A method and system for computing offloading and power allocation in LEO satellite MEC network

By constructing a computing offload and power allocation method in the LEO satellite MEC network, and optimizing resource allocation using Lagrangian multiplication method and KKT conditions, the high latency and cost problems introduced by traditional cloud computing are solved, and ubiquitous communication is achieved and computing power improve user experience quality.

CN118611742BActive Publication Date: 2024-11-12JILIN UNIVERSITY
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411080315.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-08
Publication Date
2024-11-12
Estimated Expiration
2044-08-08

AI Technical Summary

Technical Problem

Due to the long distance, traditional mobile cloud computing has a large burden on access networks and backhaul links, and has introduced high latency, especially in more than half of the world's regions such as oceans, deserts and remote mountainous areas. It is necessary to design ubiquitous communications and real-time. Computing 6G wireless networks, especially in remote areas, are expensive to increase base station backhaul capacity, resulting in a degradation in wireless access network performance.

Method used

Design a computational offloading and power allocation method for LEO satellite MEC network. By constructing a LEO satellite MEC network model, the offloading mode and transmission power allocation are converted into convex function form using Lagrangian multiplication method and KKT conditions, and the computing resources are optimized. Allocate and spectrum resources to minimize the total overhead of task equipment in the system.

Benefits of technology

It realizes that while making full use of system resources, meets user communication needs, reduces the total overhead of task equipment in the system, improves user experience quality, and is suitable for wireless access networks in remote areas.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118611742B_ABST
    Figure CN118611742B_ABST
Patent Text Reader

Abstract

The present invention is applicable to the field of communication technology, and provides a method and system for computing offloading and power allocation in a LEO satellite MEC network. The method includes: constructing a LEO satellite MEC network model and initializing system parameters; according to the constructed model, calculating the weighted sum of user equipment delay energy consumption of local computing, LEO satellite MEC offloading computing, and cloud offloading computing, and establishing an optimization problem; according to the established optimization problem, constructing the user offloading mode selection problem as a potential game process, using the Lagrange multiplier method and KKT conditions to obtain a resource allocation plan; using a quadratic transformation to optimize the transmission power of the offloading user; alternately iterating and optimizing to obtain the best offloading decision and resource allocation plan and the best user transmission power allocation. This method can meet the communication needs of users in the system while effectively utilizing system resources to minimize the total execution overhead of task equipment in the system, thereby improving the quality of experience of users in the system.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the field of communication technology, and in particular to a method and system for computing offloading and power allocation in a LEO satellite MEC network. Background Art

[0002] Wireless communication has become an indispensable way of connection in modern society. With the commercialization of 5G and the development of 6G technology, people's expectations for communication equipment and network quality are increasing. Traditional mobile cloud computing (MCC) is far away from the remote centralized cloud, so the introduction of MCC will bring huge burdens to the access network and backhaul link of the mobile network, and will also introduce high latency problems. Therefore, Mobile Edge Computing (MEC) technology has emerged, allowing users to obtain computing services anywhere, alleviating traffic congestion on the backbone network and improving the quality of experience (QoE) of end users. However, more than half of the world's regions, especially oceans, deserts, and remote mountainous areas, still face the problem of insufficient Internet access. Therefore, extending connectivity to other populations has become a top priority for the development of future communication networks. Recently, enhanced on-board processing capabilities are transforming the role of satellites from transmission relays to on-orbit computing, which allows tasks on satellites to be processed directly, preventing long propagation delays from satellites to ground data centers and congestion of limited bandwidth downlinks, providing remote areas with ubiquitous, broadband and reliable computing services.

[0003] Therefore, it is necessary to design the sixth generation (6G) wireless network with ubiquitous communication and real-time computing capabilities. On the one hand, for special areas that cannot be covered by ground communication, satellite communication can be used as a supplement to provide access services. Compared with ground communication, satellite communication has the characteristics of large coverage area and more flexible deployment due to its geographical advantages, and is widely regarded as one of the typical application scenarios of 6G wireless networks. In particular, low earth orbit (LEO) satellites with low orbital altitudes can provide efficient communication due to low transmission delay and propagation path loss. On the other hand, due to the complex natural environment and long-distance construction, it is economically costly to increase the ground backhaul capacity of existing base stations (BS) in remote areas, resulting in the degradation of wireless access network performance. In this context, LEO satellites equipped with MEC servers can be regarded as space network nodes to provide additional computing capabilities. Therefore, an effective MEC satellite-ground collaborative computing network system must consider the limitations of power consumption and latency, and optimize and coordinate spectrum resources, computing resources, and storage resources. Summary of the invention

[0004] The object of the present invention is to provide a method and system for computing offloading and power allocation in a LEO satellite MEC network, aiming to solve the problems raised in the above-mentioned background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] A method for computing offloading and power allocation in a LEO satellite MEC network comprises the following steps:

[0007] Step S101: constructing a LEO satellite MEC network model, including mission equipment, a LEO satellite MEC server and a cloud server; initializing system parameters, wherein the system parameters include the mission data volume and computing power of the mission equipment, and the computing power of the LEO satellite MEC server and the cloud server;

[0008] Step S102: According to the system model constructed in step S101, the weighted sum of the delay energy consumption of local computing, LEO satellite MEC offloading computing and cloud offloading computing is obtained, and a specific optimization problem is established with the goal of minimizing the total overhead of the mission equipment in the system;

[0009] Step S103: According to the optimization problem established in step S102, the offloading mode selection problem is constructed as a potential game process, and the offloading mode of a mission device is updated in each iteration; for the mission device that selects the LEO satellite MEC offloading mode, the computing resource allocation scheme is obtained by using the Lagrange multiplier method and the KKT condition;

[0010] Step S104: According to the offloading mode obtained in step S103, for the mission equipment that selects the LEO satellite MEC offloading mode and the cloud offloading mode, the transmission power allocation problem is converted into a subtraction form of two convex functions by a quadratic transformation method for solution;

[0011] Step S105: Repeat the iterative execution of step S103 and step S104 until convergence or reaching the maximum number of iterations, and find the best offloading mode, the best computing resource allocation, and the best transmission power allocation.

[0012] Furthermore, the step S101 specifically includes:

[0013] Step S1011: In the LEO satellite MEC network model, there are several mission devices within the LEO satellite coverage area. Ground mission devices with certain computing capabilities can choose to process computing tasks locally, offload all computing tasks to the LEO satellite for edge computing, and further offload computing tasks to the cloud server for processing through LEO satellite backhaul transmission;

[0014] Step S1012: define the task set as , and each task device has a computing task to perform ,in Indicates the data size of the computing task, in bytes. Indicates the amount of computing resources required for the computing task, that is, the number of CPU cycles;

[0015] Step S1013: Definition The selectable offloading mode for the task device; define the mode selection factor as ,in ,when When the value is 1, it means that the task device will calculate the task locally, on the LEO satellite MEC server, or on the cloud server.

[0016] Furthermore, the step S102 specifically includes:

[0017] Step S1021: The task equipment in the satellite coverage area transmits the computing task to the LEO satellite via the Ka-band using orthogonal frequency division multiplexing. The uplink transmission rate of the satellite-to-ground link is:

[0018] ;

[0019] in, p i Indicates the task device i The transmission power, B is the uplink bandwidth of the satellite-to-ground link, is the channel gain between the mission equipment and the LEO satellite, is the background noise power;

[0020] Step S1022: When the task device chooses to perform calculation locally, ; The execution delay required for local calculation of the task is:

[0021] ;

[0022] The energy consumption required for local computing is:

[0023] ;

[0024] Among them, the definition Equipment for the mission local computing power, is the energy consumption coefficient of the chip; the user overhead for local computing is: ,in and , and Respectively represent the delay and energy consumption of the task equipment i The relative weight of

[0025] Step S1023: When the task device chooses to offload the calculation on the LEO satellite MEC server, ; The upload delay when the task is offloaded is:

[0026] ;

[0027] The propagation delay of the mission offloaded to the LEO satellite is:

[0028] ;

[0029] in, Indicates the task device i The distance from the LEO satellite, c is the speed of light; the computing resources allocated by the LEO satellite MEC server to each offload task are defined as , then the satellite edge computing task delay is:

[0030] ;

[0031] The task is offloaded to the LEO satellite MEC server and the total delay required is calculated as:

[0032] ;

[0033] For task equipment offloading energy consumption, it is expressed as transmission energy consumption ; The user cost when choosing to offload the calculation on the LEO satellite MEC server is: ;

[0034] Step S1024: When the task device selects a cloud server to offload computing, ; After uploading the task to the LEO satellite, it is transmitted to the cloud server through the backhaul for calculation; the upload delay when the task is offloaded is:

[0035] ;

[0036] The propagation delay of the mission offloaded to the LEO satellite is:

[0037] ;

[0038] The propagation delay from LEO satellite to cloud server is

[0039] ;

[0040] in, Indicates the distance between the LEO satellite and the cloud server, c is the speed of light; assuming that the delay of computing tasks transmitted from the core network to the cloud server is constant, it is defined as ; The computing resources allocated by the cloud server to each offload task are defined as , then the cloud server processing delay is:

[0041] ;

[0042] The total delay required to offload the task to the cloud server is:

[0043] ;

[0044] For task equipment offloading energy consumption, it is expressed as transmission energy consumption ; The user cost when choosing to offload computing to the cloud server is: ;

[0045] Step S1025: Establish the optimization problem as:

[0046] ;

[0047] ;

[0048] in, The maximum computing resource for the LEO satellite MEC server, is the maximum transmission power of the mission equipment, The corresponding overhead when selecting different offloading modes for each task device.

[0049] Furthermore, in step S103, the selection of the offloading mode is constructed as a potential game process, and the specific steps of obtaining the offloading decision and resource allocation scheme include:

[0050] Step S1031: There are three unloading modes for the mission equipment in the LEO satellite coverage area in the system model. When the unloading mode of other mission equipment in the area within the system is given, When the task equipment i The offloading mode is selected with the goal of minimizing the user overhead. Under the given task equipment's transmit power, the offloading mode selection problem is expressed as: ; Build the uninstall mode selection into a game ,in N Representation Game G Participants, i.e., mission equipment within the LEO satellite coverage area of ​​the system; for The strategy space, i.e., the uninstallation mode; V i represents the user cost function of the participants;

[0051] Step S1032: After the unloading strategy and power selection of the tasks in the system are determined, the optimization problem for the task equipment that selects the LEO satellite MEC unloading mode is expressed as follows:

[0052] ;

[0053] ;

[0054] Construct the Lagrangian function of the above optimization problem:

[0055] ;

[0056] in, β represents a non-negative Lagrange multiplier;

[0057] And obtain the mission equipment through KKT conditions i Computational resource allocation for determining MEC offloading mode and transmit power for LEO satellites .

[0058] Furthermore, in step S104, the specific steps of solving the task equipment transmission power allocation are as follows:

[0059] For the task devices in the system, when the offloading mode selection and resource allocation scheme are given, the optimization problem is expressed as:

[0060] ;

[0061] ;

[0062] The optimization problem is equivalent to the following through the quadratic transformation method:

[0063] ;

[0064] ;

[0065] ;

[0066] in, is an auxiliary variable introduced; in At the iteration, the calculation formula is:

[0067] ;

[0068] in, Y i It is the subtraction form of two convex functions. By transforming the original optimization problem into a convex difference problem, the transformed power allocation optimization problem is solved to obtain the task equipment transmission power under the given offloading mode selection and resource allocation. pi .

[0069] Furthermore, in step S105, by solving the sub-problems in step S103 and step S104 respectively, step S103 and step S104 are alternately optimized until convergence or the maximum number of iterations is reached, and the optimal offloading mode, optimal computing resource allocation and optimal transmission power allocation are obtained.

[0070] A computing offloading and power distribution system in a LEO satellite MEC network, comprising:

[0071] The first construction module is used to construct a LEO satellite MEC network model, including mission equipment, a LEO satellite MEC server and a cloud server; initialize system parameters, wherein the system parameters include the mission data volume and computing power of the mission equipment, and the computing power of the LEO satellite MEC server and the cloud server;

[0072] The second building module is used to obtain the weighted sum of the delay energy consumption of local computing, LEO satellite MEC offloading computing and cloud offloading computing according to the constructed system model, and establish a specific optimization problem with the goal of minimizing the total overhead of the mission equipment in the system;

[0073] The first conversion module is used to construct the offloading mode selection problem as a potential game process according to the established optimization problem, and update the offloading mode of a mission device in each iteration; for the mission device that selects the LEO satellite MEC offloading mode, the computing resource allocation scheme is obtained by using the Lagrange multiplier method and the KKT condition;

[0074] The second conversion module is used to convert the transmission power allocation problem into a subtraction form of two convex functions for solving the problem through a quadratic transformation method according to the obtained unloading mode and for the mission equipment that selects the LEO satellite MEC unloading mode and the cloud unloading mode;

[0075] The iterative calculation module is used to repeatedly iterate the construction and solution process until convergence or the maximum number of iterations is reached, and to find the best offloading mode, the best computing resource allocation, and the best transmission power allocation.

[0076] Compared with the prior art, the present invention has the following beneficial effects:

[0077] The method provided by the present invention can, on the one hand, meet the communication needs of users in the system while making full use of system resources, and on the other hand, can also minimize the total execution overhead of task equipment in the system and improve the experience quality of users in the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] Figure 1 It is a schematic diagram of the LEO satellite MEC network model in the present invention.

[0079] Figure 2 Schematic diagram of the method of the present invention.

[0080] Figure 3 Schematic diagram of the system in the present invention. DETAILED DESCRIPTION

[0081] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0082] The specific implementation of the present invention is described in detail below in conjunction with specific embodiments.

[0083] Embodiment 1, as Figure 1 and Figure 2 As shown, a method for computing offloading and power allocation in a LEO satellite MEC network provided by an embodiment of the present invention includes the following steps:

[0084] Step S101: constructing a LEO satellite MEC network model, including mission equipment, a LEO satellite MEC server and a cloud server; initializing system parameters, wherein the system parameters include the mission data volume and computing power of the mission equipment, and the computing power of the LEO satellite MEC server and the cloud server;

[0085] The step S101 specifically includes:

[0086] Step S1011: In the LEO satellite MEC network model, there are several mission devices within the LEO satellite coverage area. Ground mission devices with certain computing capabilities can choose to process computing tasks locally, offload all computing tasks to the LEO satellite for edge computing, and further offload computing tasks to the cloud server for processing through LEO satellite backhaul transmission;

[0087] Step S1012: define the task set as , and each task device has a computing task to perform ,in Indicates the data size of the computing task, in bytes. Indicates the amount of computing resources required for the computing task, that is, the number of CPU cycles;

[0088] Step S1013: Definition The selectable offloading mode for the task device; define the mode selection factor as ,in ,when When the value is 1, it means that the task device will calculate the task locally, on the LEO satellite MEC server, or on the cloud server.

[0089] Step S102: According to the system model constructed in step S101, the weighted sum of the delay energy consumption of local computing, LEO satellite MEC offloading computing and cloud offloading computing is obtained, and a specific optimization problem is established with the goal of minimizing the total overhead of the mission equipment in the system;

[0090] The step S102 specifically includes:

[0091] Step S1021: The task equipment in the satellite coverage area transmits the computing task to the LEO satellite via the Ka-band using orthogonal frequency division multiplexing. The uplink transmission rate of the satellite-to-ground link is:

[0092] ;

[0093] in, Indicates the task device The transmission power, B is the uplink bandwidth of the satellite-to-ground link, is the channel gain between the mission equipment and the LEO satellite, is the background noise power;

[0094] Step S1022: When the task device chooses to perform calculation locally, ; The execution delay required for local calculation of the task is:

[0095] ;

[0096] The energy consumption required for local computing is:

[0097] ;

[0098] Among them, the definition Equipment for the mission local computing power, is the energy consumption coefficient of the chip; the user overhead for local computing is: ,in and , and Respectively represent the delay and energy consumption of the task equipment The relative weight of

[0099] Step S1023: When the task device chooses to offload the calculation on the LEO satellite MEC server, ; The upload delay when the task is offloaded is:

[0100] ;

[0101] The propagation delay of the mission offloaded to the LEO satellite is:

[0102] ;

[0103] in, Indicates the task device The distance from the LEO satellite, is the speed of light; the computing resources allocated by the LEO satellite MEC server to each offload task are defined as , then the satellite edge computing task delay is:

[0104] ;

[0105] The task is offloaded to the LEO satellite MEC server and the total delay required is calculated as:

[0106] ;

[0107] For task equipment offloading energy consumption, it is expressed as transmission energy consumption ; The user cost when choosing to offload the calculation on the LEO satellite MEC server is: ;

[0108] Step S1024: When the task device selects a cloud server to offload computing, ; After uploading the task to the LEO satellite, it is transmitted to the cloud server through the backhaul for calculation; the upload delay when the task is offloaded is:

[0109] ;

[0110] The propagation delay of the mission offloaded to the LEO satellite is:

[0111] ;

[0112] The propagation delay from LEO satellite to cloud server is

[0113] ;

[0114] in, Indicates the distance between the LEO satellite and the cloud server, is the speed of light; assuming that the delay of computing tasks transmitted from the core network to the cloud server is constant, it is defined as ; The computing resources allocated by the cloud server to each offload task are defined as , then the cloud server processing delay is:

[0115] ;

[0116] The total delay required to offload the task to the cloud server is:

[0117] ;

[0118] For task equipment offloading energy consumption, it is expressed as transmission energy consumption ; The user cost when choosing to offload computing to the cloud server is: ;

[0119] Step S1025: Establish the optimization problem as:

[0120] ;

[0121] ;

[0122] in, The maximum computing resource for the LEO satellite MEC server, is the maximum transmission power of the mission equipment, The corresponding overhead when selecting different offloading modes for each task device.

[0123] Step S103: According to the optimization problem established in step S102, the offloading mode selection problem is constructed as a potential game process, and the offloading mode of a mission device is updated in each iteration; for the mission device that selects the LEO satellite MEC offloading mode, the computing resource allocation scheme is obtained by using the Lagrange multiplier method and the KKT condition;

[0124] In step S103, the selection of the offloading mode is constructed as a potential game process, and the specific steps of obtaining the offloading decision and resource allocation scheme include:

[0125] Step S1031: There are three unloading modes for the mission equipment in the LEO satellite coverage area in the system model. When the unloading mode of other mission equipment in the area within the system is given, When the task equipment The offloading mode is selected with the goal of minimizing the user overhead. Under the given task equipment's transmit power, the offloading mode selection problem is expressed as: ; Build the uninstall mode selection into a game ,in N Representation Game G Participants, i.e., mission equipment within the LEO satellite coverage area of ​​the system; for The strategy space, i.e., the uninstallation mode; represents the user cost function of the participants;

[0126] Step S1032: After the unloading strategy and power selection of the tasks in the system are determined, the optimization problem for the task equipment that selects the LEO satellite MEC unloading mode is expressed as follows:

[0127] ;

[0128] ;

[0129] Construct the Lagrangian function of the above optimization problem:

[0130] ;

[0131] in, β represents a non-negative Lagrange multiplier;

[0132] And obtain the mission equipment through KKT conditions Computational resource allocation for determining MEC offloading mode and transmit power for LEO satellites .

[0133] Step S104: According to the offloading mode obtained in step S103, for the mission equipment that selects the LEO satellite MEC offloading mode and the cloud offloading mode, the transmission power allocation problem is converted into a subtraction form of two convex functions by a quadratic transformation method for solution;

[0134] In step S104, the specific steps of solving the task equipment transmission power allocation are as follows:

[0135] For the task devices in the system, when the offloading mode selection and resource allocation scheme are given, the optimization problem is expressed as:

[0136] ;

[0137] ;

[0138] The optimization problem is equivalent to the following through the quadratic transformation method:

[0139] ;

[0140] ;

[0141] ;

[0142] in, is an auxiliary variable introduced; in At the iteration, the calculation formula is:

[0143] ;

[0144] in, It is the subtraction form of two convex functions. By transforming the original optimization problem into a convex difference problem, the transformed power allocation optimization problem is solved to obtain the task equipment transmission power under the given offloading mode selection and resource allocation. .

[0145] Step S105: Repeat the iterative execution of step S103 and step S104 until convergence or reaching the maximum number of iterations, and find the best offloading mode, the best computing resource allocation, and the best transmission power allocation.

[0146] In step S105, by solving the sub-problems in step S103 and step S104 respectively, step S103 and step S104 are alternately optimized until convergence or the maximum number of iterations is reached, and the optimal offloading mode, optimal computing resource allocation and optimal transmission power allocation are obtained.

[0147] In an embodiment of the present invention, a LEO satellite MEC network model is first constructed. On the basis of local calculation, two unloading calculation modes, namely, LEO satellite MEC unloading calculation and cloud unloading calculation, are introduced. Different unloading modes assist each other to improve the experience quality of users in the system. By obtaining the weighted sum of the delay energy consumption of local calculation, LEO satellite MEC unloading calculation and cloud unloading calculation, an optimization problem is constructed with the goal of minimizing the total overhead of task equipment in the system. A potential game process is constructed to select the unloading mode for the task equipment in the system, and the computing resources in the system are allocated using the Lagrange multiplier method and the KKT condition. Then, the obtained unloading decision and resource allocation scheme are brought into the iteration of solving the transmission power allocation of the task equipment using the quadratic transformation. By alternately optimizing the unloading decision, resource allocation and transmission power allocation, the iterative execution is repeated until convergence or the maximum number of iterations, so that the result is optimal. The method proposed in the present invention can meet the communication needs of users in the system while making full use of system resources, and can also minimize the total execution overhead of task equipment in the system and improve the experience quality of users in the system.

[0148] Embodiment 2, as Figure 3 As shown, a computing offloading and power distribution system in a LEO satellite MEC network provided by an embodiment of the present invention includes:

[0149] The first construction module is used to construct a LEO satellite MEC network model, including mission equipment, a LEO satellite MEC server and a cloud server; initialize system parameters, wherein the system parameters include the mission data volume and computing power of the mission equipment, and the computing power of the LEO satellite MEC server and the cloud server;

[0150] The second building module is used to obtain the weighted sum of the delay energy consumption of local computing, LEO satellite MEC offloading computing and cloud offloading computing according to the constructed system model, and establish a specific optimization problem with the goal of minimizing the total overhead of the mission equipment in the system;

[0151] The first conversion module is used to construct the offloading mode selection problem as a potential game process according to the established optimization problem, and update the offloading mode of a mission device in each iteration; for the mission device that selects the LEO satellite MEC offloading mode, the computing resource allocation scheme is obtained by using the Lagrange multiplier method and the KKT condition;

[0152] The second conversion module is used to convert the transmission power allocation problem into a subtraction form of two convex functions for solving the problem through a quadratic transformation method according to the obtained unloading mode and for the mission equipment that selects the LEO satellite MEC unloading mode and the cloud unloading mode;

[0153] The iterative calculation module is used to repeatedly iterate the construction and solution process until convergence or the maximum number of iterations is reached, and to find the best offloading mode, the best computing resource allocation, and the best transmission power allocation.

[0154] The specific implementation process of the functions realized by each module in Example 2 is the same as the implementation process of each step in Example 1, and will not be repeated here.

[0155] The above are only preferred embodiments of the present invention. It should be pointed out that, for those skilled in the art, several modifications and improvements can be made without departing from the concept of the present invention. These should also be regarded as the protection scope of the present invention. These will not affect the effect of the implementation of the present invention and the practicality of the patent.

Claims

1. A method for computing offloading and power allocation in a LEO satellite MEC network, characterized in that: The following steps are involved: Step S101: constructing a LEO satellite MEC network model, including mission equipment, LEO satellite MEC server and cloud server; Initialize system parameters, including the mission data volume and computing power of the mission equipment, and the computing power of the LEO satellite MEC server and cloud server; Step S102: According to the system model constructed in step S101, the weighted sum of the delay energy consumption of local computing, LEO satellite MEC offloading computing and cloud offloading computing is obtained, and a specific optimization problem is established with the goal of minimizing the total overhead of the mission equipment in the system; Step S103: According to the optimization problem established in step S102, the offloading mode selection problem is constructed as a potential game process, and the offloading mode of a mission device is updated in each iteration; for the mission device that selects the LEO satellite MEC offloading mode, the computing resource allocation scheme is obtained by using the Lagrange multiplier method and the KKT condition; Step S104: According to the offloading mode obtained in step S103, for the mission equipment that selects the LEO satellite MEC offloading mode and the cloud offloading mode, the transmission power allocation problem is converted into a subtraction form of two convex functions by a quadratic transformation method for solution; Step S105: Repeat the iteration of step S103 and step S104 until convergence or the maximum number of iterations is reached, and find the best offloading mode, the best computing resource allocation, and the best transmission power allocation; The step S101 specifically includes: Step S1011: In the LEO satellite MEC network model, there are several mission devices within the LEO satellite coverage area. Ground mission devices with certain computing capabilities can choose to process computing tasks locally, offload all computing tasks to the LEO satellite for edge computing, and further offload computing tasks to the cloud server for processing through LEO satellite backhaul transmission; Step S1012: define a task set as I = {1, 2, ... N}, and each task device has a computing task T to be executed i = {W i ,C i }(i∈N), where W i Indicates the data size of the computing task in bytes, C i Indicates the amount of computing resources required for the computing task, that is, the number of CPU cycles; Step S1013: define O = {L, LEO, C} as the selectable unloading mode for the task device; define the mode selection factor as in when When the value is 1, it means that the task equipment will calculate the task locally, on the LEO satellite MEC server, or on the cloud server; The step S102 specifically includes: Step S1021: The task equipment in the satellite coverage area transmits the computing task to the LEO satellite via the Ka-band using orthogonal frequency division multiplexing. The uplink transmission rate of the satellite-to-ground link is: Among them, p i represents the transmission power of mission equipment i, B is the uplink bandwidth of the satellite-to-ground link, is the channel gain between the mission equipment and the LEO satellite, σ 2 is the background noise power; Step S1022: When the task device chooses to perform calculation locally, The execution delay required for local calculation of the task is: The energy consumption required for local computing is: In which, define f i L is the local computing capability of task device i, κ is the energy consumption coefficient of the chip; the user overhead during local computing is: in and and They represent the relative weights of delay and energy consumption to task device i respectively; Step S1023: When the task device chooses to offload the calculation on the LEO satellite MEC server, The upload delay when the task is offloaded is: The propagation delay of the mission offloaded to the LEO satellite is: in, represents the distance between the mission device i and the LEO satellite, c is the speed of light; the computing resources allocated by the LEO satellite MEC server to each offloaded task are defined as f i m , then the satellite edge computing task delay is: The task is offloaded to the LEO satellite MEC server and the total delay required is calculated as: For task equipment offloading energy consumption, it is expressed as transmission energy consumption The user cost when choosing to offload the calculation on the LEO satellite MEC server is: Step S1024: When the task device selects a cloud server to offload computing, After uploading the task to the LEO satellite, it is transmitted to the cloud server for calculation via the backhaul. The upload delay when the task is offloaded is: The propagation delay of the mission offloaded to the LEO satellite is: The propagation delay from LEO satellite to cloud server is: Among them, d S,C represents the distance between the LEO satellite and the cloud server, c is the speed of light; assuming that the delay of the computing task transmitted from the core network to the cloud server is a constant, defined as t cloud ; The computing resources allocated by the cloud server to each offload task are defined as f i C , then the cloud server processing delay is: The total delay required to offload the task to the cloud server is: For task equipment offloading energy consumption, it is expressed as transmission energy consumption The user cost when choosing to offload computing to the cloud server is: Step S1025: Establish the optimization problem as: Among them, F LEO is the maximum computing resource of the LEO satellite MEC server, P i,max is the maximum transmission power of the mission equipment, V i o The corresponding overhead when selecting different offloading modes for each task device; In step S103, the selection of the offloading mode is constructed as a potential game process, and the specific steps of obtaining the offloading decision and resource allocation scheme include: Step S1031: There are three unloading modes for the mission equipment in the LEO satellite coverage area in the system model. When the unloading mode of other mission equipment in the area within the system is given, When , task device i selects the offloading mode with the goal of minimizing its own user overhead; under the given task device's transmit power, the offloading mode selection problem is expressed as: Framing uninstall mode selection as a game Where N represents the participants in the game G, i.e., the mission equipment within the LEO satellite coverage area of ​​the system; is the strategy space of i∈N, i.e., the unloading mode; V i represents the user cost function of the participants; Step S1032: After the unloading strategy and power selection of the tasks in the system are determined, the optimization problem for the task equipment that selects the LEO satellite MEC unloading mode is expressed as follows: Construct the Lagrangian function of the above optimization problem: Where β represents a non-negative Lagrange multiplier; The KKT condition is used to obtain the computing resource allocation f of mission equipment i under the determined LEO satellite MEC offloading mode and transmission power. i m ; In step S104, the specific steps of solving the task equipment transmission power allocation are as follows: For the task devices in the system, when the offloading mode selection and resource allocation scheme are given, the optimization problem is expressed as: The optimization problem is equivalent to the following through the quadratic transformation method: Among them, y i is an auxiliary variable introduced; at the qth iteration, the calculation formula is: Among them, Y i It is the subtraction form of two convex functions. By transforming the original optimization problem into a convex difference problem, the transformed power allocation optimization problem is solved to obtain the task equipment transmission power p under the given offloading mode selection and resource allocation. i .

2. The method according to claim 1, characterized in that In step S105, by solving the sub-problems in step S103 and step S104 respectively, step S103 and step S104 are alternately optimized until convergence or the maximum number of iterations is reached, and the optimal offloading mode, optimal computing resource allocation and optimal transmission power allocation are obtained.

3. A computational offloading and power distribution system in a LEO satellite MEC network, the system being used to execute the method of claim 1 or 2, characterized in that: include: The first building module is used to build a LEO satellite MEC network model, including mission equipment, LEO satellite MEC server and cloud server; Initialize system parameters, including the mission data volume and computing power of the mission equipment, and the computing power of the LEO satellite MEC server and cloud server; The second building module is used to obtain the weighted sum of the delay energy consumption of local computing, LEO satellite MEC offloading computing and cloud offloading computing according to the constructed system model, and establish a specific optimization problem with the goal of minimizing the total overhead of the mission equipment in the system; The first conversion module is used to construct the offloading mode selection problem as a potential game process according to the established optimization problem, and update the offloading mode of a mission device in each iteration; for the mission device that selects the LEO satellite MEC offloading mode, the computing resource allocation scheme is obtained by using the Lagrange multiplier method and the KKT condition; The second conversion module is used to convert the transmission power allocation problem into a subtraction form of two convex functions for solving the problem through a quadratic transformation method according to the obtained unloading mode and for the mission equipment that selects the LEO satellite MEC unloading mode and the cloud unloading mode; The iterative calculation module is used to repeatedly iterate the construction and solution process until convergence or the maximum number of iterations is reached, and to find the best offloading mode, the best computing resource allocation, and the best transmission power allocation.