Task unloading method based on medium and low orbit hybrid satellite network

Abstract

The invention requests to protect a task unloading method based on a medium and low orbit hybrid satellite network, and belongs to the technical field of communication. The task offloading method based on the medium and low orbit hybrid satellite network is provided for solving the problem that the task offloading performance is deteriorated due to the fact that high time-varying characteristics of a low orbit inter-satellite link may cause low resource utilization rate and service load imbalance of a low orbit satellite network. According to the method, an optimal low-orbit access satellite and an optimal medium-orbit cooperative satellite are selected according to the position of a ground user, the medium-orbit cooperative satellite selects a corresponding low-orbit unloading satellite according to the type and the data volume of each subtask in a task request and the resource state of the low-orbit satellite, a task unloading decision is constructed and optimized, and the task unloading efficiency is improved. Therefore, the resource utilization rate and the task completion rate of the network are maximized, and the unit task unloading time delay of the system is effectively reduced.

Description

Technical Field

[0001] The present invention belongs to the field of communication technology, and in particular relates to a task offloading method based on a medium- and low-orbit hybrid satellite network. Background Art

[0002] With the rapid development of satellite networks, using satellites to directly provide edge computing services to ground terminals has become an important direction for future network development. Using low Earth Orbit (LEO) satellite networks to achieve broad ground coverage can provide good communication and edge computing services for areas with weak network facilities. However, due to the high-speed operation of LEO satellites, the state of inter-satellite links in LEO satellite networks is constantly changing, which may cause frequent routing interruptions during task offloading, leading to task failure and reduced task offloading performance.

[0003] In view of the above problems, the present invention proposes a dual-layer satellite network task offloading method based on medium and low orbit hybrid, which realizes task scheduling and resource allocation by introducing medium Earth Orbit (MEO) satellite-assisted task offloading in LEO satellite network. The method first selects the best low-orbit access satellite and medium-orbit cooperative satellite according to the ground user position and the current star network status. The medium-orbit cooperative satellite constructs a set of candidate low-orbit offloading satellites according to the type and data volume of each subtask in the task offloading request and the distribution of star network resource status. The task offloading decision is optimized by differential evolution algorithm, thereby improving the utilization rate of star network resources, maximizing the task completion rate and reducing the unit task offloading delay. Summary of the invention

[0004] The present invention aims to solve the above problems of the prior art. A task offloading method based on a medium- and low-orbit hybrid satellite network is proposed. The technical solution of the present invention is as follows:

[0005] A task offloading method based on a medium- and low-orbit hybrid satellite network comprises the following steps:

[0006] 101. Assume that the task offloading request of user i on the ground is J = {j}, where j is a parallel subtask, and initialize the candidate medium-orbit cooperative satellite set Candidate LEO satellite collection

[0007] 102. According to the location of user i and the current star network status, obtain the best low-orbit access satellite s a , if successful, jump to step 103, otherwise, jump to step 109;

[0008] 103. Will be connected to low-orbit satellites a The medium-orbit satellites with intersatellite links are added to the set S c, and according to the access satellite s a The inter-satellite link distance with the medium-orbit satellite, for the candidate medium-orbit cooperative satellite set S c Arrange the elements in ascending order;

[0009] 104. If Take out the set S c The first medium-orbit satellite s in the c =s, will be with s c All low-orbit satellites with intersatellite links are added to the candidate low-orbit satellite set S o , and according to s c The intersatellite link distance with the low-orbit satellite, for the set S o Arrange the elements in ascending order, jump to step 105, otherwise, jump to step 109;

[0010] 105. Let the temporary set J′=J, if Jump to step 106, otherwise, jump to step 104;

[0011] 106. If Take out the first element j in the set J′ and jump to step 107; otherwise, jump to step 108;

[0012] 107. According to the candidate low-orbit cooperative satellite set S o , the task amount m of subtask j j and task type v j , construct a set of candidate low-orbit unloading satellites S for subtask j j , if successful, jump to step 106, otherwise, jump to step 104;

[0013] 108. According to the candidate low-orbit unloading satellite set S of each subtask j j , construct the optimal task offloading decision n for task offloading request J based on differential evolution algorithm * If successful, output the uninstall decision n * , jump to step 109, otherwise, jump to step 104;

[0014] 109. The steps are completed.

[0015] Furthermore, in step 102, the best low-orbit access satellite s is obtained. a The method comprises the following steps:

[0016] 1) Initialize the candidate low-orbit access satellite set

[0017] 2) Add the low-orbit satellites that meet the access conditions of user i to the set S a ,if Jump to step

[0018] 3), otherwise, jump to step 6);

[0019] 3) According to the set S a The remaining computing resources of each satellite s in For S a Arrange the elements in descending order and calculate the relationship between user i and set S a The distance d between medium and low orbit satellites s i,s , according to d i,s The value pair Identical satellites are arranged in ascending order;

[0020] 4) Take out the set S a The first low-orbit satellite s in the low-orbit access satellite s a =s, and output s a ;

[0021] 5) The step ends.

[0022] Furthermore, in step 2), the low-orbit satellites that meet the access conditions of user i are added to the set S a The methods specifically include:

[0023] The elevation angle γ of ground user i to low-orbit satellite s i,s Greater than or equal to the minimum elevation angle γ for satellite-to-ground communication min , and the remaining coverage time t of low-orbit satellite s for ground user i i,s Greater than or equal to the mission tolerance delay of ground user i The low-orbit satellite s is added to the set S a , where the elevation angle γ of ground user i to low-orbit satellite s i,s The calculation method of is shown in formula (1);

[0024]

[0025] In formula (1), Δφ=φ i -φ s ,φ i and denote the longitude and latitude of ground user i, respectively, φ s and φ s They represent the longitude and latitude of the low-orbit satellite s, R represents the radius of the earth, and h represents the orbital height of the low-orbit satellite.

[0026] Furthermore, in step 3), user i and set S a The distance d between medium and low orbit satellites s i,s The calculation method is shown in formula (2):

[0027]

[0028] In formula (2), R represents the radius of the earth, h represents the orbital height of the low-orbit satellite, and θ i,s represents the geocentric angle of ground user i relative to low-orbit satellite s, θ i,s The calculation of is shown in formula (3), where γ i,s represents the elevation angle of ground user i relative to low-orbit satellite s.

[0029] Furthermore, in step 107, a candidate low-orbit unloading satellite set S is constructed for subtask j. j The method comprises the following steps:

[0030] 6) Let the candidate low-orbit satellite set S of subtask j be j =S o , temporary set S j ′=S j ;

[0031] 7) If Take out the set S j ′, according to the data volume m of subtask j j and task tolerance delay t max , calculate the computing resources z required by satellite s to process subtask j j,s , jump to step 8), otherwise, jump to step 10);

[0032] 8) If satellite s has no v j Type of virtual network function VNF instance, let z j,s =z j,s +z0, where z0 is the basic computing resource required for VNF instantiation, jump to step 9), otherwise, jump to step 9);

[0033] 9) If the remaining computing resources on satellite s Less than z j,s , remove satellite s from the set S j If removed, jump to step 7), otherwise, jump to step 7);

[0034] 10) The steps are completed.

[0035] Furthermore, in step 7), the computing resources z required for satellite s to process subtask j j,s The calculation method is shown in formula (4):

[0036]

[0037] In formula (4), m j and jThey represent the task data volume and task complexity of subtask j in task offloading request J, respectively, and t max represents the task tolerance delay of task offloading request J, represents the uplink transmission delay of subtask j in task offloading request J, which is calculated as shown in formula (5): represents the uplink and downlink propagation delay of subtask j in task offloading request J, which is calculated as shown in formula (6);

[0038]

[0039] In formula (5), R i,s represents the transmission rate between ground user i and low-orbit satellite s, is a binary variable, indicating whether subtask j is forwarded via medium-orbit cooperative satellite s. If subtask j is forwarded via medium-orbit cooperative satellite s, let Otherwise, let R e represents the intersatellite transmission rate. In formula (6), c is the speed of light, d e Indicates the distance between stars.

[0040] Furthermore, in step 108, the optimal task offloading decision n is constructed for the task offloading request J based on the differential evolution algorithm. * The method comprises the following steps:

[0041] 11) Let the population be a set N = {n}, where the individuals in the population n = {nj|j ∈J} is the unloading decision vector, nj is the unloading satellite of subtask j in set J, and the candidate low-orbit unloading satellite set S for each subtask j is j Initialize n randomly j , calculate the objective function value t of n n , let the optimal objective function value t * =min{t n} n∈N}, the best individual The maximum number of iterations is C, and the iteration count variable c = 0;

[0042] 12) Let c = c + 1, if c ≤ C, let the temporary set N' = N, jump to step 13), otherwise, jump to step 16);

[0043] 13) If Take out the first element n in the set N′ and calculate the objective function value t n , perform mutation and crossover operations on n to obtain individual n after mutation and crossover cross , calculate the objective function value Jump to step 14), otherwise, jump to step 12);

[0044] 14) If make n=n cross , jump to step 15), otherwise, jump to step 13);

[0045] 15) If t n ≤t * , let t * =t n , n * =n, jump to step 13), otherwise, jump to step 13);

[0046] 16) Output the optimal unloading decision n * , the step ends.

[0047] Furthermore, in step 11), the objective function value t of n is calculated n The calculation method of is shown in formula (7):

[0048]

[0049] In formula (7), t j represents the unloading completion delay of subtask j, which is calculated as shown in formula (8):

[0050]

[0051] In formula (8), represents the computational delay of subtask j, which is calculated as shown in formula (9):

[0052]

[0053] In formula (9), represents the computing resources allocated by the low-orbit satellite s to the subtask j, and the calculation is shown in formula (10):

[0054]

[0055] In formula (10), represents the remaining available computing resources of satellite s, z j,s represents the computing resources required for subtask j to complete the task at satellite s.

[0056] Furthermore, the method of performing mutation and crossover operations on n in step 13) specifically includes:

[0057] (1) Mutation operation: Use formula (11) to perform mutation on each n in individual n. j Perform mutation processing to obtain Thus, the mutated individuals are obtained

[0058]

[0059] In formula (11), Indicates rounding down, ε1, ε2, ε3 indicate the value interval [1,S j ], and ε1≠ε2≠ε3, α1∈[0,1] represents the mutation scaling factor;

[0060] (2) Crossover operation: Use formula (12) to perform crossover on each n in individual n. j Perform cross processing to obtain So as to obtain the individual after crossover

[0061]

[0062] In formula (12), rand(0,1) represents a random number in the numerical interval [0,1), and α2∈[0,1] represents the crossover probability.

[0063] The advantages and beneficial effects of the present invention are as follows:

[0064] The present invention discloses a task offloading method based on a medium- and low-orbit hybrid satellite network. Most of the existing research on satellite network task offloading only considers the task offloading of ground users in a single-layer satellite network, such as a LEO low-orbit network, and the high time variability of the LEO inter-satellite link may pose a great challenge to the continuity and stability of the task offloading route, thereby reducing the resource utilization of the satellite network and causing the problem of deterioration of the task offloading performance. In response to this problem, the present invention proposes a double-layer satellite network task offloading method based on a medium- and low-orbit hybrid, which provides cross-layer collaboration by introducing a MEO medium-orbit satellite to assist the LEO network in implementing task scheduling and resource allocation decisions. The method first selects the best low-orbit access satellite and medium-orbit cooperative satellite according to the location of the ground user, and the medium-orbit cooperative satellite constructs a set of candidate low-orbit offloading satellites according to the type and computing resource requirements of each subtask in the offloading request, and optimizes the task offloading decision through a differential evolution algorithm, thereby improving the resource utilization of the satellite network, maximizing the task completion rate and reducing the system unit task offloading delay. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 The present invention provides a flow chart of a task offloading method based on a medium- and low-orbit hybrid satellite network in a preferred embodiment. DETAILED DESCRIPTION

[0066] The following will describe the technical solutions in the embodiments of the present invention in detail in conjunction with the accompanying drawings in the embodiments of the present invention. The described embodiments are only part of the embodiments of the present invention.

[0067] The technical solution of the present invention to solve the above technical problems is:

[0068] The concepts and models involved in the present invention are as follows:

[0069] The present invention assumes that the medium- and low-orbit satellite network is composed of several low-orbit LEO satellites and several medium-orbit MEO satellites, and the satellites can communicate through line-of-sight links. Ground users access the satellite network through LEO satellites, and the LEO access satellite selects a medium-orbit satellite as a task scheduling cooperation satellite. The medium-orbit cooperation satellite selects a low-orbit unloading satellite for each subtask of the user according to the resource status distribution of the star network and the link connection status. Each subtask can only have one unloading satellite, and each satellite can perform multiple subtasks. Considering that the amount of result data returned from each unloading satellite is small, the task unloading delay only calculates the downlink propagation delay of the result data returned, and ignores its transmission delay.

[0070] The technical solution of the present invention is described as follows:

[0071] 1. The elevation angle γ of ground user i to low-orbit satellite s i,s , calculated as shown in formula (1):

[0072]

[0073] In formula (1), Δφ=φ i -φ s ,φ i and denote the longitude and latitude of ground user i, respectively, φ s and They represent the longitude and latitude of the low-orbit satellite s, R represents the radius of the earth, and h represents the orbital height of the low-orbit satellite;

[0074] 2. User i and set S a The distance d between medium and low orbit satellites s i,s , calculated as shown in formula (2):

[0075]

[0076] In formula (2), R represents the radius of the earth, h represents the orbital height of the low-orbit satellite, and θ i,s represents the geocentric angle of ground user i relative to low-orbit satellite s;

[0077] 3. The geocentric angle θ between ground user i and low-orbit satellite s i,s , calculated as shown in formula (3):

[0078]

[0079] 4. Computational resources z required by satellite s to process subtask j j,s , calculated as shown in formula (4):

[0080]

[0081] In formula (4), m j and j They represent the task data volume and task complexity of subtask j in task offloading request J, respectively, and t max represents the task tolerance delay of task offloading request J, represents the uplink transmission delay of subtask j in task offloading request J, represents the uplink and downlink propagation delay of subtask j in task offloading request J;

[0082] 5. Uplink transmission delay of subtask j offloaded to satellite s The calculation method is shown in formula (5):

[0083]

[0084] In formula (5), R i,s represents the transmission rate between ground user i and low-orbit satellite s, is a binary variable, indicating whether subtask j is forwarded via medium-orbit cooperative satellite s. If subtask j is forwarded via medium-orbit cooperative satellite s, let Otherwise, let R e Indicates the intersatellite transmission rate;

[0085] 6. Uplink and downlink propagation delay of subtask j offloaded to satellite s The calculation method is shown in formula (6):

[0086]

[0087] In formula (6), c is the speed of light, d is e Indicates the distance between stars;

[0088] 7. The objective function value t of population individual n n , the calculation method is shown in formula (7):

[0089]

[0090] 8. Delay t for completion of unloading of subtask j j , calculated as shown in formula (8):

[0091]

[0092] 9. Computation delay of subtask j The calculation is shown in formula (9):

[0093]

[0094] 10. Computational resources allocated by low-orbit satellite s to subtask j The calculation is shown in formula (10):

[0095]

[0096] In formula (10), represents the remaining available computing resources of satellite s, z j,s represents the computing resources required for subtask j to complete the task at satellite s;

[0097] 11. Methods for mutation and crossover operations on population individuals n

[0098] (1) Mutation operation: Use formula (11) to perform mutation on each n in individual n. j Perform mutation processing to obtain Thus, the mutated individuals are obtained

[0099]

[0100] In formula (11), Indicates rounding down, ε1, ε2, ε3 indicate the value interval [1,S j ], and ε1≠ε2≠ε3, represents the variation scaling factor;

[0101] (2) Crossover operation: Use formula (12) to perform crossover on each n in individual n. j Perform cross processing to obtain So as to obtain the individual after crossover

[0102]

[0103] In formula (12), rand(0,1) represents a random number in the numerical interval [0,1), and α2∈[0,1] represents the crossover probability;

[0104] 13. Sub-algorithm 1: Obtain the best low-orbit access satellite s a

[0105] Step 1: Initialize the candidate low-orbit access satellite set

[0106] Step 2: Add low-orbit satellites that meet the access conditions of user i to the set S a ,if Jump to step 3, otherwise, jump to step 5;

[0107] Step 3: According to the set S a The remaining computing resources of each satellite s in For S a Arrange the elements in descending order and calculate the relationship between user i and set S a The distance d between medium and low orbit satellites s i,s , according to d i,s The value pair Identical satellites are arranged in ascending order;

[0108] Step 4: Take out the set S a The first low-orbit satellite s in the low-orbit access satellite s a =s, and output s a ;

[0109] Step 5: End of step.

[0110] 14. Sub-algorithm 2: Construct a set S of candidate low-orbit unloading satellites for subtask j j

[0111] Step 1: Let the candidate low-orbit satellite set S of subtask j be j =S o , temporary set S j ′=S j ;

[0112] Step 2: If Take out the set S j ′, according to the data volume m of subtask j j and task tolerance delay t max , calculate the computing resources z required by satellite s to process subtask j j,s , jump to step 3, otherwise, jump to step 5;

[0113] Step 3: If satellite s has no v j Type of virtual network function VNF instance, let z j,s =z j,s +z0, where z0 is the basic computing resource required for VNF instantiation, jump to step 4; otherwise, jump to step 4;

[0114] Step 4: If the remaining computing resources on satellite s Less than z j,s , remove satellite s from the set S j If removed, jump to step 2; otherwise, jump to step 2;

[0115] Step 5: End of step.

[0116] 15. Sub-algorithm 3: Constructing the optimal task offloading decision n for task offloading request J based on differential evolution algorithm *

[0117] Step 1: Let the population be a set N = {n}, where the individual n = {nj|j ∈J} is the offloading decision vector, n j is the unloading satellite of subtask j in set J, according to the candidate low-orbit unloading satellite set S of each subtask j j Initialize n randomly j , calculate the objective function value t of n n , let the optimal objective function value t * =min{t n | n∈N}, the best individual The maximum number of iterations is C, and the iteration count variable c = 0;

[0118] Step 2: Let c = c + 1. If c ≤ C, let the temporary set N′ = N and jump to step 3. Otherwise, jump to step 6.

[0119] Step 3: If Take out the first element n in the set N′ and calculate the objective function value t n , perform mutation and crossover operations on n to obtain individual n after mutation and crossover cross , calculate the objective function value Jump to step 4, otherwise, jump to step 2;

[0120] Step 4: If make n=n cross , jump to step 5, otherwise, jump to step 3;

[0121] Step 5: If t n ≤t * , let t * =t n , n * =n, jump to step 3, otherwise, jump to step 3;

[0122] Step 6: Output the optimal offloading decision n * , the step ends.

[0123] A task offloading method based on a medium- and low-orbit hybrid satellite network, the specific implementation method includes the following steps:

[0124] Step 1: Assume that the task offloading request of user i on the ground is J = {j}, where j is a parallel subtask, and initialize the candidate medium-orbit cooperative satellite set Candidate LEO satellite collection

[0125] Step 2: According to the location of user i and the current satellite network status, call sub-algorithm 1 to obtain the best low-orbit access satellite s a , if successful, jump to step 3, otherwise, jump to step 9;

[0126] Step 3: Connect to low-orbit access satellites a The medium-orbit satellites with intersatellite links are added to the set S c , and according to the access satellite s a The inter-satellite link distance with the medium-orbit satellite, for the candidate medium-orbit cooperative satellite set S c Arrange the elements in ascending order;

[0127] Step 4: If Take out the set S c The first medium-orbit satellite s in the c =s, will be with s c All low-orbit satellites with intersatellite links are added to the candidate low-orbit satellite set S o , and according to s c The intersatellite link distance with the low-orbit satellite, for the set S o Arrange the elements in ascending order, jump to step 5, otherwise, jump to step 9;

[0128] Step 5: Let temporary set J′=J, if Jump to step 6, otherwise, jump to step 4;

[0129] Step 6: If Take out the first element j in the set J′ and jump to step 7, otherwise, jump to step 8;

[0130] Step 7: Based on the candidate low-orbit satellite set S o , the task amount m of subtask j j and task type v j , call sub-algorithm 2 to construct a set of candidate low-orbit unloading satellites S for subtask j j , if successful, jump to step 6, otherwise, jump to step 4;

[0131] Step 8: According to the candidate low-orbit unloading satellite set S of each subtask j j , calling sub-algorithm 3, constructing the optimal task offloading decision n for task offloading request J based on differential evolution algorithm * If successful, output the uninstall decision n * , jump to step 9, otherwise, jump to step 4;

[0132] Step 9: End of the step.

[0133] The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products with certain functions.

[0134] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0135] The above embodiments should be understood to be only used to illustrate the present invention and not to limit the protection scope of the present invention. After reading the contents of the present invention, technicians can make various changes or modifications to the present invention, and these equivalent changes and modifications also fall within the scope defined by the claims of the present invention.

Claims

1. A task offloading method based on a medium- and low-orbit hybrid satellite network, characterized in that: The following steps are involved:

101. Assume that the task offloading request of user i on the ground is J = {j}, where j is a parallel subtask, and initialize the candidate medium-orbit cooperative satellite set Candidate LEO satellite collection 102. According to the location of user i and the current star network status, obtain the best low-orbit access satellite s a , if successful, jump to step 103, otherwise, jump to step 109; 103. Will be connected to low-orbit satellites a The medium-orbit satellites with intersatellite links are added to the set S c , and according to the access satellite s a The inter-satellite link distance with the medium-orbit satellite, for the candidate medium-orbit cooperative satellite set S c Arrange the elements in ascending order; 104. If Take out the set S c The first medium-orbit satellite s in the c =s, will be with s c All low-orbit satellites with intersatellite links are added to the candidate low-orbit satellite set S o , and according to s c The intersatellite link distance with the low-orbit satellite, for the set S o Arrange the elements in ascending order, jump to step 105, otherwise, jump to step 109; 105. Let the temporary set J′=J, if Jump to step 106, otherwise, jump to step 104; 106. If Take out the first element j in the set J′ and jump to step 107; otherwise, jump to step 108; 107. According to the candidate low-orbit cooperative satellite set S o , the task amount m of subtask j j and task type v j , construct a set of candidate low-orbit unloading satellites S for subtask j j , if successful, jump to step 106, otherwise, jump to step 104; 108. According to the candidate low-orbit unloading satellite set S of each subtask j j , construct the optimal task offloading decision n for task offloading request J based on differential evolution algorithm * If successful, output the uninstall decision n * , jump to step 109, otherwise, jump to step 104; 109. The steps are completed.

2. According to claim 1, a task offloading method based on a medium- and low-orbit hybrid satellite network is characterized in that: In step 102, the best low-orbit access satellite s is obtained. a The method comprises the following steps: 1) Initialize the candidate low-orbit access satellite set 2) Add the low-orbit satellites that meet the access conditions of user i to the set S a ,if Jump to step 3), otherwise, jump to step 6); 3) According to the set S a The remaining computing resources of each satellite s in For S a Arrange the elements in descending order and calculate the relationship between user i and set S a The distance d between medium and low orbit satellites s i,s , according to d i,s The value pair Identical satellites are arranged in ascending order; 4) Take out the set S a The first low-orbit satellite s in the low-orbit access satellite s a =s, and output s a ; 5) The step ends.

3. The task offloading method based on a medium- and low-orbit hybrid satellite network according to claim 2 is characterized in that: In step 2), the low-orbit satellites that meet the access conditions of user i are added to the set S a The methods specifically include: The elevation angle γ of ground user i to low-orbit satellite s i,s Greater than or equal to the minimum elevation angle γ for satellite-to-ground communication min , and the remaining coverage time t of low-orbit satellite s for ground user i i,s Greater than or equal to the mission tolerance delay of ground user i The low-orbit satellite s is added to the set S a , where the elevation angle γ of ground user i to low-orbit satellite s i,s The calculation method of is shown in formula (1); In formula (1), Δφ=φ i -φ s ,φ i and denote the longitude and latitude of ground user i, respectively, φ s and They represent the longitude and latitude of the low-orbit satellite s, R represents the radius of the earth, and h represents the orbital height of the low-orbit satellite.

4. The task offloading method based on a medium- and low-orbit hybrid satellite network according to claim 2 is characterized in that: In step 3), user i and set S a The distance d between medium and low orbit satellites s i,s The calculation method is shown in formula (2): In formula (2), R represents the radius of the earth, h represents the orbital height of the low-orbit satellite, and θ i,s represents the geocentric angle of ground user i relative to low-orbit satellite s, θ i,s The calculation of is shown in formula (3), where γ i,s represents the elevation angle of ground user i relative to low-orbit satellite s.

5. The task offloading method based on a medium- and low-orbit hybrid satellite network according to claim 1 is characterized in that: In step 107, a candidate low-orbit unloading satellite set S is constructed for subtask j. j The method comprises the following steps: 6) Let the candidate low-orbit satellite set S of subtask j be j =S o , temporary set S j ′=S j ; 7) If Take out the set S j ′, according to the data volume m of subtask j j and task tolerance delay t max , calculate the computing resources z required by satellite s to process subtask j j,s , jump to step 8), otherwise, jump to step 10); 8) If satellite s has no v j Type of virtual network function VNF instance, let z j,s =z j,s +z0, where z0 is the basic computing resource required for VNF instantiation, jump to step 9), otherwise, jump to step 9); 9) If the remaining computing resources on satellite s Less than z j,s , remove satellite s from the set S j If removed, jump to step 7), otherwise, jump to step 7); 10) The steps are completed.

6. The task offloading method based on a medium- and low-orbit hybrid satellite network according to claim 5 is characterized in that: The computing resources z required by satellite s to process subtask j in step 7) j,s The calculation method is shown in formula (4): In formula (4), m j and j They represent the task data volume and task complexity of subtask j in task offloading request J, respectively, and t max represents the task tolerance delay of task offloading request J, represents the uplink transmission delay of subtask j in task offloading request J, which is calculated as shown in formula (5): represents the uplink and downlink propagation delay of subtask j in task offloading request J, which is calculated as shown in formula (6); In formula (5), R i,s represents the transmission rate between ground user i and low-orbit satellite s, is a binary variable, indicating whether subtask j is forwarded via medium-orbit cooperative satellite s. If subtask j is forwarded via medium-orbit cooperative satellite s, let Otherwise, let R e represents the intersatellite transmission rate. In formula (6), c is the speed of light, d e Indicates the distance between stars.

7. The task offloading method based on a medium- and low-orbit hybrid satellite network according to claim 1 is characterized in that: In step 108, the optimal task offloading decision n is constructed for the task offloading request J based on the differential evolution algorithm. * The method comprises the following steps: 11) Let the population be a set N = {n}, where the individuals in the population n = {nj|j ∈J } is the unloading decision vector, nj is the unloading satellite of subtask j in set J, and the candidate low-orbit unloading satellite set S for each subtask j is j Initialize n randomly j , calculate the objective function value t of n n , let the optimal objective function value t * =min{t n | n∈N }, the best individual The maximum number of iterations is C, and the iteration count variable c = 0; 12) Let c = c + 1, if c ≤ C, let the temporary set N' = N, jump to step 13), otherwise, jump to step 16); 13) If Take out the first element n in the set N′ and calculate the objective function value t n , perform mutation and crossover operations on n to obtain individual n after mutation and crossover cross , calculate the objective function value Jump to step 14), otherwise, jump to step 12); 14) If make n=n cross , jump to step 15), otherwise, jump to step 13); 15) If t n ≤t * , let t * =t n , n * =n, jump to step 13), otherwise, jump to step 13); 16) Output the optimal unloading decision n * , the step ends.

8. The task offloading method based on a medium- and low-orbit hybrid satellite network according to claim 7 is characterized in that: In step 11), the objective function value t of n is calculated n The calculation method of is shown in formula (7): In formula (7), t j represents the unloading completion delay of subtask j, which is calculated as shown in formula (8): In formula (8), represents the computational delay of subtask j, which is calculated as shown in formula (9): In formula (9), represents the computing resources allocated by the low-orbit satellite s to the subtask j, and the calculation is shown in formula (10): In formula (10), represents the remaining available computing resources of satellite s, z j,s represents the computing resources required for subtask j to complete the task at satellite s.

9. The task offloading method based on a medium- and low-orbit hybrid satellite network according to claim 7, characterized in that: The method for performing mutation and crossover operations on n in step 13) specifically includes: (1) Mutation operation: Use formula (11) to perform mutation on each n in individual n. j Perform mutation processing to obtain Thus, the mutated individuals are obtained In formula (11), Indicates rounding down, ε1, ε2, ε3 indicate the value interval [1,S j ], and ε1≠ε2≠ε3, α1∈[0,1] represents the mutation scaling factor; (2) Crossover operation: Use formula (12) to perform crossover on each n in individual n. j Perform cross processing to obtain So as to obtain the individual after crossover In formula (12), rand(0,1) represents a random number in the numerical interval [0,1), and α2∈[0,1] represents the crossover probability.

Images