each satellite; Use a second classifier to classify the elements corresponding to the third channel and the fourth channel to determine the probability that laser link communication can be implemented between each satellite; and Use the results output by the first classifier and the second classifier as the second multi-dimensional matrix.

4. The method according to claim 3, wherein The matrix conversion model includes an encoding network and a decoding network. The encoding network is used to encode the first multi-dimensional matrix to generate a corresponding feature map, and the decoding network is used to decode the feature map to generate the third multi-dimensional matrix, and wherein feature fusion is also performed between the encoding network and the decoding network in a skip connection manner.

5. The method according to claim 4, wherein the encoding network includes a plurality of encoding modules arranged in cascade, and each encoding module includes a convolutional layer and a pooling layer; the decoding network includes a plurality of decoding modules arranged in cascade, and each decoding module includes a transposed convolutional layer and an upsampling layer; and the feature maps output by each encoding module are subjected to feature fusion with the feature maps to be input to the corresponding decoding module in a skip connection manner.

6. The method according to claim 4, characterized in that The operation of using a pre-trained graph neural network to perform message passing on the first graph data to determine the corresponding second graph data includes performing multi-layer message passing on the first graph data to determine the corresponding second graph data, and wherein a single message passing includes: fusing to generate a first fused feature according to the node information of the target node to be passed and the edge information of the unidirectional edge with the target node as the source node; and generating updated node information corresponding to the target node by using a first multi-layer perceptron according to the first fused feature, and wherein a single message passing further includes: fusing to generate a second fused feature according to the edge information of the target unidirectional edge to be passed and the node information of the source node and the destination node associated with the unidirectional edge; and generating updated edge information corresponding to the target unidirectional edge by using a second multi-layer perceptron according to the second fused feature.

7. The method according to claim 1, characterized in that, The operation of uniquely determining the communication strategy for one-way communication between each satellite according to the edge feature information of the unidirectional edge in the second graph data includes: inputting the edge feature of the unidirectional edge into a fully connected layer and a communication strategy classification unit based on binary classification to determine the communication strategy uniquely corresponding to the unidirectional edge.

8. A storage medium, characterized in that, The storage medium includes a stored program, wherein, when the program runs, the method according to any one of claims 1 to 7 is executed by a processor.

9. A satellite communication scheduling device based on a dual-link, characterized in that, Including: a first matrix construction module for constructing a first multi-dimensional matrix according to the position information of a plurality of satellites, and the first multi-dimensional matrix is used to indicate the positions between each satellite and the position environment information related to signal communication; a second matrix determination module for performing feature extraction on the first multi-dimensional matrix to determine a second multi-dimensional matrix, and the second multi-dimensional matrix is used to indicate the communication strategies that can be implemented between each satellite, and the communication strategies include: communicating through a microwave link and communicating through a laser link; The first graph data construction module is used to construct first graph data according to the second multi-dimensional matrix, where the nodes of the first graph data respectively correspond to each satellite, the node attribute information is used to indicate the data transmission volume and data reception volume of the cells corresponding to each satellite, and the edges of the first graph data are used to indicate that communication can be carried out between two satellites, and the edge attribute is used to indicate the data transmission volume between two satellites and the communication strategies that can be implemented; The message passing module is used to perform message passing on the first graph data by using a pre-trained graph neural network, so as to determine the corresponding second graph data; and The communication strategy scheduling module is used to uniquely determine the communication strategies for one-way communication between each satellite according to the edge feature information of the one-way edges in the second graph data.

10. A satellite communication scheduling device based on a dual-link, characterized in that, Comprising: A processor; And A memory, connected to the processor, for providing instructions for the processor to perform the following processing steps: Construct a first multi-dimensional matrix according to the position information of multiple satellites, where the first multi-dimensional matrix is used to indicate the positions between each satellite and the position environment information related to signal communication; Perform feature extraction on the first multi-dimensional matrix to determine a second multi-dimensional matrix, where the second multi-dimensional matrix is used to indicate the communication strategies that can be implemented between each satellite, and the communication strategies include: communication through a microwave link and communication through a laser link; Construct first graph data according to the second multi-dimensional matrix, where the nodes of the first graph data respectively correspond to each satellite, the node attribute information is used to indicate the data transmission volume and data reception volume of the cells corresponding to each satellite, and the edges of the first graph data are used to indicate that communication can be carried out between two satellites, and the edge attribute is used to indicate the data transmission volume between two satellites and the communication strategies that can be implemented; Perform message passing on the first graph data by using a pre-trained graph neural network, so as to determine the corresponding second graph data; and According to the edge feature information of the one-way edges in the second graph data, uniquely determine the communication strategies for one-way communication between each satellite.

Images