Satellite network multi-task scheduling method, system, device and medium based on computation collaboration

Abstract

The application discloses a satellite network multi-task scheduling method and system based on computation and communication collaboration, equipment and medium, relates to the technical field of satellite network communication and calculation, and comprises the following steps: acquiring a task set to be scheduled; mapping the task processing capacity of a satellite node into corresponding data volume; constructing a space-time model for describing the time-varying topology of a satellite network and the communication resources, the calculation resources and the storage resources based on the mapped data volume of the satellite node; generating a joint optimization model corresponding to the computation strategy and the resource allocation strategy of the task set to be scheduled according to the space-time model; solving the joint optimization model to obtain a target scheduling scheme of the task set to be scheduled in the satellite network; and scheduling tasks in the satellite network according to the target scheduling scheme. The application realizes accurate allocation of satellite network resources and reasonable task scheduling, reduces the end-to-end completion time of complex coupled tasks, and improves the task processing efficiency and the utilization rate of satellite network resources.

Description

Technical Field

[0001] This application relates to the field of satellite network communication and computing technology, and more specifically, to satellite network multi-task scheduling methods, systems, devices and media for transmission and computing collaboration. Background Technology

[0002] With the development of Low Earth Orbit (LEO) satellite constellations, satellite networks are evolving into space computing networks with space-based data processing and intelligent decision-making capabilities. In this model, satellites with sufficient computing power can preprocess, compress, extract features, or perform intelligent analysis on data before it is transmitted to ground stations via inter-satellite links (ISL). This reduces the amount of redundant data that needs to be transmitted to the ground, lowers the downlink load, and improves the end-to-end timeliness of missions from sensing to application.

[0003] In this mode, when performing multi-task scheduling on a satellite network, transmission paths are planned for each data stream on the time-spread graph of the satellite network topology, generating corresponding path sets and transmission scheduling schemes; computing nodes are selected and scheduled on the determined transmission paths. However, this approach does not consider the distribution of computing resources during the path selection phase, which may lead to insufficient computing power or excessive load on the nodes of the selected path, thus limiting the execution efficiency of on-orbit computing. Furthermore, the predetermined paths cannot be adjusted during the computing scheduling phase, making it difficult to fully utilize computing resources, resulting in high latency and low resource utilization in satellite network task processing.

[0004] In conclusion, improving the mission processing efficiency and resource utilization of satellite networks is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0005] The purpose of this application is to provide a telecomputing-based multi-task scheduling method for satellite networks, which can, to some extent, solve the technical problem of how to improve the task processing efficiency and resource utilization of satellite networks. This application also provides a telecomputing-based multi-task scheduling system for satellite networks, electronic equipment, and a computer-readable storage medium.

[0006] To achieve the above objectives, this application provides the following technical solution: A multi-task scheduling method for satellite networks with computational collaboration includes: Get the set of tasks to be scheduled; Map the link capabilities of satellite nodes to the corresponding amount of data that can be transmitted; map the sensing capabilities of satellite nodes to the corresponding amount of data that can be sensed; map the storage capabilities of satellite nodes to the corresponding amount of data that can be stored; map the computing capabilities of satellite nodes to the corresponding amount of data that can be computed. Based on the amount of data after satellite node mapping, a spatiotemporal model is constructed to describe the time-varying topology of the satellite network as well as its communication, computing, and storage resources. Based on the spatiotemporal model, a joint optimization model corresponding to the computational strategy and resource allocation strategy of the set of tasks to be scheduled is generated; Solving the joint optimization model yields the target scheduling scheme for the set of tasks to be scheduled in the satellite network; The satellite network is task-scheduled according to the target scheduling scheme.

[0007] Preferably, a spatiotemporal model is constructed to describe the time-varying topology of the satellite network, as well as its communication, computing, and storage resources, including: Determine the set time slice; In each time slice of the spatiotemporal extension graph, the satellite nodes of the satellite network are mapped to spatiotemporal nodes; Based on the transmission relationship of mission data between different satellite nodes, communication edges connecting corresponding spatiotemporal nodes within the same time slice are generated in the spatiotemporal extension graph. Based on the relationship between the processing of task data by the same satellite node between adjacent time slices, computational edges connecting corresponding spatiotemporal nodes between adjacent time slices are generated in the spatiotemporal extension graph. Based on the relationship of the same satellite node storing task data between adjacent time slices, storage edges connecting corresponding spatiotemporal nodes are generated in the spatiotemporal extension graph. Based on the mission data of satellite nodes in the time slice, information is generated, and in the spatiotemporal extension map, a perception edge connecting the corresponding spatiotemporal node and the task to be scheduled is generated within the same time slice. The spatiotemporal extension graph is used as a spatiotemporal model.

[0008] Preferably, based on the spatiotemporal model, a joint optimization model corresponding to the computational strategy and resource allocation strategy of the set of tasks to be scheduled is generated, including: Based on the spatiotemporal model, joint decision variables are generated for the computational strategy and resource allocation strategy of the set of tasks to be scheduled; Based on the joint decision variables, a global optimization objective is generated; Based on the spatiotemporal model, constraint information for the set of tasks to be scheduled is generated. The constraint information includes path constraints, link capacity constraints, computing power constraints, storage constraints, data flow conservation constraints, and task completion constraints. The joint decision variables, the global optimization objective, and the constraint information are used as a joint optimization model.

[0009] Preferably, the joint decision variables for generating the computational strategy and resource allocation strategy corresponding to the set of tasks to be scheduled include: A path selection variable is generated to form a set of tasks to be scheduled, wherein the path selection variable is used to characterize whether to select a link to transmit the tasks to be scheduled to a satellite node; A set of tasks to be scheduled is generated using computation node variables, which are used to characterize whether a network node performs on-orbit computation on the tasks to be scheduled. A transmission variable is generated for the set of tasks to be scheduled, the transmission variable being used to characterize the amount of data transmitted by the tasks to be scheduled through the link of the satellite node in the time slice; A set of computational resource variables is generated to represent the computational resources allocated by the satellite node to the tasks to be scheduled in the time slice. A completion indicator variable is generated for a set of tasks to be scheduled, wherein the completion indicator variable is used to characterize whether the tasks to be scheduled are completed within the time slice; The path selection variable, the computing node variable, the transmission variable, the computing resource variable, and the completion indication variable are used as joint decision variables.

[0010] Preferably, the data flow conservation constraints for generating the set of tasks to be scheduled include: Based on the data compression ratio of the task to be scheduled, determine the change in the amount of data of the task to be scheduled after processing by the satellite node; Based on the amount of data that a satellite node can transmit, sense, store, and compute, the amount of data consumed by the satellite node in each round of processing the scheduled task is constrained, and the amount of input data for the satellite node in the current round of processing the scheduled task is constrained to be the amount of output data after processing the scheduled task in the previous round, thereby generating information on the change in the amount of data processed by the satellite node for the scheduled task. Based on the data volume change information of satellite nodes and the data volume change information of the tasks to be scheduled, data flow conservation constraints of the tasks to be scheduled on satellite nodes are generated. Based on the data flow conservation constraints of the tasks to be scheduled on the satellite nodes, data flow conservation constraints are generated for the set of tasks to be scheduled.

[0011] Preferably, based on the data flow conservation constraints of the tasks to be scheduled on the satellite nodes, the data flow conservation constraints of the task set to be scheduled are generated, including: For any inter-satellite link, generate inter-satellite link constraint information. The inter-satellite link constraint information is used to constrain the amount of input data for the next satellite node to process the scheduled task to be equal to the amount of output data after the previous satellite node has processed the scheduled task. Based on the inter-satellite link constraint information, the data flow conservation constraints of the tasks to be scheduled at each satellite node of the inter-satellite link are integrated to generate the data flow conservation constraints of the set of tasks to be scheduled.

[0012] Preferably, the joint optimization model is solved to obtain the target scheduling scheme for the set of tasks to be scheduled in the satellite network, including: Solving the joint optimization model yields the target scheduling scheme for the set of tasks to be scheduled in the satellite network; The target scheduling scheme includes the optimal transmission path to the task to be scheduled in the satellite network, the satellite nodes for on-orbit computation of the task to be scheduled, the link bandwidth and computational resource allocation results in each time slice.

[0013] A satellite network multi-task scheduling system with computational and data transfer capabilities, comprising: The task acquisition module is used to acquire a set of tasks to be scheduled. The data volume mapping module is used to map the link capabilities of satellite nodes to the corresponding transmittable data volume; map the sensing capabilities of satellite nodes to the corresponding senseable data volume; map the storage capabilities of satellite nodes to the corresponding storeable data volume; and map the computing capabilities of satellite nodes to the corresponding computeable data volume. The graph construction module is used to construct a spatiotemporal model that describes the time-varying topology of the satellite network, as well as its communication, computing, and storage resources, based on the amount of data mapped from satellite nodes. The model generation module is used to generate a joint optimization model corresponding to the computation strategy and resource allocation strategy of the set of tasks to be scheduled, based on the spatiotemporal model. The scheme generation module is used to solve the joint optimization model to obtain the target scheduling scheme of the set of tasks to be scheduled in the satellite network; The task scheduling module is used to schedule tasks for the satellite network according to the target scheduling scheme.

[0014] An electronic device, comprising: Memory, used to store computer programs; A processor, used to execute the computer program to implement the steps of the satellite network multitasking scheduling method as described above.

[0015] A computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the satellite network multi-task scheduling method of computational collaboration as described above.

[0016] This application provides a multi-task scheduling method for satellite networks with transmission and computation coordination. The method involves: obtaining a set of tasks to be scheduled; mapping the link capabilities of satellite nodes to corresponding transmittable data volumes; mapping the sensing capabilities of satellite nodes to corresponding senseable data volumes; mapping the storage capabilities of satellite nodes to corresponding storable data volumes; mapping the computing capabilities of satellite nodes to corresponding computeable data volumes; constructing a spatiotemporal model based on the mapped data volumes of satellite nodes to describe the time-varying topology of the satellite network and its communication, computing, and storage resources; generating a joint optimization model corresponding to the transmission and computation strategies and resource allocation strategies of the task set to be scheduled according to the spatiotemporal model; solving the joint optimization model to obtain the target scheduling scheme for the task set in the satellite network; and scheduling tasks in the satellite network according to the target scheduling scheme. In this application, the task processing capabilities of each satellite node are mapped to corresponding data volumes. This allows for the characterization of each satellite node's task processing capabilities according to a unified standard of data volume, providing a foundation for subsequent scheduling based on changes in task data volume. Furthermore, a spatiotemporal model is used to simultaneously visualize the communication, computing, and storage resources of the satellite network. Since communication resources represent data transmission paths between network nodes, the transmission-computation strategy determines the transmission path and computing nodes for tasks within the satellite network, and the resource allocation strategy determines the amount of resources occupied by tasks on maintenance machine nodes, solving the joint optimization model generated based on the spatiotemporal model is equivalent to simultaneously integrating path selection, computation, and storage to generate the optimal scheduling scheme for the set of tasks to be scheduled. This achieves accurate allocation of satellite network resources and reasonable task scheduling. Compared to a phased method of routing followed by computation, this approach reduces the end-to-end completion time of complex coupled tasks, improves task processing efficiency, and enhances the utilization rate of satellite network resources. The transmission-computation collaborative satellite network multi-task scheduling system, electronic equipment, and computer-readable storage medium provided in this application also solve corresponding technical problems. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of this application. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0018] Figure 1 A flowchart illustrating a multi-task scheduling method for satellite networks with computational collaboration, provided as an embodiment of this application; Figure 2 This is a schematic diagram of the structure of the spatiotemporal extension graph; Figure 3 A schematic diagram of the structure of a satellite network multi-task scheduling system with computational collaboration provided in this application embodiment; Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application; Figure 5 This is another structural schematic diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0019] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0020] Please see Figure 1 , Figure 1 A flowchart of a satellite network multi-task scheduling method with computational collaboration provided in this application embodiment.

[0021] This application provides a satellite network multi-task scheduling method with transmission and computing collaboration, which can be executed by the network management center in the satellite network, and includes the following steps.

[0022] Step S101: Obtain the set of tasks to be scheduled.

[0023] In practical applications, the set of tasks to be scheduled that the satellite network needs to process can be obtained first. The number and type of tasks to be scheduled in the set can be flexibly determined according to the application scenario. For example, the set of tasks to be scheduled can be image data transmission tasks taken by the satellite, video data transmission tasks, etc., or remote control commands from ground terminal equipment to the satellite, etc. Each task to be scheduled can include parameters such as source satellite node, destination satellite node, data generation time window, and number of processing rounds.

[0024] It should be noted that during the processing of tasks, satellite networks need to transmit tasks between satellite nodes and perform calculations and storage on the tasks. The transmission, calculation and storage of tasks will affect the state of the satellite network. Therefore, it is necessary to obtain the state information of the satellite network. The state information may include the topology of the satellite network, the bandwidth of inter-satellite links, the on-board computing power and storage resources, etc., so that the set of tasks to be scheduled can be scheduled in the future based on this state information.

[0025] Step S102: Map the link capability of the satellite node to the corresponding amount of data that can be transmitted; map the sensing capability of the satellite node to the corresponding amount of data that can be sensed; map the storage capability of the satellite node to the corresponding amount of data that can be stored; map the computing capability of the satellite node to the corresponding amount of data that can be computed.

[0026] In practical applications, existing methods often treat the data stream of a task as a fixed size or process it using only a simple ratio during task transmission, failing to explicitly describe the dynamic impact of operations such as data compression and feature extraction on the task data volume during on-orbit computation. This leads to discrepancies between subsequent link bandwidth requirements and transmission delay estimates, causing resource allocation and path planning to be based on inaccurate load information. Consequently, it becomes impossible to accurately assess the impact of different scheduling strategies on overall performance, reducing the effectiveness of optimization results. To address this issue, this application maps the task processing capabilities of satellite nodes to corresponding data volumes. For example, it maps the link capabilities of satellite nodes to corresponding transmittable data volumes; the sensing capabilities of satellite nodes to corresponding senseable data volumes; the storage capabilities of satellite nodes to corresponding storeable data volumes; and the computing capabilities of satellite nodes to corresponding computeable data volumes. This allows for the characterization of the task processing capabilities of satellite nodes using a unified standard, and subsequently, the processing of tasks within the satellite network can be accurately characterized based on the data volume, providing a foundation for reasonable task scheduling.

[0027] In an exemplary embodiment, the task processing capabilities of a satellite node include data sensing, data storage, link bandwidth, and computing power. Based on this, in the process of mapping the sensing capabilities of a satellite node to the corresponding amount of sensing data, the satellite node can be determined according to the generation rate of the task to be scheduled in the time slice. The amount of data generated in task data f during time slice t. , The unit can be GB. Let t represent the source satellite node of the task to be scheduled, and t represent the time length. This represents the data generation rate of the task to be scheduled in time slice t, expressed in GB / s. By mapping the storage capacity of satellite nodes to the corresponding storable data volume, the amount of task data f stored by satellite node i in time slice t can be determined. The unit can be GB; mapping the link capacity of network satellite nodes to the corresponding amount of data that can be transmitted can be based on the link bandwidth of the satellite nodes. , This represents the link bandwidth between satellite node i and satellite node j in time slice t, which can be expressed in Gbps. It determines the amount of data that satellite node i transmits in time slice t, specifically the amount of mission data f. , The unit can be GB; in the process of mapping the computing power of a satellite node to the corresponding amount of computable data, the on-orbit computing power of satellite node i can be used as a reference. Determine the amount of data that the satellite node will use to compute the mission data in the time slice. , The unit can be GB. This represents the computational complexity per unit bit for the task f to be scheduled. This represents the unit conversion factor, used to convert bit operations into data.

[0028] Step S103: Based on the amount of data after the satellite nodes are mapped, construct a spatiotemporal model to describe the time-varying topology of the satellite network as well as its communication, computing, and storage resources.

[0029] In practical applications, to avoid treating the transmission and computation of tasks in satellite networks in isolation, it is necessary to centrally process the dynamic changes of satellite network resources and the static topology. For example, modeling methods such as spatiotemporal extension graphs, time-varying topology graphs, or time-layered graphs can be used to describe the dynamic characteristics of satellite networks. Taking spatiotemporal extension graphs as an example, a spatiotemporal model describing the time-varying topology, communication resources, computing resources, and storage resources of satellite networks can be constructed based on the amount of data after the satellite nodes are mapped. This spatiotemporal model can be a spatiotemporal extension graph, so that the data transmission, computation processing, storage consumption, and time-varying nature of network topology of tasks in satellite networks can all be expressed in a static graph structure through the amount of data.

[0030] In an exemplary embodiment, during the construction of a spatiotemporal model describing the time-varying topology, communication resources, computing resources, and storage resources of a satellite network, to characterize the time-varying characteristics of the satellite network, the physical network can be discretized along the time dimension, and a spatiotemporal extension graph can be constructed. Specifically, a set time slice can be determined, for example, the time window T can be discretized into N time slices of length Δt. Assuming T is 300 seconds and Δt is 30 seconds, then N is 10, and the numbering can range from t=0 to t=N-1, etc. In each time slice of the spatiotemporal extension graph, the satellite nodes of the satellite network are mapped to spatiotemporal nodes, so as to... The time-varying physical network is transformed into a static, multi-layered graph interwoven with spatial and temporal dimensions. Based on the transmission relationships of mission data between different satellite nodes, communication edges connecting corresponding spatiotemporal nodes within the same time slice are generated in the spatiotemporal extension graph. For example, if mission data is transmitted from satellite node 1 to satellite node 2 within a time slice, communication edges need to be added between the corresponding spatiotemporal nodes of satellite node 1 and satellite node 2 in the spatiotemporal extension graph to represent the transmission relationship of mission data between satellite node 1 and satellite node 2. The processing of mission data by the same satellite node between adjacent time slices is also considered. In the spatiotemporal extension graph, computational edges connecting corresponding spatiotemporal nodes are generated between adjacent time slices. For example, if satellite node 1 performs computations on task data in both time slice 1 and time slice 2, then computational edges need to be added between the time-series nodes of satellite node 1 corresponding to time slice 1 and time slice 2 in the spatiotemporal extension graph. Based on the relationship of the same satellite node storing task data between adjacent time slices, storage edges connecting corresponding spatiotemporal nodes are also generated in the spatiotemporal extension graph. For example, if satellite node 1 stores task data in both time slice 1 and time slice 2, then storage edges need to be added between adjacent time slices. Storage edges are added between the time-series nodes of satellite node 1 corresponding to time slice 1 and time slice 2. Based on this, the tasks to be scheduled by the satellite network are generated by the source satellite nodes. Therefore, information needs to be generated based on the task data of the satellite nodes in the time slices. In the spatiotemporal extension graph, sensing edges connecting the corresponding spatiotemporal nodes and the tasks to be scheduled within the same time slice are generated. For example, if satellite node 1 receives task 1 to be scheduled in time slice 1, then a sensing edge needs to be added between the time-series node corresponding to satellite node 1 in time slice 1 and task 1 to be scheduled in the spatiotemporal extension graph. The spatiotemporal extension graph is used as the spatiotemporal model. In this way, a spatiotemporal extension graph containing multiple time layers can be constructed, enabling a unified description of the entire process of task perception, storage, computation, and transmission by the satellite network. This provides a foundation for subsequent integrated communication, computation, and storage to schedule the set of tasks to be scheduled, avoiding situations where some resources are overloaded while others are idle, resulting in low overall resource utilization efficiency.

[0031] For ease of understanding, assume the constructed spatiotemporal extension graph is as follows: Figure 2As shown in the figure, from left to right, the three stages are: link relationship modeling, path and computing node selection, and resource allocation results. In the link relationship modeling stage, the inter-satellite link connection relationship between satellite nodes in each time slice is described, the visibility relationship between the source satellite node and the target satellite node, and the computing power of each satellite node in the current time slot. At the same time, the storage and maintenance process of task data on the same satellite node is represented by the edge connecting the same satellite node in adjacent time slices. In the path and computing node selection stage, based on the spatiotemporal expansion graph, the transmission path of multiple tasks to be scheduled is planned, and several satellite nodes are selected on the path to perform on-orbit computing processing. The satellite nodes performing computing are highlighted in the spatiotemporal expansion graph to indicate that the task to be scheduled has data processing operations in that time slice. In the resource allocation stage, the link bandwidth and satellite computing resources in each time slice are allocated according to the selected path and computing nodes to discretize the continuous planning time window into a series of time slices. In this way, the complete lifecycle of a scheduled task from the source satellite node to the destination satellite node can be visualized as one or more flow paths shuttling through the spatiotemporal extension graph. This may include multiple spatial jumps along the communication edge, in-situ computation processing along the computation edge, and in-situ waiting along the storage edge, which can accurately characterize the processing of the task in the satellite network.

[0032] Step S104: Based on the spatiotemporal model, generate a joint optimization model corresponding to the computational strategy and resource allocation strategy of the set of tasks to be scheduled.

[0033] Step S105: Solve the joint optimization model to obtain the target scheduling scheme for the set of tasks to be scheduled in the satellite network.

[0034] In practical applications, after constructing a spatiotemporal model to characterize the communication, computation, and storage of the satellite network, to achieve global optimal scheduling in multi-task concurrent scenarios, a joint optimization model corresponding to the transmission strategy and resource allocation strategy of the set of tasks to be scheduled needs to be generated based on the spatiotemporal model. This joint optimization model is used to characterize the transmission path, computation location, and resource configuration of tasks in the satellite network. Then, the joint optimization model is solved to obtain the target scheduling scheme for the set of tasks to be scheduled in the satellite network. It should be noted that, in addition to the joint optimization model, heuristic algorithms, hierarchical optimization methods, or learning-driven methods can also be used to approximate the solution to the target scheduling scheme; no specific limitations are made here.

[0035] In an exemplary embodiment, during the process of generating a joint optimization model corresponding to the computational strategy and resource allocation strategy of the set of tasks to be scheduled, based on the spatiotemporal model, joint decision variables of the computational strategy and resource allocation strategy of the set of tasks to be scheduled can be generated to characterize the processing method of the tasks to be scheduled in the satellite network. A global optimization objective is generated based on the joint decision variables to reflect the overall performance required for the satellite network to process tasks. Constraint information of the set of tasks to be scheduled is generated based on the spatiotemporal model. This constraint information includes path constraints, link capacity constraints, computing power constraints, storage constraints, data flow conservation constraints, and task completion constraints. Path constraints characterize that task paths must be continuous and legal. Link capacity constraints characterize that the total amount of data transmitted by any link in any time slice does not exceed its bandwidth limit; that is, for any inter-satellite link and any time slice, all tasks to be scheduled must be within that link. The sum of transmitted data within a time slice does not exceed the transmittable data capacity of the link; the computing power constraint characterizes that the computing resources allocated to any satellite node within any time slice do not exceed its computing power, that is, for any satellite node and any time slice, the sum of the computing data allocated by the satellite node to all scheduled tasks does not exceed the computable data capacity of the node; the storage constraint characterizes that the storage capacity of a satellite node does not exceed its upper limit, that is, for any satellite node and any time slice, the sum of the data of all scheduled tasks stored by the satellite node does not exceed the storable data capacity of the node; the data flow conservation constraint characterizes the conservation of task data during transmission and computation, and the task completion constraint characterizes the final transmission of each task to the destination satellite node, so as to uniformly characterize the competitive relationship of scheduled tasks in the communication, computing, and storage resources of the satellite network with the help of constraint information; the joint decision variables, global optimization objective, and constraint information are used as a joint optimization model.

[0036] In specific application scenarios, considering that tasks need to be transmitted, computed, and stored within satellite networks, path selection variables for the set of tasks to be scheduled can be generated during the process of generating joint decision variables for the computational strategy and resource allocation strategy of the set of tasks to be scheduled. The path selection variable is used to characterize whether to select a satellite node link to transmit the scheduled task, for example... Indicates whether the task to be scheduled, f, will choose the link formed by satellite nodes i and j for transmission; a computational node variable that generates the set of tasks to be scheduled. Node variables are used to characterize whether network nodes perform on-orbit computation for the task to be scheduled, for example... This indicates whether the task to be scheduled, f, is computed on satellite node i. Computation nodes for the same task to be scheduled must be selected from the transmission path of that task and their timing must be consistent. This is the transmission variable used to generate the set of tasks to be scheduled. The transmission variable is used to characterize the amount of data transmitted by the scheduled task through the satellite node link during the time slice, such as... This represents the amount of data transmitted by the task f to be scheduled during time slice t through the link formed by satellite nodes i and j; the computational resource variable that generates the set of tasks to be scheduled. Computational resource variables are used to characterize the computational resources allocated to the scheduled tasks by satellite nodes in a time slice, such as... This represents the computing resources allocated to task f by satellite node i in time slice t; it is also a completion indicator variable for generating the set of tasks to be scheduled. The completion indicator variable is used to characterize whether the scheduled task is completed within the time slice, for example... This indicates whether the task to be scheduled, f, is completed in time slice t. The path selection variable, computation node variable, transmission variable, computation resource variable, and completion indicator variable are used as joint decision variables. The completion indicator variable is affected by the path selection variable, computation node variable, transmission variable, and computation resource variable, and can be characterized by mathematical relationships such as models.

[0037] In specific application scenarios, during the process of generating the data flow conservation constraints for the set of tasks to be scheduled, the data compression ratio of the tasks to be scheduled (f) can be used as a reference. , This represents the data compression ratio of the task to be scheduled, f, during the k-th round of on-orbit computation, and determines the amount of data of the task to be scheduled, f, after processing by satellite node i. , , Let f represent the initial data volume of the task to be scheduled. Assuming only one round is performed, let the data volume of the task before the k-th round be... Then, after the k-th round of calculation, its data volume can be expressed as: Based on this, information on the change in the amount of data of the tasks to be scheduled after processing by the satellite nodes is generated. According to the transmittable, perceptible, storable, and computable data amounts of the satellite nodes, constraints are imposed on the amount of data consumed by the satellite nodes in each round of processing the tasks to be scheduled. That is, the amount of data consumed in each round of processing the tasks to be scheduled cannot exceed the upper limit of the satellite node's data amount. Furthermore, the input data amount of the satellite node in the current round of processing the tasks to be scheduled is constrained to be the output data amount after processing the tasks to be scheduled in the previous round, generating information on the change in the amount of data of the tasks to be scheduled processed by the satellite nodes. Based on the information on the change in the amount of data of the satellite nodes and the information on the change in the amount of data of the tasks to be scheduled, data flow conservation constraints for the tasks to be scheduled on the satellite nodes are generated. Finally, based on the data flow conservation constraints for the tasks to be scheduled on the satellite nodes, data flow conservation constraints for the set of tasks to be scheduled are generated.

[0038] Furthermore, based on the data flow conservation constraints of the tasks to be scheduled at the satellite nodes, during the process of generating the data flow conservation constraints of the set of tasks to be scheduled, for any inter-satellite link, inter-satellite link constraint information is generated. The inter-satellite link constraint information is used to constrain the amount of input data for the next satellite node to process the tasks to be scheduled to be equal to the amount of output data after the previous satellite node has processed the tasks to be scheduled. For example, if the amount of data for the tasks to be scheduled before the previous satellite node processes the tasks is 10GB and the amount of data for the tasks to be scheduled after the previous satellite node processes the tasks is 5GB, then the amount of data for the tasks to be scheduled that the current satellite node needs to process is 5GB. At this time, all data consumption of the current satellite node, such as transmission load, buffer status, and resource contention, must be based on 5GB, not 10GB. Then, according to this inter-satellite link constraint information, the data flow conservation constraints of the tasks to be scheduled at each satellite node on the inter-satellite link are integrated to generate the data flow conservation constraints of the set of tasks to be scheduled.

[0039] As can be seen from the implementation process, this application reflects the initial input process of mission data in the satellite network by measuring the amount of mission data generated by satellite nodes in time slices, thus characterizing the data perception process; it can characterize the storage process of mission data in the satellite network by measuring the amount of mission data stored by satellite nodes in time slices; it can characterize the link bandwidth consumption of mission data in the satellite network by measuring the amount of mission data transmitted by satellite nodes in time slices; it can characterize the computational resource consumption of mission data in the satellite network by measuring the amount of mission data calculated by satellite nodes in time slices; and it can characterize the changes in mission data during on-orbit computation by measuring the amount of data after the scheduled mission has been processed by satellite nodes. This invention enables collaborative modeling and optimization of different resources within a unified framework. Based on this, since task compression, redundancy removal, and feature extraction all reduce data volume, and there is a correlation between the data volume of satellite nodes and the data volume of the tasks to be scheduled, generating data flow conservation constraints for the set of tasks to be scheduled based on these data volumes allows for precise characterization of real-time data volume changes during multi-stage computation. Because data volume changes directly affect subsequent transmission load and resource allocation requirements, subsequent scheduling decisions can be based on the real dynamic evolution of data during computation, transmission, and storage, improving the accuracy of path selection and resource allocation, reducing resource conflicts and idle waste, and enhancing resource utilization. It should be noted that, in the resource modeling process, in addition to applying the unified data volume modeling method of this application, multi-dimensional resource modeling can also be used to characterize communication, computing, and storage resources separately, which will not be elaborated upon here.

[0040] In specific application scenarios, the global optimization objective can be flexibly set as needed. For example, in the process of generating the global optimization objective based on joint decision variables, a global optimization objective that minimizes the maximum completion time of the set of tasks to be scheduled can be generated based on the joint decision variables. For example, the global optimization objective can be represented as follows: , F Represents the set of tasks to be scheduled. T This represents the total processing time of the set of tasks to be scheduled. This allows multiple tasks to be integrated into a single optimization framework for joint modeling and solving, enabling coordinated scheduling of communication, computing, and storage resources globally. It explicitly characterizes resource competition among multiple tasks, thus avoiding performance losses associated with traditional phased or local optimization methods. Furthermore, by using the overall completion time as the optimization objective, it achieves global optimization of the multi-task execution process, avoiding local optima problems caused by local rules or single-task optimization. This effectively reduces the overall task completion time, improves the utilization efficiency of link bandwidth and on-orbit computing resources in the satellite network, and ultimately achieves globally optimal scheduling results in multi-task concurrent scenarios.

[0041] In an exemplary embodiment, during the process of solving the joint optimization model to obtain the target scheduling scheme for the set of tasks to be scheduled in the satellite network, the joint optimization model can be solved to obtain the target scheduling scheme for the set of tasks to be scheduled in the satellite network. The target scheduling scheme can include the optimal transmission path for each task to be scheduled, the on-orbit computing node selection scheme, and the communication and computing resource allocation scheme within each time slice. The optimal transmission path represents the optimal transmission path for the task to be scheduled in the satellite network; the on-orbit computing node selection scheme represents the satellite nodes that perform on-orbit computing for the task to be scheduled; and the communication and computing resource allocation scheme within each time slice represents the link bandwidth and computing resource allocation results within each time slice. In this way, the transmission, computing, and resource consumption of the tasks to be scheduled in the satellite network can be accurately configured.

[0042] Step S106: Perform task scheduling for the satellite network according to the target scheduling scheme.

[0043] In practical applications, after obtaining the target scheduling scheme, the satellite network can be scheduled according to the target scheduling scheme. For example, the target scheduling scheme indicates that the task to be scheduled 1 starts from the source satellite node A, is transmitted through satellite node C, satellite node D, and satellite node G to the target satellite node S, and needs to be calculated at satellite node D. The link bandwidth for transmitting the task to be scheduled between satellite nodes is set to 10GB, and the computing resources of satellite node D are set to 744TOPS (Tera Operations Per Second). Then, the corresponding settings for the task to be scheduled 1 in the satellite network need to be made accordingly.

[0044] This application provides a multi-task scheduling method for satellite networks with transmission and computation coordination. The method involves: obtaining a set of tasks to be scheduled; mapping the link capabilities of satellite nodes to corresponding transmittable data volumes; mapping the sensing capabilities of satellite nodes to corresponding senseable data volumes; mapping the storage capabilities of satellite nodes to corresponding storable data volumes; mapping the computing capabilities of satellite nodes to corresponding computeable data volumes; constructing a spatiotemporal model based on the mapped data volumes of satellite nodes to describe the time-varying topology of the satellite network and its communication, computing, and storage resources; generating a joint optimization model corresponding to the transmission and computation strategies and resource allocation strategies of the task set to be scheduled according to the spatiotemporal model; solving the joint optimization model to obtain the target scheduling scheme for the task set in the satellite network; and scheduling tasks in the satellite network according to the target scheduling scheme. In this application, the task processing capacity of satellite nodes is mapped to the corresponding data volume. The task processing capacity of each satellite node can be characterized according to a unified standard data volume, providing a foundation for subsequent scheduling based on changes in task data volume. Furthermore, the spatiotemporal model is used to simultaneously display the communication, computing, and storage resources of the satellite network. Since communication resources represent the data transmission paths between network nodes, the transmission and computing strategies determine the transmission paths and computing nodes of tasks in the satellite network, and the resource allocation strategies determine the amount of resources occupied by tasks on maintenance machine nodes, solving the joint optimization model generated based on the spatiotemporal model is equivalent to simultaneously integrating path selection, computing, and storage to generate the optimal scheduling scheme for the set of tasks to be scheduled. This achieves accurate allocation of satellite network resources and reasonable task scheduling. Compared with the phased method of routing first and then computing, it can reduce the end-to-end completion time of complex coupled tasks, improve task processing efficiency, and enhance the utilization rate of satellite network resources.

[0045] Please see Figure 3 , Figure 3 This is a schematic diagram of the structure of a satellite network multi-task scheduling system with transmission and computing collaboration, provided as an embodiment of this application.

[0046] This application provides an embodiment of a satellite network multi-task scheduling system with computational collaboration, which may include: Task acquisition module 101 is used to acquire a set of tasks to be scheduled; The data volume mapping module 102 is used to map the link capability of the satellite node to the corresponding transmittable data volume; map the sensing capability of the satellite node to the corresponding senseable data volume; map the storage capability of the satellite node to the corresponding storeable data volume; and map the computing capability of the satellite node to the corresponding computeable data volume. Graph construction module 103 is used to construct a spatiotemporal model that describes the time-varying topology of the satellite network and its communication, computing and storage resources based on the amount of data after the satellite nodes are mapped. The model generation module 104 is used to generate a joint optimization model corresponding to the computation strategy and resource allocation strategy of the set of tasks to be scheduled, based on the spatiotemporal model. The scheme generation module 105 is used to solve the joint optimization model to obtain the target scheduling scheme of the set of tasks to be scheduled in the satellite network; The task scheduling module 106 is used to schedule tasks for the satellite network according to the target scheduling scheme.

[0047] This application provides a satellite network multi-task scheduling system with computational collaboration, wherein the graph construction module may include: The time slice determination unit is used to determine the set time slice; The node mapping unit is used to map satellite nodes of the satellite network to spatiotemporal nodes in each time slice of the spatiotemporal extension graph; The graph construction unit is used to generate communication edges connecting corresponding spatiotemporal nodes within the same time slice in the spatiotemporal extended graph based on the transmission relationship of task data between different satellite nodes; to generate computation edges connecting corresponding spatiotemporal nodes between adjacent time slices based on the relationship of task data processing by the same satellite node between adjacent time slices; to generate storage edges connecting corresponding spatiotemporal nodes between adjacent time slices based on the relationship of task data storage by the same satellite node between adjacent time slices; and to generate perception edges connecting corresponding spatiotemporal nodes and the task to be scheduled within the same time slice in the spatiotemporal extended graph based on the task data generation information of satellite nodes in time slices; and to use the spatiotemporal extended graph as a spatiotemporal model.

[0048] This application provides an embodiment of a satellite network multi-task scheduling system with computational and data transfer capabilities. The model generation module may include: The variable generation unit is used to generate joint decision variables corresponding to the computation strategy and resource allocation strategy of the set of tasks to be scheduled, based on the spatiotemporal model. The objective generation unit is used to generate a global optimization objective based on the joint decision variables; The constraint generation unit is used to generate constraint information for the set of tasks to be scheduled based on the spatiotemporal model. The constraint information includes path constraints, link capacity constraints, computing power constraints, storage constraints, data flow conservation constraints, and task completion constraints. The model generation unit is used to take the joint decision variables, global optimization objective, and constraint information as a joint optimization model.

[0049] This application provides a satellite network multi-task scheduling system with transmission and computation coordination. A variable generation unit is used to: generate path selection variables for a set of tasks to be scheduled, whereby the path selection variables characterize whether to select a link between satellite nodes to transmit the tasks to be scheduled; generate computing node variables for the set of tasks to be scheduled, whereby the computing node variables characterize whether network nodes perform on-orbit computation on the tasks to be scheduled; generate transmission variables for the set of tasks to be scheduled, whereby the transmission variables characterize the amount of data transmitted by the tasks to be scheduled through the links between satellite nodes in a time slice; generate computing resource variables for the set of tasks to be scheduled, whereby the computing resource variables characterize the computing resources allocated by satellite nodes to the tasks to be scheduled in a time slice; and generate completion indicator variables for the set of tasks to be scheduled, whereby the completion indicator variables characterize whether the tasks to be scheduled are completed in a time slice; and use the path selection variables, computing node variables, transmission variables, computing resource variables, and completion indicator variables as joint decision variables.

[0050] This application provides a satellite network multi-task scheduling system with transmission and computation collaboration. The constraint generation unit is used to: determine the data volume change information of the task to be scheduled after processing by satellite nodes based on the data compression ratio of the task to be scheduled; constrain the data volume consumed by the satellite nodes in each round of processing the task to be scheduled based on the satellite nodes' transmittable data volume, sensed data volume, storable data volume, and computable data volume, and constrain the input data volume of the satellite nodes in the current round of processing the task to be scheduled to be the output data volume after the previous round of processing the task to be scheduled, thereby generating data volume change information of the satellite nodes processing the task to be scheduled; generate data flow conservation constraints for the task to be scheduled on the satellite nodes based on the data volume change information of the satellite nodes and the data volume change information of the task to be scheduled; and generate data flow conservation constraints for the set of tasks to be scheduled based on the data flow conservation constraints of the task to be scheduled on the satellite nodes.

[0051] This application provides a satellite network multi-task scheduling system with computational collaboration. The constraint generation unit is used to: generate inter-satellite link constraint information for any inter-satellite link, wherein the inter-satellite link constraint information is used to constrain the amount of input data for the next satellite node to process the task to be scheduled to be equal to the amount of output data after the previous satellite node has processed the task to be scheduled; and integrate the data flow conservation constraints of the task to be scheduled at each satellite node of the inter-satellite link according to the inter-satellite link constraint information to generate data flow conservation constraints for the set of tasks to be scheduled.

[0052] This application provides an embodiment of a satellite network multi-task scheduling system with computational and data transfer capabilities. The scheme generation module may include: The scheme generation unit is used to solve the joint optimization model to obtain the target scheduling scheme of the set of tasks to be scheduled in the satellite network. The target scheduling scheme includes the optimal transmission path to the tasks to be scheduled in the satellite network, the satellite nodes for on-orbit computation of the tasks to be scheduled, the link bandwidth and computational resource allocation results in each time slice.

[0053] This application also provides an electronic device and a computer-readable storage medium, both of which have the corresponding effects of the satellite network multi-task scheduling method with computational collaboration provided in the embodiments of this application. Please refer to... Figure 4 , Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0054] An electronic device provided in this application includes a memory 201 and a processor 202. The memory 201 stores a computer program, and when the processor 202 executes the computer program, it implements the steps of the satellite network multi-task scheduling method with computing and transmission coordination as described in any of the above embodiments.

[0055] Please see Figure 5 Another electronic device provided in this application embodiment may further include: an input port 203 connected to the processor 202 for transmitting commands input from the outside to the processor 202; a display unit 204 connected to the processor 202 for displaying the processing results of the processor 202 to the outside; and a communication module 205 connected to the processor 202 for enabling communication between the electronic device and the outside. The display unit 204 may be a display panel, a laser scanning display, etc.; the communication method adopted by the communication module 205 includes, but is not limited to, Mobile High-Definition Link (MHL), Universal Serial Bus (USB), High-Definition Multimedia Interface (HDMI), wireless connection: Wireless Fidelity (WiFi), Bluetooth communication technology, Bluetooth Low Energy communication technology, and communication technology based on IEEE 802.11s.

[0056] This application provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the steps of the satellite network multi-task scheduling method with computing and transmission coordination as described in any of the above embodiments.

[0057] The computer-readable storage media involved in this application include random access memory (RAM), memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disks, removable disks, CD-ROMs (compact disc read-only memory), or any other form of storage media known in the art.

[0058] This application provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of the satellite network multi-task scheduling method with computational collaboration as described in any of the above embodiments.

[0059] For descriptions of relevant parts of the satellite network multi-task scheduling system, electronic device, and computer-readable storage medium with transmission and computing collaboration provided in this application, please refer to the detailed description of the corresponding parts in the satellite network multi-task scheduling method with transmission and computing collaboration provided in this application, which will not be repeated here. Furthermore, parts of the technical solutions provided in this application that are consistent with the implementation principles of corresponding technical solutions in the prior art have not been described in detail to avoid excessive elaboration.

[0060] It should also be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0061] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A multi-task scheduling method for satellite networks with collaborative computing and transmission, characterized in that, include: Get the set of tasks to be scheduled; Map the link capabilities of satellite nodes to the corresponding amount of data that can be transmitted; Will The sensing capabilities of satellite nodes are mapped to the corresponding amount of perceptible data. Map the storage capacity of satellite nodes to the corresponding amount of data that can be stored; map the computing capacity of satellite nodes to the corresponding amount of data that can be computed. Based on the amount of data after satellite node mapping, a spatiotemporal model is constructed to describe the time-varying topology of the satellite network as well as its communication, computing, and storage resources. Based on the spatiotemporal model, a joint optimization model corresponding to the computational strategy and resource allocation strategy of the set of tasks to be scheduled is generated; Solving the joint optimization model yields the target scheduling scheme for the set of tasks to be scheduled in the satellite network; The satellite network is task-scheduled according to the target scheduling scheme.

2. The method according to claim 1, characterized in that, Construct a spatiotemporal model to describe the time-varying topology of the satellite network, as well as its communication, computing, and storage resources, including: Determine the set time slice; In each time slice of the spatiotemporal extension graph, the satellite nodes of the satellite network are mapped to spatiotemporal nodes; Based on the transmission relationship of mission data between different satellite nodes, communication edges connecting corresponding spatiotemporal nodes within the same time slice are generated in the spatiotemporal extension graph. Based on the relationship between the processing of task data by the same satellite node between adjacent time slices, computational edges connecting corresponding spatiotemporal nodes between adjacent time slices are generated in the spatiotemporal extension graph. Based on the relationship of the same satellite node storing task data between adjacent time slices, storage edges connecting corresponding spatiotemporal nodes are generated in the spatiotemporal extension graph. Based on the mission data of satellite nodes in the time slice, information is generated, and in the spatiotemporal extension map, a perception edge connecting the corresponding spatiotemporal node and the task to be scheduled is generated within the same time slice. The spatiotemporal extension graph is used as a spatiotemporal model.

3. The method according to claim 1, characterized in that, Based on the spatiotemporal model, a joint optimization model corresponding to the computational strategy and resource allocation strategy of the set of tasks to be scheduled is generated, including: Based on the spatiotemporal model, joint decision variables corresponding to the computational strategy and resource allocation strategy of the set of tasks to be scheduled are generated; Based on the joint decision variables, a global optimization objective is generated; Based on the spatiotemporal model, constraint information for the set of tasks to be scheduled is generated. The constraint information includes path constraints, link capacity constraints, computing power constraints, storage constraints, data flow conservation constraints, and task completion constraints. The joint decision variables, the global optimization objective, and the constraint information are used as a joint optimization model.

4. The method according to claim 3, characterized in that, The joint decision variables corresponding to the computational strategy and resource allocation strategy for the set of tasks to be scheduled include: A path selection variable is generated to form a set of tasks to be scheduled. The path selection variable is used to characterize whether to select a link to a satellite node to transmit the tasks to be scheduled. A set of tasks to be scheduled is generated using computation node variables, which are used to characterize whether a network node performs on-orbit computation on the tasks to be scheduled. A transmission variable is generated for the set of tasks to be scheduled, the transmission variable being used to characterize the amount of data transmitted by the tasks to be scheduled through the link of the satellite node in the time slice; A set of computational resource variables is generated to represent the computational resources allocated by the satellite node to the tasks to be scheduled in the time slice. A completion indicator variable is generated for a set of tasks to be scheduled, wherein the completion indicator variable is used to characterize whether the tasks to be scheduled are completed within the time slice; The path selection variable, the computing node variable, the transmission variable, the computing resource variable, and the completion indication variable are used as joint decision variables.

5. The method according to claim 4, characterized in that, The data flow conservation constraints for generating the set of tasks to be scheduled include: Based on the data compression ratio of the task to be scheduled, determine the change in the amount of data of the task to be scheduled after processing by the satellite node; Based on the amount of data that a satellite node can transmit, sense, store, and compute, the amount of data consumed by the satellite node in each round of processing the scheduled task is constrained, and the amount of input data for the satellite node in the current round of processing the scheduled task is constrained to be the amount of output data after processing the scheduled task in the previous round, thereby generating information on the change in the amount of data processed by the satellite node for the scheduled task. Based on the data volume change information of satellite nodes and the data volume change information of the tasks to be scheduled, data flow conservation constraints of the tasks to be scheduled on satellite nodes are generated. Based on the data flow conservation constraints of the tasks to be scheduled on the satellite nodes, data flow conservation constraints are generated for the set of tasks to be scheduled.

6. The method according to claim 5, characterized in that, Based on the data flow conservation constraints of the tasks to be scheduled on the satellite nodes, the data flow conservation constraints of the set of tasks to be scheduled are generated, including: For any inter-satellite link, generate inter-satellite link constraint information. The inter-satellite link constraint information is used to constrain the amount of input data for the next satellite node to process the scheduled task to be equal to the amount of output data after the previous satellite node has processed the scheduled task. Based on the inter-satellite link constraint information, the data flow conservation constraints of the tasks to be scheduled at each satellite node of the inter-satellite link are integrated to generate the data flow conservation constraints of the set of tasks to be scheduled.

7. The method according to claim 1, characterized in that, Solving the joint optimization model yields the target scheduling scheme for the set of tasks to be scheduled in the satellite network, including: Solving the joint optimization model yields the target scheduling scheme for the set of tasks to be scheduled in the satellite network; The target scheduling scheme includes the optimal transmission path to the task to be scheduled in the satellite network, the satellite nodes for on-orbit computation of the task to be scheduled, the link bandwidth and computational resource allocation results in each time slice.

8. A satellite network multi-task scheduling system with computational collaboration, characterized in that, include: The task acquisition module is used to acquire a set of tasks to be scheduled. The data volume mapping module is used to map the link capabilities of satellite nodes to the corresponding amount of data that can be transmitted. Will The sensing capabilities of satellite nodes are mapped to the corresponding amount of perceptible data. Map the storage capacity of satellite nodes to the corresponding amount of data that can be stored; map the computing capacity of satellite nodes to the corresponding amount of data that can be computed. The graph construction module is used to construct a spatiotemporal model that describes the time-varying topology of the satellite network, as well as its communication, computing, and storage resources, based on the amount of data mapped from satellite nodes. The model generation module is used to generate a joint optimization model corresponding to the computation strategy and resource allocation strategy of the set of tasks to be scheduled, based on the spatiotemporal model. The scheme generation module is used to solve the joint optimization model to obtain the target scheduling scheme of the set of tasks to be scheduled in the satellite network; The task scheduling module is used to schedule tasks for the satellite network according to the target scheduling scheme.

9. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor, configured to implement the steps of the satellite network multitasking scheduling method of claim 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the satellite network multi-task scheduling method of computational collaboration as described in any one of claims 1 to 7.

Images