A satellite multi-beam cooperative scheduling method and device and a storage medium

Abstract

This application discloses a satellite multi-beam collaborative scheduling method, apparatus, and storage medium. The method involves determining the initial number of beam positions for each beam based on service-related data. Then, based on the service-related data, the initial number of beam positions, and historical beam position migration data, graph data representing the adjacency relationships between beams is generated. Furthermore, based on the graph data, node features corresponding to each beam are generated. These node features are then sorted according to the determined priority of each beam, generating an ordered feature sequence. Based on this ordered feature sequence, a unidirectional LSTM model is used to determine the final number of beam positions (target number of beam positions) for each beam, and beam position scheduling is performed. Therefore, this method can determine the appropriate number of beam positions for each beam by considering the service characteristics of the service area served by the beam, thereby improving the utilization rate of satellite beam resources.

Description

Technical Field

[0001] This application relates to the field of beam position scheduling technology, and in particular to a satellite multi-beam collaborative scheduling method, device and storage medium. Background Technology

[0002] Low-Earth orbit (LEO) satellite communication systems, with their advantages of low latency and wide coverage, have become one of the core supporting technologies for achieving seamless global communication. Among them, multi-beam technology is a key means to improve satellite communication capacity and adapt to regional service differences. However, in existing technologies, the coverage range of each beam within a single satellite usually adopts a fixed design mode, that is, the number of beam positions and the coverage range of each beam position remain unchanged after the satellite is deployed.

[0003] However, in practical applications, the service conditions in different locations within the area served by the satellite may vary. That is, there may be two types of areas in the area served by the satellite: densely populated areas and sparsely populated areas. With existing technologies, beam resources in densely populated areas will be strained and transmission efficiency will be low, while beam resources in sparsely populated areas will be idle and underutilized.

[0004] There is currently no effective solution to the technical problems of scarce beam resources and low transmission efficiency in densely populated areas, and idle and underutilized beam resources in sparsely populated areas in the existing technologies. Summary of the Invention

[0005] The embodiments of this disclosure provide a satellite multi-beam collaborative scheduling method, apparatus, and storage medium to at least solve the technical problems of scarce beam resources in densely populated service areas, low transmission efficiency, and idle beam resources in sparsely populated service areas with insufficient utilization in the prior art.

[0006] According to one aspect of the embodiments of this disclosure, a satellite multi-beam cooperative scheduling method is provided, comprising: determining service-related data and historical position migration data for each satellite beam; determining the initial number of positions corresponding to each beam based on the service-related data and a position quantity evaluation model; generating graph data based on the service-related data, the initial number of positions, and the historical position migration data; performing message passing based on the graph data to generate node features corresponding to each beam; generating a priority corresponding to each beam and sorting the node features corresponding to each beam according to the priority to generate an ordered feature sequence; determining the target number of positions corresponding to each beam based on a unidirectional LSTM model and the ordered feature sequence; and performing position scheduling for the beams according to the target number of positions.

[0007] According to another aspect of the present disclosure, a storage medium is also provided, the storage medium including a stored program, wherein the methods described above are executed by a processor when the program is running.

[0008] According to another aspect of the present disclosure, a satellite multi-beam cooperative scheduling device is also provided, comprising: a determination module, configured to determine service-related data and historical position migration data for each satellite beam; an initial evaluation module, configured to determine the initial number of positions corresponding to each beam based on the service-related data and a position quantity evaluation model; a graph data module, configured to generate graph data based on the service-related data, the initial number of positions, and the historical position migration data; a message transmission module, configured to perform message transmission based on the graph data and generate node features corresponding to each beam; a sorting module, configured to generate a priority corresponding to each beam and sort the node features corresponding to each beam according to the priority to generate an ordered feature sequence; a position quantity determination module, configured to determine the target number of positions corresponding to each beam based on a unidirectional LSTM model and the ordered feature sequence; and a scheduling module, configured to perform position scheduling of the beams according to the target number of positions.

[0009] According to another aspect of the present disclosure, a satellite multi-beam cooperative scheduling device is also provided, comprising: a processor; and a memory connected to the processor, configured to provide the processor with instructions for processing the following steps: determining service-related data and historical position migration data for each satellite beam; determining the initial number of positions corresponding to each beam based on the service-related data and a position quantity evaluation model; generating graph data based on the service-related data, the initial number of positions, and the historical position migration data; performing message passing based on the graph data to generate node features corresponding to each beam; generating a priority corresponding to each beam, and sorting the node features corresponding to each beam according to the priority to generate an ordered feature sequence; determining the target number of positions corresponding to each beam based on a unidirectional LSTM model and the ordered feature sequence; and performing position scheduling for the beams according to the target number of positions.

[0010] In this embodiment, the initial number of beam positions for each beam is determined based on the service-related data of each beam. Then, based on the service-related data, the initial number of beam positions, and historical beam position migration data, graph data representing the adjacency relationships between beams is generated. Furthermore, based on the graph data, node features corresponding to each beam are generated. According to the determined priority corresponding to each beam, the node features corresponding to each beam are sorted according to the priority to generate an ordered feature sequence. Based on the ordered feature sequence, the final number of beam positions (target number of beam positions) corresponding to each beam is determined through a unidirectional LSTM model, and beam position scheduling is performed. Thus, this method can determine the appropriate number of beam positions for each beam by combining the service characteristics of the service area served by the beam, thereby improving the utilization rate of satellite beam resources. Attached Figure Description

[0011] The accompanying drawings, which are included to provide a further understanding of this disclosure and form part of this application, illustrate exemplary embodiments of this disclosure and are used to explain this disclosure, but do not constitute an undue limitation of this disclosure. In the drawings: Figure 1 This is a schematic diagram of the hardware architecture of the satellite according to Embodiment 1 of this disclosure; Figure 2 This is a schematic flowchart of the satellite multi-beam cooperative scheduling method according to the first aspect of Embodiment 1 of this disclosure; Figure 3 This is a schematic diagram of the process for generating the allocation ratio of wave position resources according to the first aspect of Embodiment 1 of this disclosure; Figure 4 This is a schematic diagram of the process for generating graph data according to the first aspect of Embodiment 1 of this disclosure; Figure 5 This is a schematic diagram of a satellite multi-beam coordinated scheduling device according to the first aspect of Embodiment 2 of this disclosure; Figure 6 This is a schematic diagram of a satellite multi-beam coordinated scheduling device according to the first aspect of Embodiment 3 of this disclosure. Detailed Implementation

[0012] To enable those skilled in the art to better understand the technical solutions of this disclosure, the technical solutions of the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this disclosure, and not all embodiments. Based on the embodiments of this disclosure, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this disclosure.

[0013] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this disclosure are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this disclosure described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0014] Example 1

[0015] According to this embodiment, a satellite multi-beam cooperative scheduling method and apparatus are provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0016] Figure 1 A schematic diagram of the satellite's hardware architecture is shown. (Reference) Figure 1 As shown, the satellite includes an integrated electronic system, which comprises a processor, a memory, a bus management module, and a communication interface. The memory is connected to the processor, allowing the processor to access the memory, read program instructions stored in the memory, and read or write data to the memory. The bus management module is connected to the processor and also to a bus such as a CAN bus. Thus, the processor can communicate with onboard peripherals connected to the bus through the bus managed by the bus management module. Furthermore, the processor also communicates with devices such as cameras, star sensors, telemetry and command transponders, and data transmission equipment via the communication interface. Those skilled in the art will understand that… Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the aforementioned electronic device. For example, a satellite may also include components that are larger than... Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown.

[0017] It is worth noting that the aforementioned spaceborne peripherals connected to the CAN bus can be one or more. These spaceborne peripherals include, but are not limited to, GNSS modules, fiber optic gyroscopes, and high-torque flywheels. Further details will not be elaborated upon here.

[0018] It should be noted that, Figure 1One or more processors and / or other data processing circuits shown herein may generally be referred to as "data processing circuitry". This data processing circuitry may be embodied, in whole or in part, in software, hardware, firmware, or any other combination thereof. Furthermore, the data processing circuitry may be a single, independent processing module, or may be integrated, in whole or in part, into any other element in a computing device. As involved in embodiments of this disclosure, the data processing circuitry serves as processor control (e.g., selection of a variable resistor termination path connected to an interface).

[0019] Figure 1 The memory shown can be used to store software programs and modules of application software, such as the program instruction / data storage device corresponding to the satellite multi-beam cooperative scheduling method in this embodiment of the present disclosure. The processor executes various functional applications and data processing by running the software programs and modules stored in the memory, thereby realizing the satellite multi-beam cooperative scheduling method of the above-mentioned application. The memory may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory.

[0020] It should be noted here that, in some optional embodiments, the above... Figure 1 The device shown may include hardware elements (including circuitry), software elements (including computer code stored on a computer-readable medium), or a combination of both hardware and software elements. It should be noted that... Figure 1 This is only one instance of a specific particular instance, and is intended to illustrate the types of components that may exist in the aforementioned devices.

[0021] Under the aforementioned operating environment, according to the first aspect of this embodiment, a satellite multi-beam cooperative scheduling method is provided, which can be... Figure 1 The terminal implementation shown. Figure 2 A flowchart illustrating the method is shown below. (Refer to...) Figure 2 As shown, the method includes: S202: Determine the service-related data and historical position migration data for each satellite beam; S204: Based on business-related data and the wavelet number evaluation model, determine the initial number of wavelets corresponding to each beam; S206: Generate graph data based on business-related data, initial wave position count, and historical wave position migration data; S208: Message passing is based on graph data to generate node features corresponding to each beam; S210: Generate a priority corresponding to each beam, and sort the node features corresponding to each beam according to the priority to generate an ordered feature sequence; S212: Based on a unidirectional LSTM model, the number of target wave positions corresponding to each beam is determined according to the ordered feature sequence; and S214: Perform beam position scheduling based on the number of target beam positions.

[0022] First, the satellite can determine the service-related data for each beam and the historical beam position migration data (S202).

[0023] Among these, service-related data can be used to represent the attribute information of the service area corresponding to the corresponding beam in terms of communication services. For example, service-related data can indicate whether the corresponding service area is a service-intensive area or a service-sparse area.

[0024] Historical position migration data is used to represent the number of position migrations between adjacent beams.

[0025] For example, suppose the satellite in this embodiment has two beams, totaling 10 positions. At time t, beam a has 4 positions and beam b has 6 positions. At time t+1, to balance ground services, the number of positions for beam a is changed to 6, and the number of positions for beam b is changed to 4. Then, for the time from t to t+1, the historical position migration data from beam a to beam b is +2, and the historical position migration data from beam b to beam a is -2.

[0026] Then, the satellite can determine the initial number of beams corresponding to each beam based on the beam number evaluation model, according to the service-related data (S204).

[0027] refer to Figure 3 As shown, satellites can input service-related data into the preamplitude number assessment model to determine the initial preamplitude number corresponding to each beam. The initial preamplitude number can be used to construct map data.

[0028] Then, the satellite generates map data based on operational data, the initial number of positions, and historical position migration data (S206).

[0029] The graph data is used to represent the adjacency relationships between beams, the service attributes of the beams themselves, and the changes in the number of beam positions between beams throughout history.

[0030] Then, the satellite transmits messages based on the graph data and generates node features corresponding to each beam (S208).

[0031] After generating the node features corresponding to each beam, the satellite generates a priority corresponding to each beam and sorts the node features corresponding to each beam according to the priority to generate an ordered feature sequence (S210).

[0032] The aforementioned priorities can be determined based on the beam's service-related data, and the priority indicates the beam's importance. For example, the denser the service area served by the beam, the higher its priority can be.

[0033] Then, based on the one-way LSTM model, the satellite determines the number of target positions corresponding to each beam according to the ordered feature sequence (S212), and performs beam position scheduling according to the number of target positions (S214).

[0034] In the steps described above, the node features of each beam are sorted according to beam priority to generate an ordered feature sequence, which carries information about the priority of each beam. Then, the satellite uses a one-way LSTM to determine the number of target beams based on the ordered feature sequence. Thus, this method combines the characteristic of the one-way LSTM model that it can predict based on the order of the input sequence (i.e., the ordered feature sequence), ensuring that the number of generated target beams conforms to the priority of each beam.

[0035] Once the target number of wave positions is determined, wave position scheduling can be performed based on this number. Specifically, the number of wave positions for a given beam can be adjusted to match the target number of wave positions for that beam. The specific wave position adjustment method will be explained later.

[0036] As described in the background section, in practical applications, the service conditions in different locations within the area served by the satellite may vary. That is, there may be both densely populated and sparsely populated areas within the satellite's service area. In this case, the existing technology may easily lead to a shortage of beam resources and low transmission efficiency in densely populated areas, while beam resources may be idle and underutilized in sparsely populated areas.

[0037] In view of this, in this embodiment, the initial number of positions for each beam is determined based on the service-related data of each beam. Then, based on the service-related data, the initial number of positions, and historical position migration data, graph data representing the adjacency relationships between beams is generated. Furthermore, based on the graph data, node features corresponding to each beam are generated. According to the determined priority corresponding to each beam, the node features corresponding to each beam are sorted according to the priority to generate an ordered feature sequence. Based on the ordered feature sequence, the final number of positions (target number of positions) corresponding to each beam is determined through a unidirectional LSTM model, and position scheduling is performed. Thus, this method can determine the appropriate number of positions for each beam by combining the service characteristics of the service area served by the beam, thereby improving the utilization rate of satellite beam resources.

[0038] Optionally, the business-related data includes at least one of the following: user density information, business load information, and latency constraint information.

[0039] Among them, user density information is used to quantify the user density in the service area corresponding to the beam. Service load information refers to the communication service load of each user in the service area corresponding to the beam. Delay constraint information represents the delay constraints set by relevant personnel for the service area corresponding to the beam (i.e., the strictness of the delay requirements). It can be seen that service-related data mainly represents attributes related to communication services in the service area corresponding to the beam. Therefore, this method uses service-related data in determining the initial number of beam positions, generating map data, and determining the priority corresponding to the beam, thus enabling the adjustment of beam positions for each beam based on the actual situation of all satellite service areas.

[0040] Optionally, the operation of generating graph data based on business-related data, the initial number of beams, and historical beam migration data includes: obtaining a graph structure, wherein nodes in the graph structure represent corresponding beams, and edges in the graph structure represent adjacency relationships between beams; generating the graph data based on the graph structure according to the business-related data, the initial number of beams, and the historical beam migration data, wherein the node attribute information of the nodes in the graph data includes the business-related data and the initial number of beams, and the edge attribute information of the edges in the graph data includes the historical beam migration data.

[0041] Specifically, firstly, satellites can acquire graph structures. Since the number of satellite beams is fixed, the graph structure can be pre-constructed. (Reference) Figure 4 As shown, a node in the graph structure represents a corresponding beam, and the edges between beams represent their adjacency. Then, graph data can be generated based on the graph structure, using business-related data, the initial number of beams, and historical beam migration data. Figure 4 The diagram shows a satellite with four beams, so the diagram structure contains four nodes, corresponding to the satellite's beams a, b, c, and d, respectively.

[0042] The graph data includes node attribute information such as business-related data and the initial number of beam positions, and edge attribute information such as historical beam position migration data. In other words, node attribute information is related to the beam itself, while edge attribute information is related to beam position adjustments between beams. Figure 4 As can be seen, the process of generating graph data through a graph structure is the process of generating node attribute information corresponding to nodes and edge attribute information corresponding to edges.

[0043] Optionally, nodes in the graph structure are connected by directed edges, the direction of which is used to indicate the wave position migration direction.

[0044] It should be noted that in this embodiment, the edges between nodes are directed edges, meaning that for any two nodes, there are two edges connecting them, and the two edges are in opposite directions, such as... Figure 4 As shown, there are two edges connecting the node corresponding to beam a and the node corresponding to beam b. One edge points from the node corresponding to beam a to the node corresponding to beam b. The edge attribute information of this edge represents the number of positions migrated from beam a to beam b (historical position migration data). The other edge points from the node corresponding to beam b to the node corresponding to beam a. The edge attribute information of this edge represents the number of positions migrated from beam b to beam a (historical position migration data).

[0045] Optionally, the operation of message passing based on graph data to generate node features corresponding to each beam includes: for each node, generating initial node features based on node attribute information corresponding to the corresponding node, and for each edge, generating initial edge features based on edge attribute information corresponding to the corresponding edge; in one round of message passing, updating the edge features of each edge based on the node features of the two nodes connected to the corresponding edge, and updating the node features of each node based on the node features of the neighboring nodes of the corresponding node and the edge features of the edges connected to the corresponding node; and generating node features corresponding to each beam by repeatedly performing message passing for a preset number of rounds.

[0046] Specifically, in this embodiment, a Graph Neural Network (GNN) can be used for message passing based on graph data. First, initial node features corresponding to each node can be generated based on node attribute information, and edge features corresponding to each edge can be generated based on edge attribute information.

[0047] In this embodiment, business-related data and historical wave position migration data can be converted into vector form to obtain a business quantification factor vector corresponding to the business-related data and a historical wave position migration vector corresponding to the historical wave position migration data. The business quantification factor vector is obtained by normalizing the business-related data, with each element in the business quantification factor vector normalized to a number between [0, 1]. The business quantification factor vector can be represented as: X = [x1, x2, ..., x...]. n The historical wave position migration vector can be represented as: P = [p1, p2, ..., p...]. k ]. x1~x n To normalize the values ​​of each dimension in the business-related data, p1~p kLet V be the number of wave positions migrated from time 1 to k. Specifically, the business quantization factor vector X corresponding to the corresponding node and the initial number of wave positions s can be integrated into a vector V=[X, s], which serves as the initial node feature. The historical wave position migration vector P corresponding to the corresponding edge is used as the initial edge feature T=P. For node i, the initial node feature is expressed as... For the edge from node i to node j In other words, with that side The corresponding initial edge features are represented as follows .

[0048] Then, by repeatedly performing a preset number of message passes, node features corresponding to each beam can be generated. That is, the node features corresponding to each beam obtained from the last round of message passing are used as the final node features for determining the number of target beams.

[0049] Specifically, in a round of message passing, the satellite updates the node features of each node and the edge features of each edge. Specifically, the node features are updated using the following formula (1), and the edge features are updated using the following formula (2).

[0050] (1) (2) in, This represents the updated node characteristics of node i in this round of message passing. This represents the node characteristics of node i before its update in this round of message passing. and This is the weight matrix. For activation function, This represents the summation of the node features and edge features of all neighboring nodes of node i, thus achieving the aggregation of neighborhood information. Activation functions are used to mitigate gradient vanishing. This indicates that in this round of message transmission, the side Updated edge features This indicates that in this round of message transmission, the side Edge features before update and This is the weight matrix. and They are the edges The node characteristics of the source node and the node characteristics of the target node.

[0051] By repeatedly transmitting messages in pre-set cycles, satellites can generate node features corresponding to each beam.

[0052] Optionally, the operation of generating a priority corresponding to each beam includes: for each beam, inputting the service-related data of the corresponding beam into a priority ranking model, and generating a priority corresponding to the corresponding beam based on the weights corresponding to different service-related data through the priority ranking model.

[0053] For details, please refer to Figure 3 As shown, before sorting the node features generated by GNN, the satellite can input the service-related data of the corresponding beam into the priority ranking model to generate the priority corresponding to the corresponding beam. The priority ranking model can be a linear model. The priority corresponding to the corresponding beam can be generated using the following formula (3).

[0054] (3) This represents the priority and weight corresponding to the i-th beam. ~ The sum of these is 1. Where, the weights... ~ It can be manually set based on the importance of different dimensions of data in business-related data; of course, the weights... ~ It can also be a learnable parameter. The values ​​of each dimension in the business quantization factor vector corresponding to the i-th beam correspond to the data of each dimension in the business-related data.

[0055] The beam priority determined by the above formula (3) can indicate the importance of the corresponding beam. For example, the denser the service area served by the beam, the higher the beam priority may be.

[0056] After determining the priority corresponding to each beam, the node features of each beam can be sorted according to the priority. That is, the node features of each beam are sorted from high to low priority to generate an ordered feature sequence.

[0057] Optionally, the operation of determining the number of target saturations corresponding to each beam based on the ordered feature sequence using a unidirectional LSTM model includes: inputting the ordered feature sequence into the unidirectional LSTM model to generate a temporal feature vector; generating a saturation resource allocation ratio corresponding to each beam based on the temporal feature vector; and determining the number of target saturations corresponding to each beam based on the saturation resource allocation ratio and the total number of saturations of the satellite.

[0058] Specifically, satellites can input ordered feature sequences into a unidirectional LSTM to generate temporal feature vectors. The reason for using a unidirectional LSTM is that the ordered feature sequences have already been generated according to priority. The unidirectional LSTM can retain the feature memory of higher-priority beams, which can be used to constrain subsequent lower-priority beams. This achieves a collaborative logic of prioritizing high-priority beams and adaptively yielding to low-priority beams, preventing low-priority beams from crowding out the resources of high-priority beams.

[0059] Then, the satellite can generate the preamplitude allocation ratio for each beam based on the temporal feature vector, and then determine the preamplitude allocation ratio based on the total number of preamplitudes of the satellite. The number of target positions corresponding to each beam is determined by multiplication. This involves inputting the temporal feature vector into a fully connected layer to generate a corresponding resource allocation feature vector. This resource allocation feature vector is then input into a Softmax function to output the percentage of target positions allocated to each beam. m represents the number of beams.

[0060] Finally, the number of target wave positions is determined using the following formula.

[0061] (4) in, This is a rounding function. This represents the number of target positions corresponding to the i-th beam.

[0062] Therefore, the method of determining the number of target wavelengths based on the proportion of wavelength resources allocated and the total number of wavelengths of the satellite can ensure that the sum of the number of target wavelengths determined for each wavelength is equal to the total number of wavelengths of the satellite.

[0063] After determining the number of target positions corresponding to each beam, the satellite can perform position scheduling based on this number. For example, in this embodiment, the service area of ​​each beam can be fixed, and the total number of positions is also fixed. After determining the number of target positions, the number of positions for the corresponding beam is directly adjusted to the target number. As another example, the satellite can be configured with a fixed total number of positions and a fixed service area for each position. After determining the number of target positions corresponding to each beam, based on the current number of positions for each beam, if the current number of positions is less than the target number, positions from adjacent beams are allocated to the corresponding beam; if the current number of positions is more than the target number, the excess positions are allocated to adjacent beams. The allocation is completed once the satellite determines that the number of positions for each beam meets the target number.

[0064] Alternatively, the satellite's various positions can be set to fixed values. That is, the total number of satellite positions is fixed, the ground area corresponding to each position is also fixed, and each beam corresponds to a subset of positions. After determining the target number of positions for each beam, the satellite performs position migration between beams according to the target number of positions. Specifically, following the example above, the satellite divides the positions of beams with an original number of positions less than the target number of positions from adjacent beams according to the target number of positions. The goal of this division is to ensure that the number of positions for each beam meets the target number of positions. Once this goal is achieved, the division process ends.

[0065] It should be noted that in this embodiment, the beam number evaluation model, GNN, priority ranking model, and the network layer used to generate the beam resource allocation ratio can be pre-trained jointly (specifically, supervised training). Specifically, a dataset representing the adjustment of beam positions for each beam can be pre-constructed. This dataset includes service-related data samples for each beam, historical beam position migration data samples, and annotation information. The annotation information represents the result after adjusting the beam positions for each beam. Supervised training is then performed using this dataset.

[0066] In addition, refer to Figure 1 As shown, according to a second aspect of this embodiment, a storage medium is provided. The storage medium includes a stored program, wherein the methods described above are executed by a processor when the program is running.

[0067] Therefore, according to this embodiment, the initial number of beam positions for each beam is determined based on the service-related data of each beam. Then, based on the service-related data, the initial number of beam positions, and historical beam position migration data, graph data representing the adjacency relationships between beams is generated. Furthermore, based on the graph data, node features corresponding to each beam are generated, and the node features corresponding to each beam are sorted according to the determined priority, generating an ordered feature sequence. Based on the ordered feature sequence, the final number of beam positions (target number of beam positions) corresponding to each beam is determined through a unidirectional LSTM model, and beam position scheduling is performed. Thus, this method can determine the appropriate number of beam positions for each beam by combining the service characteristics of the service area served by the beam, thereby improving the utilization rate of satellite beam resources.

[0068] Furthermore, this embodiment models the spatial adjacency relationships and historical beam position migration patterns between beams using GNN, combined with LSTM timing constraints. This enables adjacent beams to allocate resources based on changes in their respective services and historical scheduling patterns, improving the stability and reliability of data transmission. Moreover, by introducing a priority evaluation mechanism, beam priorities are determined based on service-related data (such as latency constraints and user density). Unidirectional LSTM ensures that high-priority beams receive priority resource guarantees, preventing low-priority services from crowding out high-priority service resources. This is particularly suitable for latency-sensitive services such as emergency communication and real-time data transmission, improving service quality and user experience.

[0069] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that the present invention is not limited to the described order of actions, because according to the present invention, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to the present invention.

[0070] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this 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 disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0071] Example 2

[0072] Figure 5 A satellite multi-beam cooperative scheduling apparatus according to the first aspect of this embodiment is shown, which corresponds to the method according to the first aspect of Embodiment 1. (Reference) Figure 5As shown, the device includes: a determination module 510, used to determine the service-related data and historical position migration data of each satellite beam; an initial evaluation module 520, used to determine the initial number of positions corresponding to each beam based on the service-related data and a position quantity evaluation model; a graph data module 530, used to generate graph data based on the service-related data, the initial number of positions, and the historical position migration data; a message transmission module 540, used to perform message transmission based on the graph data and generate node features corresponding to each beam; a sorting module 550, used to generate a priority corresponding to each beam and sort the node features corresponding to each beam according to the priority to generate an ordered feature sequence; a position quantity determination module 560, used to determine the target number of positions corresponding to each beam based on a unidirectional LSTM model and the ordered feature sequence; and a scheduling module 570, used to perform position scheduling between beams according to the target number of positions.

[0073] Optionally, the business-related data includes at least one of the following: user density information, business load information, and latency constraint information.

[0074] Optionally, the graph data module 530 is used to obtain a graph structure, where nodes in the graph structure represent corresponding beams, and edges in the graph structure represent the adjacency relationships between beams; based on the graph structure, graph data is generated according to business-related data, the initial number of beams, and historical beam migration data. The node attribute information of the nodes in the graph data includes business-related data and the initial number of beams, and the edge attribute information of the edges in the graph data includes historical beam migration data.

[0075] Optionally, nodes in the graph structure are connected by directed edges, the direction of which is used to indicate the wave position migration direction.

[0076] Optionally, the message passing module 540 is used to generate initial node features for each node based on the node attribute information corresponding to the corresponding node, and to generate initial edge features for each edge based on the edge attribute information corresponding to the corresponding edge; in one round of message passing, for each edge, the edge features of the corresponding edge are updated based on the node features of the two nodes connected to the corresponding edge, and for each node, the node features of the corresponding node are updated based on the node features of the neighboring nodes of the corresponding node and the edge features of the edges connected to the corresponding node; by repeatedly performing message passing for a preset number of rounds, node features corresponding to each beam are generated.

[0077] Optionally, the sorting module 550 is used to input the service-related data of each beam into the priority sorting model, and generate the priority corresponding to the beam based on the weights corresponding to different service-related data through the priority sorting model.

[0078] Optionally, the wavelet number determination module 560 is used to input an ordered feature sequence into a unidirectional LSTM model to generate a temporal feature vector; generate a wavelet resource allocation ratio corresponding to each beam based on the temporal feature vector; and determine the target wavelet number corresponding to each beam based on the wavelet resource allocation ratio and the total number of wavelets for the satellite.

[0079] Therefore, according to this embodiment, the initial number of beam positions for each beam is determined based on the service-related data of each beam. Then, based on the service-related data, the initial number of beam positions, and historical beam position migration data, graph data representing the adjacency relationships between beams is generated. Furthermore, based on the graph data, node features corresponding to each beam are generated, and the node features corresponding to each beam are sorted according to the determined priority, generating an ordered feature sequence. Based on the ordered feature sequence, the final number of beam positions (target number of beam positions) corresponding to each beam is determined through a unidirectional LSTM model, and beam position scheduling is performed. Thus, this method can determine the appropriate number of beam positions for each beam by combining the service characteristics of the service area served by the beam, thereby improving the utilization rate of satellite beam resources.

[0080] Furthermore, this embodiment models the spatial adjacency relationships and historical beam position migration patterns between beams using GNN, combined with LSTM timing constraints. This enables adjacent beams to allocate resources based on changes in their respective services and historical scheduling patterns, improving the stability and reliability of data transmission. Moreover, by introducing a priority evaluation mechanism, beam priorities are determined based on service-related data (such as latency constraints and user density). Unidirectional LSTM ensures that high-priority beams receive priority resource guarantees, preventing low-priority services from crowding out high-priority service resources. This is particularly suitable for latency-sensitive services such as emergency communication and real-time data transmission, improving service quality and user experience.

[0081] Example 3

[0082] Figure 6 A satellite multi-beam cooperative scheduling apparatus according to a first aspect of this embodiment is shown, which corresponds to the method according to a first aspect of Embodiment 1. (Reference) Figure 6As shown, the device includes: a processor 610; and a memory 620 connected to the processor 610, used to provide the processor 610 with instructions to process the following steps: determining service-related data and historical position migration data for each satellite beam; determining the initial number of positions corresponding to each beam based on the service-related data and a position quantity evaluation model; generating graph data based on the service-related data, the initial number of positions, and the historical position migration data; performing message passing based on the graph data to generate node features corresponding to each beam; generating priorities corresponding to each beam and sorting the node features corresponding to each beam according to the priorities to generate an ordered feature sequence; determining the target number of positions corresponding to each beam based on a unidirectional LSTM model and the ordered feature sequence; and performing position scheduling for the beams according to the target number of positions.

[0083] Optionally, the business-related data includes at least one of the following: user density information, business load information, and latency constraint information.

[0084] Optionally, the operation of generating graph data based on business-related data, the initial number of beams, and historical beam migration data includes: obtaining a graph structure, where nodes in the graph structure represent corresponding beams, and edges in the graph structure represent the adjacency relationships between beams; and generating graph data based on the graph structure according to the business-related data, the initial number of beams, and historical beam migration data, wherein the node attribute information of the nodes in the graph data includes business-related data and the initial number of beams, and the edge attribute information of the edges in the graph data includes historical beam migration data.

[0085] Optionally, nodes in the graph structure are connected by directed edges, the direction of which is used to indicate the wave position migration direction.

[0086] Optionally, the operation of message passing based on graph data to generate node features corresponding to each beam includes: for each node, generating initial node features based on node attribute information corresponding to the corresponding node, and for each edge, generating initial edge features based on edge attribute information corresponding to the corresponding edge; in one round of message passing, updating the edge features of each edge based on the node features of the two nodes connected to the corresponding edge, and updating the node features of each node based on the node features of the neighboring nodes of the corresponding node and the edge features of the edges connected to the corresponding node; and generating node features corresponding to each beam by repeatedly performing message passing for a preset number of rounds.

[0087] Optionally, the operation of generating a priority corresponding to each beam includes: for each beam, inputting the service-related data of the corresponding beam into a priority ranking model, and generating a priority corresponding to the corresponding beam based on the weights corresponding to different service-related data through the priority ranking model.

[0088] Optionally, the operation of determining the number of target saturations corresponding to each beam based on the ordered feature sequence using a unidirectional LSTM model includes: inputting the ordered feature sequence into the unidirectional LSTM model to generate a temporal feature vector; generating the saturation resource allocation ratio corresponding to each beam based on the temporal feature vector; and determining the number of target saturations corresponding to each beam based on the saturation resource allocation ratio and the total number of saturations of the satellite.

[0089] Therefore, according to this embodiment, the initial number of beam positions for each beam is determined based on the service-related data of each beam. Then, based on the service-related data, the initial number of beam positions, and historical beam position migration data, graph data representing the adjacency relationships between beams is generated. Furthermore, based on the graph data, node features corresponding to each beam are generated, and the node features corresponding to each beam are sorted according to the determined priority, generating an ordered feature sequence. Based on the ordered feature sequence, the final number of beam positions (target number of beam positions) corresponding to each beam is determined through a unidirectional LSTM model, and beam position scheduling is performed. Thus, this method can determine the appropriate number of beam positions for each beam by combining the service characteristics of the service area served by the beam, thereby improving the utilization rate of satellite beam resources.

[0090] Furthermore, this embodiment models the spatial adjacency relationships and historical beam position migration patterns between beams using GNN, combined with LSTM timing constraints. This enables adjacent beams to allocate resources based on changes in their respective services and historical scheduling patterns, improving the stability and reliability of data transmission. Moreover, by introducing a priority evaluation mechanism, beam priorities are determined based on service-related data (such as latency constraints and user density). Unidirectional LSTM ensures that high-priority beams receive priority resource guarantees, preventing low-priority services from crowding out high-priority service resources. This is particularly suitable for latency-sensitive services such as emergency communication and real-time data transmission, improving service quality and user experience.

[0091] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0092] In the above embodiments of the present invention, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0093] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0094] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0095] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0096] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the 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 to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0097] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method of satellite multi-beam coordinated scheduling, the method comprising: include: Determine the service-related data and historical position migration data for each satellite beam; Based on the aforementioned business-related data and the wavelet number evaluation model, the initial number of wavelets corresponding to each beam is determined. Based on the business-related data, the initial number of wave positions, and the historical wave position migration data, graph data is generated; Based on the graph data, message transmission is performed to generate node features corresponding to each beam; A priority corresponding to each beam is generated, and the node features corresponding to each beam are sorted according to the priority to generate an ordered feature sequence; Based on the unidirectional LSTM model, the number of target wave positions corresponding to each beam is determined according to the ordered feature sequence. Based on the number of target wave positions, beam position scheduling is performed.

2. The method according to claim 1, characterized in that, The business-related data includes at least one of the following: user density information, business load information, and latency constraint information.

3. The method according to claim 1, characterized in that, The operation of generating graph data based on the business-related data, the initial number of wave positions, and the historical wave position migration data includes: Obtain a graph structure, where nodes in the graph structure represent corresponding beams, and edges in the graph structure represent the adjacency relationships between beams; Based on the business-related data, the initial number of wavelets, and the historical wavelet migration data, the graph data is generated according to the graph structure. The node attribute information of the nodes in the graph data includes the business-related data and the initial number of wavelets, and the edge attribute information of the edges in the graph data includes the historical wavelet migration data.

4. The method according to claim 3, characterized in that, In the graph structure, nodes are connected by directed edges, and the direction of the directed edges is used to indicate the wave position migration direction.

5. The method according to claim 3, characterized in that, The operation of message passing based on the graph data and generating node features corresponding to each beam includes: For each node, initial node features are generated based on the node attribute information corresponding to the corresponding node; and for each edge, initial edge features are generated based on the edge attribute information corresponding to the corresponding edge. In one round of message passing, for each edge, the edge characteristics of the corresponding edge are updated based on the node characteristics of the two nodes connected to the corresponding edge, and for each node, the node characteristics of the corresponding node are updated based on the node characteristics of the neighboring nodes of the corresponding node and the edge characteristics of the edges connected to the corresponding node. By repeatedly transmitting messages in a preset number of rounds, node features corresponding to each beam are generated.

6. The method according to claim 1, characterized in that, The operation of generating a priority corresponding to each beam includes: For each beam, the service-related data of the corresponding beam is input into the priority ranking model. The priority ranking model generates the priority corresponding to the corresponding beam based on the weights corresponding to different service-related data.

7. The method according to claim 1, characterized in that, Based on the unidirectional LSTM model, the operation of determining the number of target wave positions corresponding to each beam according to the ordered feature sequence includes: The ordered feature sequence is input into the unidirectional LSTM model to generate a temporal feature vector. Based on the time-series feature vector, generate the wave position resource allocation ratio corresponding to each beam; The number of target wavelengths corresponding to each beam is determined based on the wavelength resource allocation ratio and the total number of wavelengths of the satellite.

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

9. A satellite multi-beam collaborative scheduling device, characterized in that, include: The determination module is used to determine the service-related data and historical position migration data of each satellite beam; The initial evaluation module is used to determine the initial number of beams corresponding to each beam based on the business-related data and the beam number evaluation model. The graph data module is used to generate graph data based on the business-related data, the initial number of wave positions, and the historical wave position migration data. The message passing module is used to pass messages based on the graph data and generate node features corresponding to each beam. The sorting module is used to generate a priority corresponding to each beam, and sort the node features corresponding to each beam according to the priority to generate an ordered feature sequence; The wave position number determination module is used to determine the number of target wave positions corresponding to each beam based on the ordered feature sequence using a unidirectional LSTM model. The scheduling module is used to schedule the beam positions according to the number of target beam positions.

10. A satellite multi-beam cooperative scheduling device, characterized in that, include: processor; as well as A memory, connected to the processor, for providing the processor with instructions to perform the following processing steps: Determine the service-related data and historical position migration data for each satellite beam; Based on the aforementioned business-related data and the wavelet number evaluation model, the initial number of wavelets corresponding to each beam is determined. Based on the business-related data, the initial number of wave positions, and the historical wave position migration data, graph data is generated; Based on the graph data, message transmission is performed to generate node features corresponding to each beam; A priority corresponding to each beam is generated, and the node features corresponding to each beam are sorted according to the priority to generate an ordered feature sequence; Based on the unidirectional LSTM model, the number of target wave positions corresponding to each beam is determined according to the ordered feature sequence. Based on the number of target wave positions, beam position scheduling is performed.

Images