A resource allocation method for intensive tasks

Abstract

This invention relates to a resource allocation method for intensive tasks, belonging to the field of wireless communication technology. The invention collects intensive tasks from ground terminal equipment and processes them on an high-altitude platform or transmits them to low-Earth orbit satellites, constructing a weighted sum mathematical model of total system energy consumption. By establishing a bidirectional preference relationship between ground terminal equipment and the high-altitude platform, and an association between the high-altitude platform and the low-Earth orbit satellite, the optimal path selection is determined. The weighted sum mathematical model of total system energy consumption is decoupled into three sub-problems: high-altitude platform selection association problem, task offloading, and resource allocation, which are then solved. The results of high-altitude platform selection association, task offloading decision, and resource allocation are jointly iteratively optimized, using the weighted sum of total system energy consumption obtained in each iteration as the fitness evaluation criterion, until convergence to the optimal weighted sum of total system energy consumption. This invention proposes a highly energy-efficient hybrid offloading strategy, which not only allocates resources efficiently and flexibly but also effectively improves system energy efficiency.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of communication and relates to a resource allocation method for dense task-oriented space-air-ground integrated networks. BACKGROUND

[0002] With the rapid development of global informatization, the fifth generation communication system (5G) has gradually matured and is widely used in various fields, and the research on the sixth generation communication system (6G) is also actively promoted. Under the development of the 6G communication system, the communication and computing demands between terminal devices have increased dramatically, prompting the significant development of Mobile Edge Computing (MEC) technology. Compared with the single MEC server deployment in the ground network, the deployment of MEC servers in the Space-Air-Ground Integrated Network (SAGIN) is more diversified. According to different network requirements, MEC servers can be deployed on Low Earth Orbit (LEO) satellites or High Altitude Platforms (HAPs) to provide more flexible and efficient computing services. Although SAGIN has advantages such as wide coverage, high throughput, and strong flexibility, its high dynamicity and complex heterogeneous environment also bring significant technical challenges. For example, frequent switching between nodes and users, long-distance transmission of satellite communication, dynamic connection between HAPs and LEOs, and optimization of computing resources and energy consumption in a multi-layer network structure are key problems that need to be studied in depth. In addition, ground wireless networks are often limited by energy and computing resources, making it difficult to meet the growing computing demands of Internet of Things devices. Therefore, in order to fully utilize the advantages of SAGIN and cope with the multiple challenges in dense task scenarios, the resource allocation method for space-air-ground integrated dense tasks proposed in this paper can dynamically adjust the association relationship and reasonably allocate resources to minimize system energy consumption and guarantee service quality, effectively solving the resource allocation challenges of SAGIN in complex environments, which has important research significance for building an efficient, stable, and low-energy space-air-ground integrated network. SUMMARY

[0003] Therefore, the purpose of the present application is to provide a resource allocation method for dense task-oriented space-air-ground integrated networks, aiming to solve the problems raised in the background.

[0004] To achieve the above-mentioned purpose, the present application provides the following technical solutions:

[0005] A resource allocation method for dense tasks, comprising the following steps:

[0006] S1. The intensive tasks of the ground terminal device are collected to the high altitude platform (HAP) or the task is forwarded by the HAP to the low earth orbit (LEO) satellite for processing, and a system total energy consumption weighted sum mathematical model is constructed;

[0007] S2. By constructing the bidirectional preference relationship between the ground terminal device and the HAP and the association between the HAP and the LEO, the optimal path selection is determined; the system total energy consumption weighted sum mathematical model is decoupled into three sub-problems of HAP selection association, task offloading decision and resource allocation for solving respectively;

[0008] S3. The solving results of the three sub-problems of HAP selection association, task offloading decision and resource allocation are jointly iterated and optimized, and the system total energy consumption weighted sum obtained in each iteration is taken as the fitness evaluation standard, until the optimal system total energy consumption weighted sum is converged.

[0009] Further, the HAP acts as a base station and a relay node, and the system total energy consumption weighted sum mathematical model is constructed under the maximum delay constraint, and is expressed as:

[0010]

[0011] Wherein, E U,H and E H,L respectively represent the transmission energy consumption of the terminal device to the HAP and the transmission energy consumption of the HAP to the LEO, and α1 and α2 are weight coefficients, respectively measuring the different contributions of the transmission energy consumption from the terminal device to the HAP and the transmission energy consumption of the HAP to the LEO to the optimization target; E H represents the HAP computing energy consumption; E L represents the LEO computing energy consumption; β and γ are respectively constant positive weights of the HAP and LEO transmission energy consumption; H represents H HAPs; h represents the HAP index, from 1 to H; t represents the time slot index, from 1 to T; o represents the HAP offloading decision constraint: P represents the maximum power constraint of the terminal device and the HAP; represents whether the task is processed locally by the HAP; represents whether the task is offloaded to the LEO for processing; P represents the maximum power constraint set of the terminal device and the HAP; represents that the power of the path X should not exceed the maximum power constraint; U represents U terminal devices; u represents the terminal device index, from 1 to U; S represents S LEOs; s represents the LEO index, from 1 to S; represents the bit size of the current computing task; represents the computing resource possessed by the current path X; F represents the maximum computing resource constraint of the HAP and the LEO; d u represents the actual completion delay of the task; C1 represents that the terminal device can only select one task offloading mode in one time slot; C2 represents that the power of the terminal device and the HAP does not exceed the maximum power limit; C3 represents that the resources used by the HAP and the LEO do not exceed the maximum resource constraint; C4 represents the maximum tolerable time delay constraint of task processing;

[0012] The association constraint is represented as:

[0013]

[0014] Wherein, represents whether the terminal device u is associated with the HAP index at time slot t; represents whether the HAP index is associated with the LEO index at time slot t;

[0015] The maximum power constraint P of the terminal device and the HAP is represented as:

[0016]

[0017] Wherein, represents the maximum power of the terminal device; represents the maximum power of the HAP;

[0018] o is represented as:

[0019]

[0020] Wherein, When the task will be directly processed locally by the HAP; when the task will be indirectly processed by the HAP to the LEO;

[0021] F is represented as:

[0022]

[0023] Wherein, represents the maximum available computing resource of the HAP; represents the maximum available computing resource of the LEO;

[0024] The terminal device to HAP and HAP to LEO path set X is represented as:

[0025] X={(i,j)|i∈U∪H,j∈H∪S}.

[0026] Further, the is represented as:

[0027]

[0028] The is denoted as:

[0029]

[0030] where u t denotes the terminal device u that needs to process a task in time slot t; h t denotes the HAP index participating in processing a task in time slot t; s t denotes the LEO index participating in processing a task in time slot t;

[0031] Each terminal device can only be associated with one HAP, denoted as:

[0032]

[0033] Each HAP can only be associated with one satellite, denoted as:

[0034]

[0035] Each HAP has a limited load, satisfying:

[0036]

[0037] where N H is the maximum number of associations of HAPs.

[0038] Further, the S2 is specifically:

[0039] The preference list P of terminal devices on HAP u,h is denoted as:

[0040]

[0041] where φ u is denoted as the task priority weight; is denoted as the transmission rate of terminal device to HAP; is denoted as the available computing resources of HAP; t denotes the time slot index, from 1 to T; U denotes the number of terminal devices with U; u denotes the terminal device index, from 1 to U; u t denotes the terminal device u that needs to process a task in time slot t; U t denotes the set of all terminal devices that need to process a task in time slot t;

[0042] The HAP preference list P of terminal devices h,u is denoted as:

[0043]

[0044] where, is denoted as the bit size of the current computing task; ht denotes the HAPs participating in processing tasks in time slot t; H t denotes the set of all HAPs participating in processing tasks in time slot t;

[0045] The comprehensive preference target is defined as:

[0046]

[0047] where M is the matching set, and λ and μ are weight coefficients; using the blocking pair theory, the matching set M is gradually adjusted by determining whether the matching relationship between the terminal device and the HAP satisfies the blocking pair condition; in each iteration, if a blocking pair is found, resources are re-allocated until all blocking pairs are eliminated, so that M reaches a stable state and satisfies the constraint condition;

[0048] The matching problem of the HAP and the LEO is modeled as a many-to-many matching problem, and the GUROBI optimizer is used to solve the minimum transmission energy consumption between the HAP and the LEO;

[0049] Based on the transmission and computing characteristics of the task, a task offloading decision model is established, and a particle swarm optimization algorithm is used to solve the optimal offloading path;

[0050] Based on the HAP selection association and the task offloading decision, the resource allocation problem is optimized to an optimal computing resource and power allocation problem; in the case of ensuring that the allocated computing resource does not exceed the maximum computing capacity, a whale optimization algorithm is used, a penalty function is introduced to handle the constraint condition, local and global searches are performed, the global optimum is explored, and the optimal solution is gradually approached through fitness evaluation, and the penalty function Λ(p) is represented as:

[0051]

[0052] where δ1(t) and η1(t) represent the adaptive penalty factors of the key constraint condition, which dynamically change with time slots; δ2(t) and η2(t) are used for the adaptive penalty factors of the secondary constraint condition; denotes the indicator function, which is used to determine whether the constraint condition is violated; denotes the indicator function, which is used to determine whether the constraint condition is violated; r and z are adjustment violation degree index parameters, which are used to increase the penalty of solutions with greater violation degrees; denotes the logarithmic processing penalty of the violation degree for a smaller range of violation degrees; denotes the exponential penalty of the violation degree for a rapidly growing violation degree; Q(p) denotes the objective function, denotes the power weight factor.

[0053] Further, the task offloading decision model, in particular:

[0054]

[0055]

[0056]

[0057] Further, if the indicator function G h = 1; otherwise G h = 0.

[0058] Further, if the indicator function H s = 1; otherwise H s = 0.

[0059] Further, the objective function Q(p), in particular:

[0060]

[0061] Further, the S3, in particular: taking the iteration obtained system total energy consumption weighted sum as fitness evaluation standard, in each iteration, by optimizing HAP association selection, task offloading path and resource allocation configuration, gradually approaching the optimal solution; the iteration process updates the fitness evaluation standard after each iteration, until the system total energy consumption weighted sum reaches the convergence condition to obtain the optimal system total energy consumption weighted sum.

[0062] The beneficial effects of the present application are:

[0063] For the task offloading problem of intensive tasks, in view of the sharp increase in the number of terminal devices and calculation, the complexity of communication infrastructure in some areas and other factors; considering the dynamic association relationship of HAP, and its large-scale access simultaneously acting as a relay and an air base station to provide edge computing services for intensive tasks, and the edge server on the satellite can also provide edge computing services; a high energy efficient hybrid offloading strategy is proposed, by jointly optimizing the selection association of HAP, the task offloading path and reasonably allocating the computing resources, the efficiency and flexibility of resource allocation are realized. The present application effectively improves the system energy efficiency.

[0064] Other advantages, objects, and features of the present application will be apparent to those skilled in the art from the following specification, and will be learned from the practice of the present application. The objects and other advantages of the present application can be realized and obtained by the following specification. BRIEF DESCRIPTION OF DRAWINGS

[0065] In order to make the objects, technical solutions and advantages of the present application clearer, the preferred embodiments of the present application will be described in detail below with reference to the drawings, in which:

[0066] Figure 1 A network architecture diagram of a specific embodiment of the present application;

[0067] Figure 2 A resource allocation method flow chart for intensive tasks of a specific embodiment of the present application. DETAILED DESCRIPTION

[0068] The embodiments of the present application will be described below through specific examples. Other advantages and effects of the present application can be easily understood by those skilled in the art from the content disclosed in the specification. The present application can also be implemented or applied through other different embodiments, and various modifications or changes can be made to the details in the specification without departing from the spirit of the present application. It should be noted that the drawings provided in the following embodiments only schematically illustrate the basic concepts of the present application, and the features in the following embodiments and examples can be combined with each other without conflict.

[0069] It should be noted that the drawings are only used for illustrative purposes, and the representation is only a schematic diagram, not a physical diagram, and should not be understood as a limitation of the present application. In order to better illustrate the embodiments of the present application, some components in the drawings may be omitted, enlarged or reduced, and do not represent the actual size of the product. It is understandable for those skilled in the art that some well-known structures and their descriptions in the drawings may be omitted.

[0070] The same or similar reference numerals in the drawings of the embodiments of the present application correspond to the same or similar components; in the description of the present application, it should be understood that the orientation or positional relationship indicated by the terms "upper", "lower", "left", "right", "front", "back" and the like is based on the orientation or positional relationship shown in the drawings, and is only for the convenience of describing the present application and simplifying the description, and does not indicate or imply that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, therefore the terms describing the positional relationship in the drawings are only used for illustrative purposes, and should not be understood as a limitation of the present application. For those skilled in the art, the specific meaning of the above terms can be understood according to the specific circumstances.

[0071] Please refer to Figure 1 A network architecture diagram of a specific embodiment of the present application; please refer to Figure 2 A resource allocation method flow chart for intensive tasks of a specific embodiment of the present application.

[0072] The resource allocation method for intensive tasks of the present embodiment specifically includes the following steps:

[0073] S1: The intensive tasks of the ground terminal device are collected to high altitude platform (HAP) processing or offloaded to low earth orbit satellite (LEO) processing by HAP, and a system total energy consumption weighted sum mathematical model is built;

[0074] S1.1: The intensive tasks of the terminal device are collected to HAP through link , HAP acts as a base station and a relay node, and processes locally or transfers tasks to LEO through link ; the task processing delay of terminal device u can be expressed as T u , wherein the delay mainly consists of task transmission delay and calculation delay for different offloading modes; the energy consumption of task processing consists of transmission energy consumption of terminal device and transmission energy consumption and calculation energy consumption of different offloading modes, expressed as W E ;

[0075] S1.2: A system total energy consumption weighted sum mathematical model is built under the constraint of maximum delay, expressed as:

[0076]

[0077] , wherein E U,H and E H,L represent transmission energy consumption of terminal device to HAP and HAP to LEO respectively, and α1 and α2 are weight coefficients, respectively measuring different contributions of transmission energy consumption from terminal device to HAP and HAP to LEO to the optimization objective; E H represents HAP calculation energy consumption; E L represents LEO calculation energy consumption; β and γ are constant positive weights of HAP and LEO transmission energy consumption respectively; H represents H HAPs; h represents HAP index from 1 to H; t represents time slot index from 1 to T; o represents HAP offloading decision constraint: P represents maximum power constraint of terminal device and HAP; represents whether the task is processed locally by HAP; represents whether the task is offloaded to LEO for processing; P represents a set of maximum power constraints of terminal device and HAP; represents that the power of path X should not exceed the maximum power constraint; U represents U terminal devices; u represents terminal device index from 1 to U; S represents S LEOs; s represents LEO index from 1 to S; represents the bit size of the current computing task; represents the computing resources possessed by the current path X; F represents the maximum computing resource constraint of HAP and LEO; d u represents the actual completion delay of the task; represents the maximum tolerable latency of a task; C1 represents that the terminal device can only select one task offloading mode in a time slot; C2 represents that the power of the terminal device and the HAP does not exceed the maximum power limit; C3 represents that the resources used by the HAP and the LEO do not exceed the maximum resource constraint; and C4 represents the maximum tolerable latency constraint of task processing.

[0078] The association set is represented as:

[0079]

[0080] wherein, represents whether the terminal device u is associated with the HAP index at time slot t; represents whether the HAP index is associated with the LEO index at time slot t;

[0081]

[0082]

[0083] wherein, u t represents the terminal device u that needs to process a task at time slot t; h t represents the HAP index participating in processing a task at time slot t; s t represents the LEO index participating in processing a task at time slot t;

[0084] It is stipulated that each terminal device can only be associated with one HAP, which is represented as:

[0085]

[0086] It is stipulated that each HAP can only be associated with one satellite, which is represented as:

[0087]

[0088] The load of each HAP is limited, satisfying:

[0089]

[0090] wherein, N H is the maximum number of associations of the HAP;

[0091] The maximum power constraint P of the terminal device and the HAP is represented as:

[0092]

[0093] wherein, represents the maximum power of the terminal device; represents the maximum power of the HAP;

[0094] o is represented as:

[0095]

[0096] where, If then the task will be handled directly at HAP; If then the task will be relayed by HAP to LEO for indirect handling;

[0097] F is denoted as:

[0098]

[0099] where, denotes the maximum available computing resource of HAP; denotes the maximum available computing resource of LEO;

[0100] The set of terminal device to HAP and HAP to LEO paths X is denoted as:

[0101] X = {(i, j) | i e U U H, j e H U S}

[0102] S2: By constructing the bidirectional preference relationship between ground terminal devices and HAPs and the association between HAPs and LEOs, the optimal path selection is determined, and the system total energy consumption weighted sum mathematical model is decoupled into three sub-problems of HAP selection association problem, task offloading and resource allocation, which are solved respectively;

[0103] S2.1: If the task makes offloading decisions at HAP, then the two association problems of terminal device and HAP association and HAP and LEO satellite association are independent; The terminal device and HAP association problem can be modeled as a many-to-one matching game problem, and the preference list P u,h of terminal devices on HAP is denoted as:

[0104]

[0105] where, φ u denotes the task priority weight; denotes the transmission rate of terminal device to HAP; denotes the available computing resource of HAP; t denotes the time slot index, from 1 to T; U denotes the terminal devices with U; u denotes the terminal device index, from 1 to U; u t denotes the terminal device u that needs to process tasks within time slot t; U t denotes the set of all terminal devices that need to process tasks within time slot t;

[0106] The HAP preference list P h,u of terminal device is denoted as:

[0107]

[0108] wherein, denotes the bit size of the current computing task; h t denotes the HAP index participating in processing the task in time slot t; H t denotes the set of all HAPs participating in processing the task in time slot t;

[0109] Considering the preferences of the terminal device for the HAP and the preferences of the HAP for the terminal device, the comprehensive preference target is defined as:

[0110]

[0111] wherein, M is a matching set, and λ and μ are weight coefficients; using the blocking pair theory, the matching relationship between the terminal device and the HAP is determined whether it satisfies the blocking pair condition, and M is gradually adjusted; in each iteration, if a blocking pair is found, resources are re-allocated until all blocking pairs are eliminated, so that M reaches a stable state and satisfies the constraint condition;

[0112] The matching problem between the HAP and the LEO is modeled as a many-to-many matching problem, which is solved by using the GUROBI optimizer, the goal of which is to minimize the transmission energy consumption between the HAP and the LEO, while satisfying the computing task offloading demand of the HAP, so that the matching relationship reaches a stable state;

[0113] S2.2: Based on the transmission and computing characteristics of the task, and each task only selects one processing method, a binary decision variable is used to represent the task offloading decision model, which is represented as:

[0114]

[0115]

[0116]

[0117] By using the particle swarm optimization algorithm, the optimal offloading decision is solved to minimize the system energy consumption under the constraint of the task delay;

[0118] S2.3: Optimizing resource allocation based on HAP selection association and task offloading decisions; First, dynamically select the most suitable HAP according to the task characteristics of the terminal device and the resource status of the HAP. Then, decide whether the task is processed locally in the HAP or offloaded to LEO to minimize energy consumption and optimize resource utilization; The optimization problem is formulated as an optimization problem of computing resources and power allocation. While ensuring that the allocated computing resources do not exceed the maximum computing capacity, the whale optimization algorithm is adopted. By introducing a penalty function to handle constraints, local and global searches are performed to explore the global optimum, and fitness evaluation is used to gradually approach the optimal solution. The penalty function Λ(p) is expressed as:

[0119]

[0120] Where δ1(t) and η1(t) represent the adaptive penalty factors for key constraints, which change dynamically with time slots; δ2(t) and η2(t) are used for the adaptive penalty factors for secondary constraints. This indicates an indicator function used to determine constraints. Whether it is violated, when At that time, the indicator function G h =1, otherwise G h =0; This indicates an indicator function used to determine constraints. Whether it is violated, when At that time, H s =1, otherwise H s =0; r and z are parameters for adjusting the degree of violation, usually taken as 2, used to increase the penalty for solutions with a higher degree of violation; Indicates the degree of violation Logarithmic penalties are suitable for a smaller range of violation severity. Indicates the degree of violation The exponential penalty applies to rapidly increasing levels of violation; Describe the objective function. This represents the power weighting factor.

[0121] S3: Perform joint iterative optimization on the results of HAP selection association, task offloading decision and resource allocation, using the weighted sum of total system energy consumption obtained in each iteration as the fitness evaluation standard, until convergence to the optimal weighted sum of total system energy consumption.

[0122] S3.1: The weighted sum of total system energy consumption obtained through iteration is used as the fitness evaluation standard. In each iteration, the optimal solution is gradually approached by optimizing HAP association selection, task unloading path and resource allocation configuration. The fitness evaluation standard is updated after each iteration until the weighted sum of total system energy consumption reaches the convergence condition, thereby obtaining the optimal weighted sum of total system energy consumption.

[0123] The application is directed to the task offloading problem of intensive tasks; due to the sharp increase in the number of terminal devices and computing, the complexity of communication infrastructure in some areas and other problems; considering the dynamic association relationship of HAP, and its large-scale access while serving as a relay and an air base station, edge computing services are provided for intensive tasks, and edge servers on satellites can also provide edge computing services, a high energy-efficient hybrid offloading strategy is proposed, by jointly optimizing the selection association of HAP, the task offloading path and reasonably allocating computing resources, the efficiency and flexibility of resource allocation are realized. The application effectively improves the system energy efficiency.

[0124] Finally, it should be pointed out that the above embodiments are only used to illustrate the technical solutions of the present application and are not limiting. Although the present application has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solutions of the present application can be modified or replaced by equivalents without departing from the spirit and scope of the present application, and they should be covered in the scope of the claims of the present application.

Claims

1. A method for resource allocation for intensive tasks, characterized in that, The method comprises the following steps: S1. The intensive tasks of the ground terminal equipment are collected to a high altitude platform (HAP) or the tasks are forwarded by the HAP to a low earth orbit (LEO) satellite for processing, and a system total energy consumption weighted sum mathematical model is constructed; S2. Optimal path selection is determined by constructing a bidirectional preference relationship between the ground terminal equipment and the HAP and an association between the HAP and the LEO, and the system total energy consumption weighted sum mathematical model is decoupled into three sub-problems of HAP selection association, task offloading decision and resource allocation for solving respectively; S3. The solving results of the three sub-problems of HAP selection association, task offloading decision and resource allocation are jointly iteratively optimized, and the system total energy consumption weighted sum obtained in each iteration is taken as an adaptability evaluation standard until the optimal system total energy consumption weighted sum is converged; In the S1, the HAP acts as a base station and a relay node, and the system total energy consumption weighted sum mathematical model is constructed under a maximum delay constraint and is expressed as: in, and These represent the transmission power consumption from the terminal device to the HAP and from the HAP to the LEO, respectively. and These are weighting coefficients, which respectively measure the different contributions of transmission energy consumption from the terminal device to the HAP and from the HAP to the LEO to the optimization objective; Indicates the energy consumption for HAP calculation; This indicates the energy consumption calculated by LEO; and These are constant positive weights for the transmission energy consumption of HAP and LEO, respectively. Indicates having H One HAP; h Represents the HAP index, from 1 to H ; Indicates the time slot index, from 1 to T ; Represents the offloading decision constraints for HAP: P This indicates the maximum power constraints for the terminal equipment and the HAP; Indicates whether the task is processed locally in the HAP; Indicates whether the task has been unloaded to LEO processing; This represents the set of maximum power constraints for terminal devices and HAPs; Representing a path The power should not exceed the maximum power constraint; U Indicates having U One terminal device; u Represents the terminal device index, from 1 to... ; S Indicates having S One LEO; s Indicates the LEO index, from 1 to S ; This represents the bit size of the current computation task; Indicates the current path The computing resources available; F This represents the maximum computational resource constraints for HAP and LEO; Indicates the actual completion delay of the task; Indicates the maximum tolerable latency for the task; C 1 indicates that the terminal device can only select one task offloading method within a time slot; C 2 indicates that the power of the terminal equipment and HAP does not exceed the maximum power limit; C 3 indicates that HAP and LEO do not use resources exceeding the maximum resource constraint; C 4 indicates the maximum tolerable latency constraint for task processing; The association constraint is expressed as: wherein, denotes whether at time slot t the terminal device u is associated with a HAP index; denotes whether at time slot t the HAP index is associated with a LEO index; Maximum power constraints of terminal devices and HAPs P , is represented as: wherein, represents the maximum power of the terminal device; represents the maximum power of the HAP; is represented as: wherein, ; when , the task will be handled directly at the HAP; when , the task will be relayed through the HAP to be handled indirectly at the LEO; F is expressed as: wherein, represents the maximum available computing resources of the HAPs; represents the maximum available computing resources of the LEOs; Terminal device to HAP and HAP to LEO path set X denoted as: 。 2. The resource allocation method for intensive tasks according to claim 1, characterized in that: The is represented as: The is represented as: wherein, represents a terminal device that needs to process a task within a time slot t ; u ; represents an HAP index that participates in processing a task within a time slot t ; represents an LEO index that participates in processing a task within a time slot t ; Each terminal equipment can be associated with only one HAP, which is expressed as: Each HAP can be associated with only one satellite, which is expressed as: The load of each HAP is limited, satisfying: wherein, is the maximum number of associations for HAPs.

3. The method of claim 1, wherein, The S2 is specifically: A preference list of terminal devices on a hap is represented as: in, Represented as task priority weight; This represents the transmission rate from the terminal device to the HAP. This represents the available computing resources for HAP. Indicates the time slot index, from 1 to T; U Indicates having U One terminal device; u Represents the terminal device index, from 1 to... ; Indicates in time slot t Terminal devices that need to process tasks u ; Indicates in time slot t The collection of all terminal devices that need to process tasks; Hap preference list of terminal device is represented as: in, This represents the bit size of the current computation task; Indicates in time slot t HAP index for internal participation in processing tasks; This represents the set of all HAPs that participate in the processing task within time slot t; The comprehensive preference target is defined as: wherein, is a matching set, and is a weight coefficient; using the blocking pair theory, by judging whether the matching relationship between the terminal device and the HAP satisfies the blocking pair condition, gradually adjusting ; in each iteration, if a blocking pair is found, re-allocate resources until all blocking pairs are eliminated, so that reach a stable state and satisfy the constraint conditions; The matching problem of the HAP and the LEO is modeled as a many-to-many matching problem, and a GUROBI optimizer is used to solve the minimum transmission energy consumption between the HAP and the LEO; Based on the transmission and calculation characteristics of the tasks, a task offloading decision model is established, and a particle swarm optimization algorithm is used to solve the optimal offloading path; Based on the HAP selection association and task offloading decision, the optimization resource allocation problem is formulated as an optimization computing resource and power allocation problem; in the case of ensuring that the allocated computing resources do not exceed the maximum computing capacity, the whale optimization algorithm is adopted, the constraint condition is handled by introducing a penalty function, local and global search is carried out, the global optimum is explored, and the optimal solution is gradually approached through fitness evaluation, the penalty function is expressed as: wherein, and denotes a key constraint adaptive penalty factor, which varies dynamically with time slots; and for a secondary constraint adaptive penalty factor; denotes an indicator function for judging whether a constraint is violated; denotes an indicator function for judging whether a constraint is violated; and is an adjustment violation degree index parameter for increasing the penalty of a solution with a larger violation degree; denotes a logarithmic processing penalty for the violation degree for a smaller range of violation degrees; denotes an exponential penalty for the violation degree for a rapidly growing violation degree; denotes an objective function, denotes a power weight factor.

4. The method of claim 3, wherein, The task offloading decision model is specifically: 。 5. The method of claim 3, wherein: If , then the indicator function ; otherwise .

6. The method of claim 3, wherein: If then ; otherwise .

7. The method of claim 3, wherein, The objective function In particular: 。 8. The method of claim 1, wherein, The S3 is specifically: the system total energy consumption weighted sum obtained in the iteration is taken as an adaptability evaluation standard, and the HAP association selection, the task offloading path and the resource allocation configuration are optimized in each iteration to gradually approach the optimal solution; The adaptability evaluation standard is updated after the end of each iteration in the iteration process until the system total energy consumption weighted sum reaches a convergence condition to obtain the optimal system total energy consumption weighted sum.

Images