A joint task offloading and power allocation method and system for satellite edge network

Abstract

The invention relates to the field of satellite communication technology, and specifically discloses a joint task offloading and power allocation method and system for a satellite edge network, wherein the method comprises: establishing a satellite task offloading model, and determining a master satellite node and a slave satellite node in a satellite constellation; the master satellite node sends a data packet, and determines whether there is a link to the slave satellite node; wherein if there is a link to the slave satellite, the slave satellite is added to a slave satellite list; if there is no communication link, the next slave satellite is continuously searched, and it is determined whether there is a communication link; according to the distance between each slave satellite and the master satellite and the computing power, a penalty dual decomposition algorithm is used to allocate computing tasks and computing resources; the master satellite transmits data to the slave satellite according to the task allocation strategy calculated by the penalty dual decomposition algorithm until the end.

Description

Technical Field

[0001] The present invention relates to the field of satellite communication technology, and in particular to a joint task offloading and power allocation method and system for a satellite edge network. Background Art

[0002] Ground network edge computing can provide low-latency services for users at the edge of the network, provide feasible solutions for new applications and services spawned by IoT technology, and provide better data privacy protection for services. However, the proportion of land is less than 20% of the global area. Due to economic factors, the coverage of ground networks will only be smaller, covering only densely populated areas. For remote areas (such as oceans, deserts, and mountains) that account for more than 80% of the global area, it is difficult to achieve coverage of ground edge computing services. In addition, ground computing equipment is vulnerable to natural disasters such as earthquakes and fires, as well as human activities such as power outages. Therefore, inspired by ground networks, using satellite networks to provide services has become a feasible solution. Onboard edge computing (OEC) introduces multi-access edge computing (MEC) into satellite networks, so that computing tasks can be processed by satellite networks. Low Earth Orbit (LEO) satellite networks show a wide range of applications in navigation, remote area network services, and disaster warning. Inspired by multi-access edge computing (MEC) technology, onboard edge computing (OEC) technology came into being. This technology deploys mobile edge computing on satellites to achieve lower service latency by taking advantage of the satellite's proximity to users. However, due to the size and power limitations of low-Earth orbit satellites, processing computationally intensive tasks with a single satellite may overload it, shortening its lifespan and causing high service latency. Satellite networks equipped with MEC are called satellite edge networks (SECN). SECN can not only provide low-latency network services to users in remote areas, but also serve as a supplement to densely populated areas when the ground network cannot meet user service requests. SECN can achieve a low-cost seamless global network and has become one of the key technologies for the future 6G (sixth generation).

[0003] SECN overcomes the problem of uneven distribution of ground computing resources and lack of network and computing services in the largest area on the earth by transferring computing from the ground to the satellite network. Since SECN can achieve seamless global coverage, users can enjoy various services provided by SECN no matter where they are in the world, including edge computing resources, wireless storage resources, etc. For users in remote areas, users do not need to apply for computing and network services from remote cloud centers, and SECN can greatly reduce user queues and transmission delays, thereby improving the user's service experience. For cloud centers and transmission backbone networks, SECN can share the processing of a considerable number of user tasks, and data does not have the ability to be transmitted through the backbone network, which reduces the work pressure of cloud centers and backbone networks. Finally, SECN processes data in satellite MEC without centralized processing, which greatly reduces the risk of data leakage and improves the security of user data.

[0004] Although SECN can improve user experience and reduce network pressure, the existing technology adopts the method of processing a task independently for each satellite. However, some tasks are too large. Large tasks handled by a single satellite are likely to cause the satellite to be overloaded, thereby reducing the service life of the satellite and causing waste of resources. Summary of the invention

[0005] To achieve the purpose of the present invention, the present application provides a joint task offloading and power allocation method for a satellite edge network, comprising:

[0006] Step S1: Establish a satellite task offloading model and determine a master satellite node and a slave satellite node in the satellite constellation;

[0007] Step S2: The master satellite node sends a data packet and determines whether there is a link to the slave satellite node;

[0008] If there is a link to the slave satellite, the slave satellite is added to the slave satellite list; if there is no communication link, the next slave satellite is searched for and it is determined whether there is a communication link;

[0009] Step S3: according to the distance between each slave satellite and the master satellite and the computing power, the penalty dual decomposition algorithm is used to allocate computing tasks and computing resources;

[0010] Step S4: The master satellite transmits data to the slave satellite according to the task allocation strategy calculated by the penalty dual decomposition algorithm until the end.

[0011] In some specific embodiments, step S1 includes:

[0012] Step S11: determining the transmission power, local computing time and offloading delay from the satellite to the main satellite in a certain time slot according to the task offloading model;

[0013] Step S12: establishing an energy consumption model according to the task offloading model to determine the energy consumption of the master satellite and the slave satellite in a certain time slot.

[0014] In some specific embodiments, the energy constraints and power constraints of the master satellite and the slave satellite are determined based on the energy consumption upper limit of the master satellite and the slave satellite to complete the data processing task among all participating low-orbit satellites and the energy consumption upper limit of any single satellite participating in the task processing.

[0015] In some specific embodiments, the offloading delay includes a transmission delay and a calculation delay, and the minimum value of the sum of the offloading delays of all time slots is obtained by determining the optimal solution of the task offloading ratio and the transmission power of each time slot.

[0016] In some specific embodiments, step S3 includes:

[0017] Step S31: Optimizing variables in the task allocation strategy by using a block coordinate descent method;

[0018] Step S32: solving the variables and projecting the obtained solutions to the feasible domain set of the variables;

[0019] If the solution falls within the feasible domain, the solution is assigned to the target variable; otherwise, the value closest to the solution in the feasible domain set is selected and assigned to the target variable;

[0020] Step S33: determining the optimization problem of the task allocation strategy according to the target variable;

[0021] Step S34: Determine a steady-state solution to the optimization problem through a constraint violation indicator function.

[0022] To achieve the same invention purpose, the present application also provides a satellite edge network joint task offloading and power allocation system, including:

[0023] Model building module: used to build a satellite task offloading model and determine the master satellite node and slave satellite node in the satellite constellation;

[0024] Link judgment module: used for the master satellite node to send data packets and determine whether there is a link to the slave satellite node;

[0025] If there is a link to the slave satellite, the slave satellite is added to the slave satellite list; if there is no communication link, the next slave satellite is searched for and it is determined whether there is a communication link;

[0026] Resource allocation module: used to allocate computing tasks and computing resources using the penalty dual decomposition algorithm according to the distance between each slave satellite and the master satellite and the computing power;

[0027] Strategy transmission module: used by the master satellite to transmit data to the slave satellite according to the task allocation strategy calculated by the penalty dual decomposition algorithm until the end.

[0028] In some specific embodiments, the model building module is used to perform the following steps:

[0029] Step S11: determining the transmission power, local computing time and offloading delay from the satellite to the main satellite in a certain time slot according to the task offloading model;

[0030] Step S12: establishing an energy consumption model according to the task offloading model to determine the energy consumption of the master satellite and the slave satellite in a certain time slot.

[0031] In some specific embodiments, the energy constraints and power constraints of the master satellite and the slave satellite are determined based on the energy consumption upper limit of the master satellite and the slave satellite to complete the data processing task among all participating low-orbit satellites and the energy consumption upper limit of any single satellite participating in the task processing.

[0032] In some specific embodiments, the offloading delay includes a transmission delay and a calculation delay, and the minimum value of the sum of the offloading delays of all time slots is obtained by determining the optimal solution of the task offloading ratio and the transmission power of each time slot.

[0033] In some specific embodiments, the policy transmission module is used to perform the following steps:

[0034] Step S31: Optimizing variables in the task allocation strategy by using a block coordinate descent method;

[0035] Step S32: solving the variables and projecting the obtained solutions to the feasible domain set of the variables;

[0036] If the solution falls within the feasible domain, the solution is assigned to the target variable; otherwise, the value closest to the solution in the feasible domain set is selected and assigned to the target variable;

[0037] Step S33: determining the optimization problem of the task allocation strategy according to the target variable;

[0038] Step S34: Determine a steady-state solution to the optimization problem through a constraint violation indicator function.

[0039] Beneficial effects of the above technical solution:

[0040] This application is based on a resource scheduling scheme of the Penalty Dual Decomposition (PDD) algorithm to minimize the maximum delay of the system when the satellite has limited power and energy. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0042] Figure 1 A schematic flow chart of a method for joint task offloading and power allocation of a satellite edge network provided by an embodiment of the present invention;

[0043] Figure 2 A schematic structural diagram of a satellite edge network joint task offloading and power distribution system provided by an embodiment of the present invention;

[0044] Figure 3 A schematic diagram of a low-orbit satellite edge computing system for a satellite edge network joint task offloading and power allocation method provided by an embodiment of the present invention;

[0045] Figure 4 A delay performance comparison diagram of various algorithms of a joint task offloading and power allocation method for a satellite edge network provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0046] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0047] Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar symbols throughout represent the same or similar elements or elements with the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present invention, and should not be construed as limiting the present invention.

[0048] Embodiment 1

[0049] An embodiment of the present invention provides a joint task offloading and power allocation method for a satellite edge network, referring to Figure 1 As shown, including:

[0050] Step S1: Establish a satellite task offloading model and determine the master satellite node and the slave satellite node in the satellite constellation.

[0051] In a specific embodiment of the present invention, step S1 includes:

[0052] Step S11: determining the transmission power, local computing time and offloading delay from the satellite to the main satellite in a certain time slot according to the task offloading model;

[0053] Step S12: establishing an energy consumption model according to the task offloading model to determine the energy consumption of the master satellite and the slave satellite in a certain time slot.

[0054] In a specific embodiment of the present invention, the offloading delay includes transmission delay and calculation delay, and the minimum value of the sum of the offloading delays of all time slots is obtained by determining the optimal solution of the task offloading ratio and the transmission power of each time slot.

[0055] Reference Figure 3 As shown, it means that the set M = {0, 1, 2, …, M} contains M + 1 LEO satellites, each satellite is equipped with an MEC server. In this case, only one master satellite generates computing tasks, and other LEO satellites within the communication range can be slave satellites s, receiving and assisting in processing computing task data. When a mobile station generates a time-sensitive task and there are not enough computing resources in the mobile station, part of the data generated by the task can be offloaded to the slave satellite. Assume that the period T is divided into N time slots, with i ∈ T = {1, 2, …, N}. Therefore, in each time slot I, the set of slave satellites is composed of k ∈ K = {1, 2, …, M}.

[0056] a) Task offloading model: The wireless channel between satellites is mainly line of sight (LOS). The channel gain from the kth slave satellite to the master satellite in the Ith time slot is given by h k，i It is expressed as follows:

[0057] (1)

[0058] in, is the channel power gain at the reference distance, and the distance between the kth slave satellite and the master satellite at the ith time slot is given by Definition Ratio of data offloaded to satellite k in time slot i.

[0059] Specifically, each satellite has the ability to participate in task offloading and computation, subject to resource constraints. During data processing, data transmission between the master and slave satellites is carried out via wireless links:

[0060] (2)

[0061] The transmission power from the master satellite to the kth slave satellite in the i-th time slot is expressed as . Expressed as the normalized transmission data rate from the master satellite to the kth slave satellite in time slot i:

[0062] (3)

[0063] in represents the additive white Gaussian noise (AWGN) power, which is assumed to be constant.

[0064] Assume that the task in the i-th time slot generates Di bits of data. The computation task remaining locally is expressed as ; C k represents the main frequency of the MEC server in the kth slave satellite, that is, the number of CPU cycles per second. C0 represents the main frequency of the MEC server in the master satellite. s represents the number of cycles of the CPU to execute one bit of data. Therefore, the local computing time It is expressed as:

[0065] (4)

[0066] The computation time from satellite k in the i-th time slot is expressed as The transmission delay from satellite k in the i-th time slot can be expressed as , B represents the bandwidth of the channel, and c represents the speed of light.

[0067] Combining the above expressions, we get the transmission delay for:

[0068] (5)

[0069] Energy consumption model: The energy consumption of the main satellite is expressed as:

[0070] (6)

[0071] Among them, r m is the effective capacitance coefficient of the chip equipped in the MEC server of the master satellite. The transmission consumption during the unloading process from the master satellite to the kth slave satellite is expressed as:

[0072] (7)

[0073] The energy consumption of the kth slave satellite is expressed as:

[0074] (8)

[0075] Similarly, r s is the effective capacitance coefficient of the kth slave satellite chip.

[0076] In a specific embodiment of the present invention, the energy constraints and power constraints of the master satellite and the slave satellite are determined based on the energy consumption upper limit of the master satellite and the slave satellite to complete the data processing task among all participating low-orbit satellites and the energy consumption upper limit of any single satellite participating in the task processing.

[0077] Specifically, E1 is represented as the upper limit of energy consumption for completing the data processing task in all participating low-orbit satellites, and E2 is the upper limit of energy consumption of any single satellite participating in the task processing. m Represents the maximum operating power of the satellite. The energy constraint and power constraint can be expressed as follows:

[0078] (9)

[0079] (10)

[0080] (11)

[0081] Among them, constraint (9) ensures that the total energy consumption is less than E1, constraint (10) indicates that the maximum operating energy consumption of the slave satellite is not higher than E2, and constraint (11) indicates that the maximum power of the master satellite is less than P m .

[0082] Step S2: The master satellite node sends a data packet and determines whether there is a link to the slave satellite node;

[0083] If there is a link to the slave satellite, the slave satellite will be added to the slave satellite list; if there is no communication link, continue to search for the next slave satellite and determine whether there is a communication link.

[0084] Step S3: According to the distance between each slave satellite and the master satellite and the computing capability, the penalty dual decomposition algorithm is used to allocate computing tasks and computing resources.

[0085] In a specific embodiment of the present invention, step S3 includes:

[0086] Step S31: Optimizing variables in the task allocation strategy by using a block coordinate descent method;

[0087] Step S32: solving the variables and projecting the obtained solutions to the feasible domain set of the variables;

[0088] If the solution falls within the feasible domain, the solution is assigned to the target variable; otherwise, the value closest to the solution in the feasible domain set is selected and assigned to the target variable;

[0089] Step S33: determining the optimization problem of the task allocation strategy according to the target variable;

[0090] Step S34: Determine a steady-state solution to the optimization problem through a constraint violation indicator function.

[0091] Specifically, by utilizing the block structure in the inner loop to optimize the main variables separately, the challenging process of optimizing all variables at the same time is avoided. In this application, a block coordinate descent method is used to solve one variable while the other variables are fixed. In each block, the first-order optimality condition is solved by setting the gradient of the objective to zero. The resulting solution is then projected onto the variable set. If the variable update value falls within the range of the variable set, the solution is the variable update value, otherwise, the variable update value should be the point in the variable set that is closest to the solution.

[0092] Once all subproblems are approximately solved, the dual variables and penalty parameters are updated in the outer loop. h ( Z n ), which measures the violation of the equation constraint in the nth cycle.

[0093] (12)

[0094] If the indicator meets Or the loop count is less than the maximum loop count N, then the AL multiplier and penalty factor Updated by the following expression:

[0095] (13)

[0096] Among them, γ is a constant between 0 and 1. It can be seen that when After n cycles of the algorithm, the violation indicator will approach some accuracy ζ, and if ζ ≤ η, the KKT solution of the problem is approximately obtained.

[0097] Step S4: The master satellite transmits data to the slave satellite according to the task allocation strategy calculated by the penalty dual decomposition algorithm until the end.

[0098] To verify the performance of the algorithm, a scenario is designed in which one master satellite and four slave satellites serve a time-sensitive mission. The distance between each slave satellite and the master satellite is randomly generated. It is assumed that all satellites have the same computing power.

[0099] The default values ​​for typical parameters are as follows:

[0100]

[0101] Reference Figure 4As shown in Figure 1, the DB-IPDD algorithm is compared with the original IPDD algorithm and the alternating optimization (AO) algorithm, which replaces the complex optimization process of a series of single variables. Under the same conditions, it can be seen that the AO algorithm lacks the ability to escape from the local optimum. The performance of the IPDD algorithm is significantly better than that of the IPDD algorithm and the AO algorithm. Therefore, the present application not only approaches the optimal point from the local gradient, but also achieves higher latency performance.

[0102] Embodiment 2

[0103] An embodiment of the present invention provides a joint task offloading and power allocation system for a satellite edge network, referring to Figure 2 As shown, including:

[0104] Model building module: used to build a satellite task offloading model and determine the master satellite node and slave satellite node in the satellite constellation;

[0105] Link judgment module: used for the master satellite node to send data packets and determine whether there is a link to the slave satellite node;

[0106] If there is a link to the slave satellite, the slave satellite is added to the slave satellite list; if there is no communication link, the next slave satellite is searched for and it is determined whether there is a communication link;

[0107] Resource allocation module: used to allocate computing tasks and computing resources using the penalty dual decomposition algorithm according to the distance between each slave satellite and the master satellite and the computing power;

[0108] Strategy transmission module: used by the master satellite to transmit data to the slave satellite according to the task allocation strategy calculated by the penalty dual decomposition algorithm until the end.

[0109] In a specific embodiment of the present invention, the model building module is used to perform the following steps:

[0110] Step S11: determining the transmission power, local computing time and offloading delay from the satellite to the main satellite in a certain time slot according to the task offloading model;

[0111] Step S12: establishing an energy consumption model according to the task offloading model to determine the energy consumption of the master satellite and the slave satellite in a certain time slot.

[0112] In a specific embodiment of the present invention, the energy constraints and power constraints of the master satellite and the slave satellite are determined based on the energy consumption upper limit of the master satellite and the slave satellite to complete the data processing task among all participating low-orbit satellites and the energy consumption upper limit of any single satellite participating in the task processing.

[0113] In a specific embodiment of the present invention, the offloading delay includes transmission delay and calculation delay, and the minimum value of the sum of the offloading delays of all time slots is obtained by determining the optimal solution of the task offloading ratio and the transmission power of each time slot.

[0114] In a specific embodiment of the present invention, the policy transmission module is used to perform the following steps:

[0115] Step S31: Optimizing variables in the task allocation strategy by using a block coordinate descent method;

[0116] Step S32: solving the variables and projecting the obtained solutions to the feasible domain set of the variables;

[0117] If the solution falls within the feasible domain, the solution is assigned to the target variable; otherwise, the value closest to the solution in the feasible domain set is selected and assigned to the target variable;

[0118] Step S33: determining the optimization problem of the task allocation strategy according to the target variable;

[0119] Step S34: Determine a steady-state solution to the optimization problem through a constraint violation indicator function.

[0120] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

[0121] Each embodiment in this specification is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referenced to each other. The embodiments of the present invention are described with reference to the flowcharts and / or block diagrams of the methods, terminal devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate instructions for implementing the functions in the process. Figure 1 A process or multiple processes and / or boxes Figure 1These computer program instructions can also be stored in a computer-readable memory that can guide a computer or other programmable data processing terminal device to work in a specific way, so that the instructions stored in the computer-readable memory produce a product including an instruction device, which implements the functions specified in the process. Figure 1 A process or multiple processes and / or boxes Figure 1 These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device, so that a series of operation steps are executed on the computer or other programmable terminal device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable terminal device provide for implementing the process in the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps of the functions specified in one or more boxes. Although the preferred embodiments of the embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they know the basic creative concept. Therefore, the attached claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the embodiments of the present invention. Finally, it should be noted that in this application, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply that there is any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, so that the process, method, article or terminal device including a series of elements includes not only those elements, but also other elements that are not explicitly listed, or also includes elements inherent to such process, method, article or terminal device. In the absence of more restrictions, the elements defined by the sentence "including one..." do not exclude the existence of other identical elements in the process, method, article or terminal device including the elements.

[0122] The method and device provided by the present invention are introduced in detail above. Specific examples are used in this application to illustrate the principle and implementation mode of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation mode and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

[0123] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", "one specific embodiment" or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0124] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A joint task offloading and power allocation method for a satellite edge network, characterized in that: include: Step S1: Establish a satellite task offloading model and determine a master satellite node and a slave satellite node in the satellite constellation; Step S2: The master satellite node sends a data packet and determines whether there is a link to the slave satellite node; if there is a link to the slave satellite, the slave satellite is added to the slave satellite list; if there is no communication link, the next slave satellite is searched and whether there is a communication link; Step S3: According to the distance between each slave satellite and the master satellite and the computing power, the penalty dual decomposition algorithm is used to allocate computing tasks and computing resources; Step S4: the master satellite transmits data to the slave satellite according to the task allocation strategy calculated by the penalty dual decomposition algorithm until the end; Step S1 includes: Step S11: determining the transmission power, local computing time and offloading delay from the satellite to the main satellite in a certain time slot according to the task offloading model; Step S12: establishing an energy consumption model according to the task offloading model to determine the energy consumption of the master satellite and the slave satellite in a certain time slot; The step S11 further includes: determining energy constraints and power constraints of the master satellite and the slave satellite according to the upper limit of energy consumption of the master satellite and the slave satellite to complete the data processing task among all participating low-orbit satellites and the upper limit of energy consumption of any single satellite participating in the task processing; Step S3 includes: Step S31: Optimizing variables in the task allocation strategy by using a block coordinate descent method; Step S32: solving the variables and projecting the obtained solutions to the feasible domain set of the variables; If the solution falls within the feasible domain, the solution is assigned to the target variable; otherwise, the value closest to the solution in the feasible domain set is selected and assigned to the target variable; Step S33: determining the optimization problem of the task allocation strategy according to the target variable; Step S34: Determine a steady-state solution to the optimization problem through a constraint violation indicator function.

2. The method for joint task offloading and power allocation of satellite edge network according to claim 1, characterized in that: The unloading delay includes transmission delay and calculation delay. The step S11 further includes: obtaining the minimum value of the sum of the unloading delays of all time slots by determining the optimal solution of the task unloading ratio and the transmission power of each time slot.

3. A joint task offloading and power allocation system for a satellite edge network, characterized in that: include: Model building module: used to build a satellite task offloading model and determine the master satellite node and slave satellite node in the satellite constellation; Link judgment module: used for the master satellite node to send data packets and determine whether there is a link to the slave satellite node; If there is a link to the slave satellite, the slave satellite is added to the slave satellite list; if there is no communication link, the next slave satellite is searched for and it is determined whether there is a communication link. Resource allocation module: used to allocate computing tasks and computing resources using the penalty dual decomposition algorithm according to the distance between each slave satellite and the master satellite and the computing power; Strategy transmission module: used by the master satellite to transmit data to the slave satellite according to the task allocation strategy calculated by the penalty dual decomposition algorithm until the end; The model building module is used to perform the following steps: Step S11: determining the transmission power, local computing time and offloading delay from the satellite to the main satellite in a certain time slot according to the task offloading model; Step S12: establishing an energy consumption model according to the task offloading model to determine the energy consumption of the master satellite and the slave satellite in a certain time slot; Determine the energy constraint and power constraint of the master satellite and the slave satellite according to the upper limit of energy consumption of the master satellite and the slave satellite to complete the data processing task among all participating low-orbit satellites and the upper limit of energy consumption of any single satellite participating in the task processing; The strategy transmission module is used to perform the following steps: Step S31: Optimizing variables in the task allocation strategy by using a block coordinate descent method; Step S32: solving the variables and projecting the obtained solutions to the feasible domain set of the variables; If the solution falls within the feasible domain, the solution is assigned to the target variable; otherwise, the value closest to the solution in the feasible domain set is selected and assigned to the target variable; Step S33: determining the optimization problem of the task allocation strategy according to the target variable; Step S34: Determine a steady-state solution to the optimization problem through a constraint violation indicator function.

4. The satellite edge network joint task offloading and power allocation system according to claim 3, characterized in that: The offloading delay includes transmission delay and calculation delay. The minimum value of the sum of the offloading delays of all time slots is obtained by determining the optimal solution of the task offloading ratio and the transmission power of each time slot.

Images