Low earth orbit satellite network task unloading and route planning combined scheduling method based on time expansion graph and related device

Abstract

The invention provides a low earth orbit satellite network task unloading and route planning combined scheduling method based on a time expansion graph and a related device, and belongs to the technical field of wireless communication. According to the method, a dynamic time-varying low earth orbit satellite network is constructed; expressing the constructed dynamic time-varying low earth orbit satellite network as a directed network topological graph based on a time sequence by using a time expansion graph; constructing a joint scheduling model by using the directed network topological graph based on the time sequence; setting constraint conditions of the combined scheduling model; based on the set constraint condition of the joint scheduling model, solving the objective function of the constructed joint scheduling model to obtain an optimal result; and performing task unloading and route planning combined scheduling on the low-orbit satellite network according to the obtained optimal result to obtain a task unloading and route planning combined scheduling result. The problem that the overall performance optimization effect is poor due to the fact that the coupling relation between task unloading and routing planning in the dynamic time-varying network scene is not comprehensively considered is solved.

Description

Technical Field

[0001] The present invention belongs to the technical field of wireless communication, and particularly relates to a joint scheduling method and related device for task offloading and routing planning of a low-earth orbit satellite network based on a time-expanded graph. Background Technique

[0002] In recent years, with the full deployment and commercialization of the fifth-generation mobile communication technology, multiple countries and regions around the world are actively promoting the research and development of the sixth-generation mobile communication system (6G). The world's first 6G white paper points out that the key to the 6G vision is ubiquitous wireless intelligence, where ubiquitous means providing seamless services for users anywhere, that is, in the future, an information network that seamlessly covers all users in the air, space, ground, and sea regions will be laid, so as to achieve the integration of air, space, ground, and sea.

[0003] Low Earth Orbit (LEO) satellites are usually deployed at an altitude between 500 km and 2000 km from the ground. Compared with medium and high-orbit satellites, they have lower transmission delay and less link loss. Deploying multiple LEO satellites can achieve global coverage and is not restricted by terrain. For example, for communication scenarios in remote or extreme areas such as the ocean and desert, the application of the LEO satellite network is sufficient to overcome the problem of the lack of effective ground mobile communication coverage in these areas. The LEO satellite network has become an important part of 6G.

[0004] On the other hand, more and more new application services have emerged, such as real-time remote sensing data processing, large-scale Internet of Things information aggregation, and intelligent disaster monitoring. The continuously emerging services provide diverse services and greatly improve the user experience. However, these computationally intensive applications pose high requirements on resource-constrained servers (such as low-earth orbit remote sensing satellites), and it is difficult to process these massive task data only relying on their own weak and limited computing power and storage resources.

[0005] In this context, mobile edge computing has become a key solution for optimizing resource utilization. By offloading compute-intensive or latency-sensitive tasks to computing devices with richer computing resources and larger data storage, such as edge computing satellites with edge servers deployed for computing and processing, the shortage of computing power (also known as computing ability) among multiple satellites can be addressed, and the performance bottleneck of a single satellite can be avoided. Usually, after collecting observation data, low-computing-power remote sensing satellites need to segment tasks with complex and large amounts of subsequent calculations to ensure the integrity and orderliness within each task flow, select appropriate segmented parts as task data to be offloaded, and decide whether to offload tasks and which tasks to offload to which edge computing satellite through strategies. Based on the current network topology, considering the security and efficiency of data transmission, a feasible communication link is established. After receiving a task request, the target edge computing satellite queues and schedules it according to factors such as task priority and executes the computing instruction. Finally, the computed data is sent back to the remote sensing satellite, and the edge computing satellite releases the allocated task computing resources. From the above, it can be seen that in the complete task offloading process, decisions such as task segmentation decision, task offloading decision, edge computing satellite selection decision, and resource allocation decision all require careful planning and scheduling by the entire network system. These decisions jointly determine the quality of service of the computing network.

[0006] Chen et al. proposed an adaptive particle swarm optimization algorithm based on genetic algorithm operators, which effectively formulates offloading strategies for deep neural network layers through layer partitioning operations, and simulation experiments verify that this algorithm strategy can significantly reduce energy consumption. Wu et al. decoupled and separated the power and computing resource allocation problems by establishing an analytical model, solved the KKT (Karush-Kuhn-Tucker) stationary points of each sub-problem respectively, and optimized the system power consumption through a heuristic algorithm. Li et al. studied the task offloading problem with statistical QoS guarantee, formulated it as a mixed-integer non-linear programming problem with constraints on CPU cycle numbers and delay exponents, and used the Gibbs sampling method for solution. Tang et al. established a dynamic scheduling model for task queues, and used Lyapunov optimization theory to decouple and transform the joint optimization problem of all time slots into computing resource allocation and power bandwidth sub-problems and solve them separately. The simulation results verify the superiority of the algorithm. Gang et al. proposed an efficient inter-satellite collaborative computing offloading algorithm for LEO satellite networks to reduce latency and energy consumption, and solved the sub-problems through the Lagrange multiplier method and the binary particle swarm optimization algorithm respectively.

[0007] In recent years, a large number of relevant literatures and achievements have emerged in the research field of task offloading technology. Although emerging technologies are emerging in an endless stream, there are still the following two problems: 1. There is a lack of attention to the joint optimization of task offloading and offloading routing planning. Task offloading not only needs to consider how to offload computing tasks from low-computing-power nodes to edge nodes or the cloud, but also needs to optimize the offloading routing simultaneously to ensure that data can be transmitted efficiently and stably in a complex network environment. During the task offloading process, the choice of offloading routing directly determines the path and efficiency of task transmission. If the choice is inappropriate, it may lead to link congestion and transmission delay, or even data loss. Currently, most research mainly focuses on a single decision of task offloading, ignoring the inherent coupling relationship between task offloading and routing selection. Even when considering routing optimization, most adopt a separate method to first plan task offloading and then independently select the transmission path, lacking means to model the interaction between the two simultaneously.

[0008] 2. The dynamic changes in the actual network environment are ignored. Currently, existing research generally assumes that the topology of the low-earth orbit satellite network is fixed, defaulting that there is always a communication link between satellites. This simplification ignores the complexity of the actual dynamic environment. In an actual low-earth orbit satellite network, the high-speed movement of satellites, link discontinuity, and channel state changes make it difficult for traditional models to adapt to the characteristics of dynamic topologies and unable to effectively address practical problems such as link interruption and resource competition, thus limiting the practicality and reliability of task scheduling and resource allocation strategies.

[0009] In summary, the current existing technologies mainly focus on task offloading decision-making or routing planning optimization, without comprehensively considering the coupling relationship between task offloading and routing planning in a dynamic time-varying network scenario, resulting in poor global performance optimization. Summary of the Invention

[0010] The purpose of the present invention is to provide a method and related device for joint scheduling of task offloading and routing planning in a low-earth orbit satellite network based on a time-expanded graph, aiming to solve the problem in the existing technology that the global performance optimization effect is poor due to the lack of comprehensive consideration of the coupling relationship between task offloading and routing planning in a dynamic time-varying network scenario.

[0011] To achieve the above purpose, the present invention adopts the following technical solutions: In the first aspect, the present invention provides a method for joint scheduling of task offloading and routing planning in a low-earth orbit satellite network based on a time-expanded graph, including the following steps: Construct a dynamic time-varying low-earth orbit satellite network; Use the time-expanded graph to represent the constructed dynamic time-varying low-earth orbit satellite network as a directed network topology graph based on a time series; Construct a joint scheduling model using the directed network topology graph based on a time series, where the joint scheduling model takes minimizing the total energy consumption as the objective function; Set the constraint conditions of the joint scheduling model; Based on the constraint conditions of the established joint scheduling model, solve the objective function of the constructed joint scheduling model to obtain the optimal result; According to the obtained optimal result, perform joint scheduling of task offloading and routing planning for the LEO satellite network to obtain the joint scheduling result of task offloading and routing planning.

[0012] A further improvement of the present invention is that the expression of the objective function for minimizing the total energy consumption is:

[0013] Wherein, is a remote sensing satellite, is an edge computing satellite, is a set of time slots, is the objective function for minimizing the total energy consumption, is the time slot is the energy consumption of local computing of the remote sensing satellite within the time slot is the transmission energy consumption of the remote sensing satellite within the time slot is the energy consumption of the edge computing satellite for computing and processing remote sensing data within the time slot is the time slot is the energy consumption of the edge computing satellite for participating in transmission relay within the time slot is the time slot is the total energy consumption of the edge computing satellite for participating in transmission relay within the time slot

[0014] A further improvement of the present invention is that the constraint conditions of the joint scheduling model include the following formulas:

[0015]

[0016]

[0017]

[0018]

[0019]

[0020] Wherein, k is the time slot, is the remote sensing satellite, is the edge computing satellite, is the task offloading strategy, is the offloading target selection strategy, is the total amount of data waiting for local computing of the remote sensing satellite, is the maximum computing storage capacity of the remote sensing satellite, is the total amount of data waiting for computing of the edge computing satellite, is the maximum computing and storage capacity of the edge computing satellite, is a constant, .

[0021] A further improvement of the present invention lies in that after constructing the joint scheduling model using the directed network topology graph, first decouple the objective function of the constructed joint scheduling model to obtain the objective function of the decoupled joint scheduling model, and then solve the objective function of the decoupled joint scheduling model.

[0022] A further improvement of the present invention lies in that the specific steps for decoupling the objective function of the constructed joint scheduling model include: Construct a routing planning model; In the constructed routing planning model, use the Dijkstra algorithm to find the optimal offloading paths from each remote sensing satellite node to the edge computing satellite node at different times; Obtain the offloading path loss based on the optimal offloading paths from each remote sensing satellite node to the edge computing satellite node at different times; Multiply the path loss by the offloading data volume, and superimpose the results of multiplying the path loss by the offloading data volume to obtain the objective function of the decoupled joint scheduling model.

[0023] A further improvement of the present invention lies in that The expression of the objective function of the decoupled joint scheduling model is:

[0024] Wherein, is the energy consumption of local computing of the remote sensing satellite within the time slot ; is the energy consumption of the edge computing satellite for computing and processing remote sensing data within the time slot ; is the total energy consumption of the overall transmission process; The total energy consumption of the overall transmission process is calculated by the formula:

[0025] Wherein, is the offloading target selection strategy, is the data volume that needs to be offloaded and processed, is the basic loss of the offloading route.

[0026] A further improvement of the present invention lies in that in the step of solving the objective function of the established joint scheduling model based on the set constraint conditions of the joint scheduling model to obtain the optimal result, the alternating optimization algorithm is specifically used to solve the objective function of the established joint scheduling model to obtain the optimal result.

[0027] In a second aspect, the present invention provides a joint scheduling system for task offloading and routing planning in a low-earth orbit satellite network based on a time-expanded graph, including a satellite network construction module, a directed network topology graph acquisition module, a joint scheduling model construction module, a constraint condition setting module, an objective function solving module, and a joint scheduling module; The satellite network construction module is used to construct a dynamic and time-varying low-earth orbit satellite network; The directed network topology graph acquisition module is used to represent the constructed dynamic and time-varying low-earth orbit satellite network as a directed network topology graph based on a time series by using a time-expanded graph; The joint scheduling model construction module is used to construct a joint scheduling model by using the directed network topology graph based on a time series, and the joint scheduling model takes minimizing the total energy consumption as the objective function; The constraint condition setting module is used to set the constraint conditions of the joint scheduling model; The objective function solving module is used to solve the objective function of the constructed joint scheduling model based on the set constraint conditions of the joint scheduling model to obtain an optimal result; The joint scheduling module is used to perform joint scheduling of task offloading and routing planning on the low-earth orbit satellite network according to the obtained optimal result to obtain a joint scheduling result of task offloading and routing planning.

[0028] In a third aspect, the present invention provides an electronic device, including a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the steps of the above-described joint scheduling method for task offloading and routing planning in a low-earth orbit satellite network based on a time-expanded graph are implemented.

[0029] In a fourth aspect, the present invention provides a storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-described joint scheduling method for task offloading and routing planning in a low-earth orbit satellite network based on a time-expanded graph are implemented.

[0030] Compared with the prior art, the present invention has the following beneficial effects: The present invention belongs to an improved invention. Compared with the existing methods for task offloading and routing planning and scheduling in low-earth orbit satellite networks, on the one hand, the present invention comprehensively considers the close coupling relationship between task offloading decisions and routing planning, and proposes a joint optimization scheduling method. Specifically, task offloading decisions directly affect the distribution of offloading routes and the link transmission status, while the status of offloading routes in turn restricts the feasibility and performance of optional offloading schemes. Therefore, when making task offloading decisions, the present invention comprehensively considers the dynamic changes of link status and network topology in real time, collaboratively optimizes task offloading and offloading routing strategies, and achieves the optimal global performance, thus effectively solving the problem in the prior art that the global performance optimization effect is not good due to the failure to comprehensively consider the coupling relationship between task offloading and routing planning in a dynamically time-varying network scenario. On the other hand, the present invention can achieve the efficient coordination of computing resources and communication resources in a dynamically changing low-earth orbit satellite network environment. It can not only avoid the overload of computing nodes or congestion of communication paths that may occur due to separate optimization, but also significantly improve the overall task processing efficiency and resource utilization rate of the system, and finally achieve better energy consumption performance.

[0031] Secondly, the present invention constructs a joint scheduling model using a directed network topology graph. Specifically, a joint scheduling model is constructed through a time series model of the time-varying network state (a directed network topology graph based on time series), effectively solving the problem of topological time-variability caused by the high-speed movement of nodes in a low-earth orbit satellite network, so that more accurate scheduling decisions can be made.

[0032] Furthermore, the present invention discloses the expression of the objective function for minimizing the total energy consumption. It can be seen from the objective function for minimizing the total energy consumption that the present invention considers various factors when establishing the joint scheduling model.

[0033] Furthermore, the present invention discloses the constraint conditions (6) of the joint scheduling model. It can be seen that the present invention comprehensively considers various constraint conditions, and comprehensively considering various constraint conditions can more comprehensively reflect the actual situation of the joint scheduling problem of task offloading and routing planning.

[0034] Furthermore, after the present invention discloses constructing a joint scheduling model using a directed network topology graph, the objective function of the constructed joint scheduling model is first decoupled to obtain the objective function of the decoupled joint scheduling model, and then the objective function of the decoupled joint scheduling model is solved. By decoupling, the objective function of the complex joint scheduling model is decomposed into multiple relatively simple sub-models (the objective function of the decoupled joint scheduling model), which can significantly reduce the computational complexity of each sub-model, thereby improving the solving efficiency of the later model. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1This is the flowchart of the joint scheduling method for task offloading and routing planning in a low-earth orbit satellite network based on a time-expanded graph of the present invention; Figure 2 This is the schematic diagram of the joint scheduling system for task offloading and routing planning in a low-earth orbit satellite network based on a time-expanded graph of the present invention; Figure 3 This is the architecture diagram of the low-earth orbit satellite network of the present invention; Figure 4 This is the directed network topology graph based on time series of the present invention; Figure 5 This is the comparison graph of the total energy consumption of the joint scheduling method of the present invention and two other algorithm systems versus the number of remote sensing satellites; Figure 6 This is the schematic diagram of the relationship between the optimal energy consumption value obtained by the joint scheduling method of the present invention and the number of remote sensing satellites under different numbers of edge computing satellites; Figure 7 This is the structural schematic diagram of the electronic device of the present invention. Specific embodiments

[0036] To further understand the content of the present invention, the following describes the present invention in detail with reference to the accompanying drawings and specific embodiments. It should be understood that the embodiments are only for explaining the present invention and not for limiting it.

[0037] The joint scheduling method for task offloading and routing planning in a low-earth orbit satellite network based on a time-expanded graph proposed by the present invention solves the objective function of the constructed joint scheduling model based on the constraint conditions of the set joint scheduling model to obtain the optimal result; and performs joint scheduling of task offloading and routing planning for the low-earth orbit satellite network according to the obtained optimal result to obtain the joint scheduling result of task offloading and routing planning. Compared with the prior art, the present invention effectively solves the problem that the global performance optimization effect is not good due to the failure to comprehensively consider the coupling relationship between task offloading and routing planning in a dynamically time-varying network scenario in the prior art.

[0038] Embodiment 1: The flowchart of the joint scheduling method for task offloading and routing planning in a low-earth orbit satellite network based on a time-expanded graph of the present invention is as Figure 1 shown. The joint scheduling method for task offloading and routing planning in a low-earth orbit satellite network based on a time-expanded graph of the present invention includes the following steps: S1. Construct a dynamically time-varying low-earth orbit satellite network.

[0039] S2. Use the time-expanded graph to represent the constructed dynamically time-varying low-earth orbit satellite network as a directed network topology graph based on time series.

[0040] S3. Use the directed network topology graph based on time series to construct a joint scheduling model, and the joint scheduling model takes minimizing the total energy consumption as the objective function.

[0041] S4. Set the constraint conditions of the joint scheduling model.

[0042] S5. Solve the objective function of the constructed joint scheduling model based on the set constraint conditions of the joint scheduling model to obtain the optimal result.

[0043] S6. Perform joint scheduling of task offloading and routing planning for the low-earth orbit satellite network according to the obtained optimal result to obtain the joint scheduling result of task offloading and routing planning.

[0044] Embodiment 2: The schematic diagram of the joint scheduling system for task offloading and routing planning of the low-earth orbit satellite network based on the time-expanded graph of the present invention is as Figure 2 shown. The joint scheduling system for task offloading and routing planning of the low-earth orbit satellite network based on the time-expanded graph of the present invention includes a satellite network construction module, a directed network topology graph acquisition module, a joint scheduling model construction module, a constraint condition setting module, an objective function solving module, and a joint scheduling module.

[0045] Among them, the satellite network construction module is used to construct a dynamic and time-varying low-earth orbit satellite network.

[0046] The directed network topology graph acquisition module is used to represent the constructed dynamic and time-varying low-earth orbit satellite network as a directed network topology graph based on time series by using the time-expanded graph.

[0047] The joint scheduling model construction module is used to construct a joint scheduling model by using the directed network topology graph based on time series, and the joint scheduling model takes minimizing the total energy consumption as the objective function.

[0048] The constraint condition setting module is used to set the constraint conditions of the joint scheduling model.

[0049] The objective function solving module is used to solve the objective function of the constructed joint scheduling model based on the set constraint conditions of the joint scheduling model to obtain the optimal result.

[0050] The joint scheduling module is used to perform joint scheduling of task offloading and routing planning for the low-earth orbit satellite network according to the obtained optimal result to obtain the joint scheduling result of task offloading and routing planning.

[0051] Embodiment 3: The joint scheduling method for task offloading and routing planning of the low-earth orbit satellite network based on the time-expanded graph of the present invention includes the following steps: S1. Construct a dynamic and time-varying low-earth orbit satellite network.

[0052] First, construct a dynamic and time-varying low-earth orbit satellite network. The architecture of the dynamic and time-varying low-earth orbit satellite network is as Figure 3As shown in the figure, the low Earth orbit satellite network includes three main functional layers: the user layer, the observation layer, and the access layer. The task requirements and data interaction between the layers are realized through communication links. The specific introduction of each layer is as follows: The core of the user layer is the ground center, which aggregates the task requirements of ground users. The requirements cover multiple fields, such as emergency rescue and environmental monitoring, etc., and organizes them into a standardized task set. These tasks are uploaded to the low Earth orbit satellite network through specific interfaces for remote sensing satellites to execute.

[0053] The observation layer consists of several remote sensing satellites operating in low orbits. Each remote sensing satellite is equipped with a high-resolution remote sensing camera for real-time acquisition of surface images or dynamic target information. Due to the high coverage rate and short revisit cycle of remote sensing satellites, they have significant flexibility in space observation tasks. However, limited by the satellite's power consumption and hardware configuration, its computing power and storage capacity are relatively limited. To complete the further processing of data, the remote sensing satellite can either use its own computing power to complete basic tasks or choose to offload the data to the edge computing satellite in the access layer through the inter-satellite link for calculation.

[0054] The access layer is composed of edge computing satellites distributed in higher orbits. These satellites are equipped with powerful computing modules and large-capacity storage systems to support the processing of compute-intensive tasks. Compared with remote sensing satellites, although the number of edge computing satellites is smaller, the edge computing satellites have stronger computing power support capabilities. Through dynamic links with remote sensing satellites, the access layer can receive offloaded task requests and return the processing results in a timely manner after completion, significantly shortening the overall task response time.

[0055] S2. Use the time-expanded graph to represent the constructed dynamic and time-varying low Earth orbit satellite network as a time-series-based directed network topology graph.

[0056] Use the time-expanded graph (TEG) to represent the constructed dynamic and time-varying low Earth orbit satellite network as a time-series-based directed network topology graph. The time-series-based directed network topology graph is as Figure 4 shown. Assume that the entire task processing time is divided into discrete time slots, and the length of each time slot is . Within the same time slot, the network topology is assumed to be fixed, while the link connection status between adjacent time slots may change with satellite movement, thus dynamically reflecting the evolution of the entire network over time.

[0057] Node ( Figure 4The black circles represent nodes. Time expansion: Assume that the satellite network consists of two types of nodes, remote sensing satellites and edge computing satellites. The reference to the time-expanded graph adds time characteristics. Each node is replicated according to different time points and expanded into multiple twin nodes, representing different states of the same node at different time points, effectively capturing the dynamic characteristics of the node's change over time and reflecting the state evolution of the node in different time slots.

[0058] Specifically, represents the set of all satellite nodes, , where and represent the set of remote sensing satellites and the set of edge computing satellites respectively. For each vertex in the set or , in the time-expanded graph, the node expands into twin nodes or according to time evolution. Therefore, the static node set can be further transformed into a node set representing different moments, where .

[0059] Spatiotemporal expansion link ( Figure 4 each arrow in represents the spatiotemporal expansion link) generation: Figure 4 The link in includes a cache link and a transmission link. The cache link represents a directed edge connecting the same satellite between two consecutive time slots, denoted by , which is used to represent the internal storage of task data in the satellite, including unfinished computing tasks and task data to be offloaded. In fact, the cache link does not exist in the real network, only meaning that the data is stored on the node. To avoid confusion in data processing, the task data in the cache link is divided into two independent queues, respectively used to store task data waiting for computing and waiting for transmission; the transmission link represents a directed edge connecting different satellites between two consecutive time slots, denoted by , including inter-satellite links within the edge computing satellite layer, inter-satellite links between remote sensing satellites, and cross-layer inter-satellite links between remote sensing satellites and edge computing satellites. Finally, the cache link and the transmission link between nodes at adjacent times together constitute a fixed network topology within a unit time slot, thereby effectively capturing the change law of satellite communication links and providing support for the optimal path selection of computing offloading tasks.

[0060] S3. Construct a joint scheduling model using the time-series based directed network topology graph.

[0061] Construct a joint scheduling model using the time-series based directed network topology graph, where the joint scheduling model takes minimizing the total energy consumption as the objective function.

[0062] The expression of the objective function for minimizing the total energy consumption (also called ) is as follows:

[0063] Where, is a remote sensing satellite, is an edge computing satellite, is the set of time slots, is the objective function for minimizing the total energy consumption (specifically minimizing the total energy consumption of two types of satellites, including remote sensing satellites and edge computing satellites), is the time slot the energy consumption of local computing of the remote sensing satellite within, is the transmission energy consumption of the remote sensing satellite within the time slot , is the time slot the energy consumed by the edge computing satellite for computing and processing remote sensing data within, is the time slot the total energy consumption of the edge computing satellite participating in transmission relay within. The following is a detailed description of each parameter ( , , and ): A. Energy consumption of local computing of remote sensing satellites

[0064] Assume that within the time slot , the generation of the original remote sensing data (i.e., tasks) to be processed by the remote sensing satellite is independent and identically distributed. Let the average task generation rate be , then the amount of newly generated data per unit time slot is , satisfying . Each remote sensing satellite can choose to perform local computing or offload to the edge computing satellite for operation and processing. Let represent the task offloading strategy, indicating the proportion of tasks selected for offloading among the newly generated tasks , and the calculation formula of is:

[0065] Where, is the amount of data processed by local computing, is the amount of data that needs to be offloaded for processing.

[0066] Assume that processing 1 bit of computational task input requires CPU cycles, and the CPU cycle frequency of the remote sensing satellite is 。Then the remote sensing satellite for the mission Local computing rate The calculation formula is as follows:

[0067] The amount of data that can be processed locally within a unit time slot It is expressed as:

[0068] The amount of unprocessed data remaining after the time slot ends Will wait for the calculation and processing in the next time slot in the calculation queue, and its calculation formula is:

[0069] Therefore, the remote sensing satellite The amount of data queuing in the queue per time slot The calculation formula is updated to:

[0070] According to circuit theory, the power consumption of the CPU is mainly determined by the dynamic power consumption, and the dynamic power consumption comes from the switching activities of the internal logic gates of the CPU and is proportional to the square of the circuit voltage And the CPU cycle frequency Of the square. In addition, during low-voltage operation, the CPU cycle frequency is approximately linearly related to the chip voltage. Therefore, the local computing power consumption of the remote sensing satellite The calculation formula is:

[0071] Among them, Is the constant power consumption coefficient related to the internal structure of the device.

[0072] Therefore, the time slot The energy consumption of local computing within The calculation formula is:

[0073] B. Transmission energy consumption of remote sensing satellite

[0074] Let Represent the offloading target selection strategy for edge computing satellites. Among them, Represents the remote sensing satellite Select the edge computing satellite As the target for task offloading calculation. It is stipulated that the remote sensing satellite can select at most one edge computing satellite to be associated as its offloading target node and select a suitable transmission route to offload Of the task, satisfying Each edge computing satellite allows receiving offloading tasks from multiple remote sensing satellites for parallel computing and processing.

[0075] For the tasks to be offloaded, the remote sensing satellite has multiple routing options available, and each routing can transmit data to the target edge computing node. Define the transmission routing set , where refers to the routing set starting from different remote sensing satellite nodes and ending at different edge computing satellite nodes, which is listed by the enumeration method. , is the routing sequence number for offloading from the same initial point to the same edge computing satellite node at different times. The binary variable judges whether the corresponding node of satellite at time (the time between time slots and ) is on the offloading route , where , . , . Indicates that is on the transmission route , indicates the non - existence.

[0076] The transmission rate of the remote sensing satellite transmitting data to the next node of the route is calculated by the formula:

[0077] where, is the transmission power of satellite . Considering the Rayleigh fading environment, , is the channel fading coefficient between satellites, is the variance of Gaussian white noise, is the transmission bandwidth of the remote sensing satellite . At the same time, the transmission delay (the amount of data to be offloaded and processed) is constrained not to be greater than the duration of each time slot, indicating that any amount of data can be transmitted within one time slot.

[0078] and respectively represent the starting remote sensing satellite node and the edge computing satellite node selected as the offloading target. Then the energy consumption of the remote sensing satellite transmitting data to satellite is calculated by the formula:

[0079] Therefore, remote sensing satellites are in the time slot Transmission energy consumption within The calculation formula is:

[0080] C. Edge computing satellite computing energy consumption

[0081] Define the CPU clock frequency of the edge computing satellite as , the unit is cycle / s, and the calculation intensity is (The number of CPU cycles required to process 1 bit of task), the unit is cycle / bit. In-house edge computing satellites Offload task processing rate The calculation formula is:

[0082] in, Represents the loss coefficient related to the architecture of the edge computing satellite, so the energy consumption is calculated The expression is:

[0083] For each edge computing satellite , there is multi-task offloading caused by the access of multiple remote sensing satellites. If the edge computing end receives the offload task before a certain time, it will automatically be incorporated into the computing queue and wait for processing in the subsequent time slot. For binary parameters ,like This means that the current edge computing satellite is the final target satellite node for unloading. Therefore, the time slot The amount of data received for the task to be calculated The calculation formula is:

[0084] Time Slot The total amount of data waiting to be processed in the queue The calculation formula is:

[0085] The remaining tasks that were not completed in the previous time slot The calculation formula is:

[0086] in, Time slot The maximum amount of task data that can be processed is calculated as follows:

[0087] Therefore, the inner edge computing satellite computing energy consumption expression is:

[0088] D. Total energy consumption of edge computing satellite transmission relay

[0089] The transmission rate calculation formula of the edge computing satellite is :

[0090] where and are the transmission bandwidth and transmission power of the edge computing satellite respectively.

[0091] The edge computing satellite to transmission energy consumption calculation formula is:

[0092] Therefore, the total energy consumption of the inner edge computing satellite participating in the transmission relay calculation formula is:

[0093] S4. Set the constraint conditions of the joint scheduling model.

[0094] The constraint conditions of the joint scheduling model include the following formulas: C1:

[0095] C2:

[0096] C3:

[0097] C4:

[0098] C5:

[0099] C6:

[0100] where k is the time slot, is the remote sensing satellite, is the edge computing satellite, is the task offloading strategy, is the offloading target selection strategy, is the total amount of data waiting for local computing by the remote sensing satellite, is the maximum computing storage capacity of the remote sensing satellite, is the total amount of data waiting for computing by the edge computing satellite, is the maximum computing storage capacity of the edge computing satellite, is a constant, .

[0101] The following is a detailed description of the constraint conditions: Specifically, C1 is the upper and lower bound constraint of the task offloading strategy C2 represents the offloading target selection strategy which is a binary variable. C3 can ensure that each remote sensing satellite offloads tasks to at most one edge computing satellite. C4 indicates the relationship of mutual restraint between and i.e., when , then . Conversely, if , then there must exist a such that

[0102] S5. Based on the constraint conditions of the set joint scheduling model, solve the objective function of the constructed joint scheduling model to obtain the optimal result.

[0103] In this step, the alternating optimization algorithm is specifically used to solve the objective function of the established joint scheduling model to obtain the optimal result.

[0104] S6. According to the obtained optimal result, perform joint scheduling of task offloading and routing planning for the low-earth orbit satellite network to obtain the joint scheduling result of task offloading and routing planning.

[0105] Example 4: After step S3 of this example uses the directed network topology graph to construct the joint scheduling model, first decouple the objective function of the constructed joint scheduling model to obtain the objective function of the decoupled joint scheduling model, and then step S4 solves the objective function of the decoupled joint scheduling model (specifically, use the alternating optimization algorithm to solve the objective function of the decoupled joint scheduling model to obtain the optimal result).

[0106] Construct a routing planning model; In the constructed routing planning model, use the Dijkstra algorithm to find the optimal offloading path from each remote sensing satellite node to the edge computing satellite node at different times; The offloading path loss is obtained based on the optimal offloading paths of each remote sensing satellite node to the edge computing satellite node at different times; Multiply the path loss by the offloaded data volume, and superimpose the results of multiplying the path loss by the offloaded data volume to obtain the objective function of the decoupled joint scheduling model.

[0107] The following provides a detailed explanation of the decoupling of the objective function of the constructed joint scheduling model: Since there is an assumption during data transmission, it means that the parallel transmission of data by either the remote sensing satellite or the edge computing satellite acting as a relay can be completed within a unit time slot. Moreover, according to and , it can be seen that the communication transmission rate of the remote sensing satellite and the communication transmission rate of the edge computing satellite are only variables related to the current node and the transmission destination node, and are independent of the time slot. As shown in the directed network topology diagram Figure 4 , the routing selection problem can be separately abstracted for optimization . The energy consumption of each path can be represented by , and the variables and are coefficients related to the nodes and generally have definite values. Therefore, it can be assumed that is the basic loss cost of each link. Specifically, the cost of the inter-satellite link with the remote sensing satellite as the sender is represented as , and the cost with the edge computing satellite as the sender is represented as . For example, the communication storage cost inside the node is set to 0, and if there is no communication link between two nodes, the cost is set to infinity. Therefore, for the data to be offloaded from each remote sensing satellite within each time slot, a unique offloading route can be independently found, defined as the set of all satellite nodes that the newly generated offloading task of the remote sensing satellite node in time slot needs to be transmitted to the edge satellite . At the same time, introduce to represent the basic energy consumption of the offloading route , which is the total basic cost of the corresponding route.

[0108] Due to the binary variable in that is difficult to determine and control the computing and transmission energy consumption simultaneously, it can be transformed into an independent transmission energy consumption solution problem, and the computing energy consumption is no longer affected by a third variable other than and .

[0109] For the variable The determination can be summarized as finding the path with the minimum cost for each remote sensing satellite to offload and transmit new tasks to the target computing satellite within each time slot. Define the path set as , and the cost set as , which correspond to each other. Each node of each remote sensing satellite in each time slot is regarded as the starting point of the path, that is, . The end points are different edge computing satellites at different times , denoted as . Therefore, there are a total of optimal transmission routes, saved as , where . In fact, is the state of the edge computing satellite node at different times, and it plays the role of offloading task computing and processing in the network. Therefore, only the route with the minimum cost needs to be selected in , indicating that the remote sensing satellite offloads the task to the target offloading node 's optimal transmission route , corresponding to the cost , generating -order matrices and . The present invention calculates (also called the optimal offloading path from each remote sensing satellite node to the edge computing satellite node at different times) and matrix (also called the offloading path loss) through the Dijkstra algorithm. The specific calculation steps are as follows: a. Assignment According to the parameters of all paths towards the starting node, sequentially assign the basic cost values of each link in the TEG; b. Route search Respectively take as the starting point, and search for the optimal path with as the end point and the minimum cost ; c. Determination of the optimal route Select the offloading point with the minimum cost in , determine and , and generate and matrices. and

[0110] The basic loss value of the path to be optimized is determined. According to the above analysis, the basic loss multiplied by the amount of offloaded data of the current road section (also called multiplying the path loss by the amount of offloaded data) is the transmission energy consumption. Therefore, considering all time slots, all remote sensing satellites are offloaded to the possible edge computing ends, and the total energy consumption of the overall transmission process can be converted into the following expression:

[0111] For edge computing satellites , a judgment variable is introduced for the time slot when the task offloading reaches the target satellite to determine whether the newly generated offloading task of the remote sensing satellite arrives at the current node at the end of the time slot . If it meets the requirements, the variable is assigned a value of 1, otherwise 0. The total amount of data waiting for operation and processing in the time slot queue is calculated as: The total amount of data waiting for operation and processing in the time slot is calculated as:

[0112] is transformed into the objective function of the decoupled joint scheduling model (also called ), and its expression is:

[0113] where is the energy consumption of local computing of the remote sensing satellite in the time slot , is the energy consumption of the edge computing satellite for computing and processing remote sensing data in the time slot , is the total energy consumption of the overall transmission process.

[0114] The specific steps for solving the objective function of the decoupled joint scheduling model by using the alternating optimization algorithm in step S4 of this embodiment include: A. Initialization First, set the upper limit values and of the two-layer loop and the initial loop number . According to the constraint relationships of C1 - C4, set the initial values of and ; B. Fix Solve the single-objective optimization problem of with respect to , and solve and update the optimal solution 。According to Calculate the system energy consumption ; C. Fix Solve Regarding for the single-objective optimization problem, solve and update the optimal solution 。According to Calculate the system energy consumption ; D. Check the convergence condition of the inner loop Compare the change amount of the alternating optimization. If the convergence condition is satisfied or the maximum number of loop iterations of the inner loop is reached , then stop the inner iteration loop and output the inner optimal value 。Otherwise, let , and return to step B to continue the iteration; E. Check the convergence condition of the outer loop Compare the change amount between the current iteration result and the previous round result. If the convergence condition is satisfied or the maximum number of loop iterations of the outer loop is reached , then stop the outer iteration loop and output the optimal solution and as well as the optimal value 。Otherwise, let , and return to step B to continue the iteration.

[0115] For other steps (step S1, step S2, step S5, and step S6) of this embodiment, refer to Embodiment 3.

[0116] To verify the effectiveness of the joint scheduling method for task offloading and routing planning in low-earth orbit satellite networks proposed by the present invention, this embodiment uses MATLAB2019 software for simulation experiments. The specific configuration of the simulation experiment parameters is as follows: The remote sensing satellite constellation includes remote sensing satellites, and the edge satellite constellation includes edge satellites. The transmission power between the remote sensing satellites and the edge computing satellites ranges from 1 to 5W and 5 to 10W respectively, and the bandwidth is 10MHz and 20MHz respectively. The computing capabilities of the two types of satellites are 1 to 2GHz and 2 to 4GHz respectively, the CPU clock frequency is 737.5 cycle / bit, and the capacitance loss coefficient , and the noise power spectral density is specified as -174 dBm / Hz. The simulation simulates the joint optimization of time slots, and the unit time slot is s.

[0117] To evaluate the superiority of the proposed joint scheduling method for task offloading and routing planning in a low-earth orbit satellite network based on a time-expanded graph, the present invention designs comparison methods: a single-slot offloading algorithm and a random offloading algorithm. Specifically, the single-slot offloading algorithm refers to optimizing the offloading ratio and transmission route only for the newly generated tasks in the current time slot on the premise that the network topology remains fixed within a unit time slot. For example, in Figure 4 For the newly generated data, if offloading is considered, a decision is made based on the inter-satellite link graph within that time slot, and only edge computing nodes with valid time windows can be selected as the offloading destination. The random offloading algorithm randomly generates the offloading ratio of each remote sensing satellite in each time slot , and then calculates the optimal offloading selection . The results of the simulation experiment are as Figure 5 shown. As the number of remote sensing satellites increases, the total system energy consumption gradually rises. However, since the two nodes are not connected during offloading route selection and the basic cost is infinite, and such situations cannot be avoided in the random offloading algorithm, the value of its total energy consumption is much greater than that of the other two algorithms, exceeding multiple orders of magnitude. The proposed algorithm (also called the method proposed in the present invention) and the single-slot offloading algorithm consider the connectivity between nodes, thus restricting the value of the offloading ratio and keeping the total system energy consumption at a relatively low level. According to Figure 5 the curve trend in

[0118] it can be seen that the energy consumption of the proposed algorithm is generally lower than that of the single-slot offloading algorithm, but the gap between them is small. This is because the difference between the two algorithms mainly lies in the transmission route selection stage, and the transmission energy consumption accounts for a relatively low percentage of the total energy consumption, so the gap between them is also very small. If the running time and the number of running time slots are increased, the energy consumption disadvantage of the single-slot offloading algorithm compared to the proposed algorithm will gradually become prominent. Figure 6 To evaluate the impact of the number of edge computing satellites on the performance of a low-earth orbit satellite network, Figure 6 shows the trend of the total system energy consumption changing with the number of remote sensing satellites under different configurations of the number of edge satellites. It can be clearly seen from Figure 6 that as the number of available edge satellites in the system increases, the total energy consumption shows a downward trend. Specifically, first, the increase in the number of edge satellites expands the range of task offloading options, thereby improving the flexibility of the offloading process and enhancing the adaptability of the system in handling various tasks, enabling an increase in the range of the best offloading target satellites that can be selected according to task requirements and network conditions. Second, in the network topology based on the time-expanded graph, the increase in the number of edge satellite nodes introduces more offloading route selections with lower transmission costs. Therefore, by optimizing the data transmission path and reducing energy-consuming long-distance communications in route selection, the total energy consumption of the system is further reduced. Additionally, in and In both cases, the curves almost overlap, the energy consumption difference is extremely small, and the optimization and improvement effect is not significant. This may be because the system has approached its energy efficiency upper limit.

[0119] Embodiment 5: Please refer to Figure 7 As shown, the present invention also provides an electronic device 100 for a joint scheduling method of task offloading and routing planning in a low-earth orbit satellite network based on a time-expanded graph; the electronic device 100 includes a memory 101, at least one processor 102, a computer program 103 stored in the memory 101 and executable on the at least one processor 102, and at least one communication bus 104.

[0120] The memory 101 can be used to store the computer program 103. The processor 102 realizes the steps of the joint scheduling method of task offloading and routing planning in a low-earth orbit satellite network based on the time-expanded graph described in Embodiment 1 by running or executing the computer program stored in the memory 101 and calling the data stored in the memory 101. The memory 101 may mainly include a program storage area and a data storage area. Among them, the program storage area can store an operating system, application programs required for at least one function (such as a sound playback function, an image playback function, etc.); the data storage area can store data created according to the use of the electronic device 100 (such as audio data, etc.). In addition, the memory 101 may include non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, at least one magnetic disk storage device, a flash memory device, or other non-volatile solid-state storage devices.

[0121] The at least one processor 102 may be a central processing unit (CPU), or may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The processor 102 may be a microprocessor or the processor 102 may also be any conventional processor, etc. The processor 102 is the control center of the electronic device 100, and connects various parts of the entire electronic device 100 using various interfaces and lines.

[0122] The memory 101 in the electronic device 100 stores a plurality of instructions to implement a joint scheduling method for low-earth orbit satellite network task offloading and routing planning based on a time-expanded graph. The processor 102 can execute the plurality of instructions to implement the following: Construct a dynamic and time-varying low-earth orbit satellite network; Use the time-expanded graph to represent the constructed dynamic and time-varying low-earth orbit satellite network as a directed network topology graph based on a time series; Construct a joint scheduling model using the directed network topology graph based on a time series, where the joint scheduling model takes minimizing the total energy consumption as the objective function; Set the constraint conditions of the joint scheduling model; Based on the set constraint conditions of the joint scheduling model, solve the objective function of the constructed joint scheduling model to obtain an optimal result; Perform joint scheduling of task offloading and routing planning for the low-earth orbit satellite network according to the obtained optimal result to obtain a joint scheduling result of task offloading and routing planning.

[0123] Example 6: If the modules / units integrated in the electronic device 100 are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, to implement all or part of the processes in the above example methods of the present invention, it can also be completed by a computer program instructing relevant hardware. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above various method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file, or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disc, computer memory, and read-only memory (ROM, Read-Only Memory).

[0124] Those skilled in the art should understand that the embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0125] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate a means for implementing the functions specified in the flow Figure 1 one flow or multiple flows and / or blocks Figure 1 a block or multiple blocks.

[0126] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including an instruction means that implements the functions specified in the flow Figure 1 one flow or multiple flows and / or blocks Figure 1 a block or multiple blocks.

[0127] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in the flow Figure 1 one flow or multiple flows and / or blocks Figure 1 a block or multiple blocks.

[0128] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: modifications or equivalent replacements can still be made to the specific embodiments of the present invention. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention shall be covered by the protection scope of the claims of the present invention.

Claims

1. A joint scheduling method for task offloading and routing planning in a low-earth orbit satellite network based on a time-expanded graph, characterized in that, It includes the following steps: Construct a dynamic and time-varying low-earth orbit satellite network; Use a time-expanded graph to represent the constructed dynamic and time-varying low-earth orbit satellite network as a directed network topology graph based on time series; Construct a joint scheduling model using the directed network topology graph based on time series, where the joint scheduling model takes minimizing the total energy consumption as the objective function; Set the constraint conditions of the joint scheduling model; Based on the set constraint conditions of the joint scheduling model, solve the objective function of the constructed joint scheduling model to obtain the optimal result; According to the obtained optimal result, perform joint scheduling of task offloading and routing planning for the low-earth orbit satellite network to obtain the joint scheduling result of task offloading and routing planning.

2. The joint scheduling method for task offloading and routing planning in a low-earth orbit satellite network based on a time-expanded graph according to claim 1, wherein The expression of the objective function for minimizing the total energy consumption is: Among them, is a remote sensing satellite, is an edge computing satellite, is a set of time slots, is the objective function of minimizing the total energy consumption, is a time slot the energy consumption of local computing of the remote sensing satellite within the time slot, is the transmission energy consumption of the remote sensing satellite within the time slot , is a time slot the energy consumed by the edge computing satellite for computing and processing remote sensing data within the time slot, is a time slot the total energy consumption of the edge computing satellite participating in transmission relaying within the time slot.

3. The joint scheduling method for task offloading and routing planning in a low-earth orbit satellite network based on a time-expanded graph according to claim 1, characterized in that The constraint conditions of the joint scheduling model include the following formulas: Among them, k is a time slot, is a remote sensing satellite, is an edge computing satellite, is a task offloading strategy, is an offloading target selection strategy, is the total amount of data that the remote sensing satellite waits for local computing, is the maximum computing storage capacity of the remote sensing satellite, is the total amount of data that the edge computing satellite waits for computing, is the maximum computing storage capacity of the edge computing satellite, is a constant, .

4. The joint scheduling method for task offloading and routing planning in a low-earth orbit satellite network based on a time-expanded graph according to claim 1, wherein After constructing the joint scheduling model using the directed network topology graph, first decouple the objective function of the constructed joint scheduling model to obtain the decoupled objective function of the joint scheduling model, and then solve the decoupled objective function of the joint scheduling model.

5. The joint scheduling method for task offloading and routing planning in a low-earth orbit satellite network based on a time-expanded graph according to claim 4, characterized in that The specific steps for decoupling the objective function of the constructed joint scheduling model include: Construct a routing planning model; In the constructed routing planning model, use the Dijkstra algorithm to find the optimal offloading paths from each remote sensing satellite node to the edge computing satellite node at different times; Obtain the offloading path loss based on the optimal offloading paths from each remote sensing satellite node to the edge computing satellite node at different times; Multiply the path loss by the offloading data volume, and superimpose the results of multiplying the path loss by the offloading data volume to obtain the objective function of the decoupled joint scheduling model.

6. The joint scheduling method for task offloading and routing planning in a low-earth orbit satellite network based on a time-expanded graph according to claim 5, wherein, The expression of the objective function of the decoupled joint scheduling model is: Among them, is the energy consumption of local computing of the remote sensing satellite within the time slot , is the energy consumption of the edge computing satellite for computing and processing remote sensing data within the time slot , and is the total energy consumption of the overall transmission process; Total energy consumption of the entire transmission process The calculation formula is: wherein, is the offloading target selection strategy, is the amount of data that needs to be offloaded, is the basic loss of the offloading route.

7. The method for joint scheduling of low-orbit satellite network task offloading and routing planning based on a time-expanded graph according to claim 1, characterized in that: In the step of solving the objective function of the established joint scheduling model based on the set constraint conditions of the joint scheduling model to obtain the optimal result, specifically use the alternating optimization algorithm to solve the objective function of the established joint scheduling model to obtain the optimal result.

8. A joint scheduling system for task offloading and routing planning in a low-earth orbit satellite network based on a time-expanded graph, characterized in that, It includes a satellite network construction module, a directed network topology graph acquisition module, a joint scheduling model construction module, a constraint condition setting module, an objective function solving module, and a joint scheduling module; The satellite network construction module is used to construct a dynamic and time-varying low-earth orbit satellite network; The directed network topology graph acquisition module is used to represent the constructed dynamic and time-varying low-earth orbit satellite network as a directed network topology graph based on time series using a time-expanded graph; The joint scheduling model construction module is used to construct a joint scheduling model using the directed network topology graph based on time series, where the joint scheduling model takes minimizing the total energy consumption as the objective function; The constraint condition setting module is used to set the constraint conditions of the joint scheduling model; The objective function solving module is used to solve the objective function of the constructed joint scheduling model based on the set constraint conditions of the joint scheduling model to obtain the optimal result; The joint scheduling module is used to perform joint scheduling of task offloading and routing planning for the low-earth orbit satellite network according to the obtained optimal result to obtain the joint scheduling result of task offloading and routing planning.

9. An electronic device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, the steps of the joint scheduling method for task offloading and routing planning in a low-earth orbit satellite network based on a time-expanded graph according to any one of claims 1 to 7 are implemented.

10. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, the steps of the joint scheduling method for task offloading and routing planning in a low-earth orbit satellite network based on a time-expanded graph according to any one of claims 1 to 7 are implemented.

Images