Multi-uav edge computing resource scheduling method and device

Abstract

The application provides a multi-unmanned aerial vehicle edge computing resource scheduling method and device, which comprises the following steps: subtracting the total computing power consumed by all user terminals to be unloaded to each unmanned aerial vehicle from the idle computing power of each unmanned aerial vehicle to obtain the residual computing power of each unmanned aerial vehicle; constructing a target function according to the number of user terminals to be unloaded to each unmanned aerial vehicle and the residual computing power of each unmanned aerial vehicle; solving the target function to obtain the unmanned aerial vehicle to which each user terminal to be unloaded to is unloaded, so that the total number of user terminals corresponding to all unmanned aerial vehicles is maximum, and the maximum value of the residual computing power of the unmanned aerial vehicle is minimum. The application maximizes the service range of unmanned aerial vehicle edge computing, simultaneously considers the load balancing of the unmanned aerial vehicle, and is suitable for a scenario in which the unmanned aerial vehicle may be difficult to provide services for all user terminals.

Description

Technical Field

[0001] The present invention relates to the field of wireless communication technologies, and in particular to a method and device for scheduling edge computing resources of multiple unmanned aerial vehicles. Background Art

[0002] Edge computing is a key 5G technology. It places edge computing servers on the mobile user side, establishing wireless communication links with the user terminals. The user terminals offload high-complexity, low-latency computational tasks directly to the edge servers, which then quickly return the results. Compared to cloud computing, edge computing avoids multiple layers of internet processing and offers greater real-time performance.

[0003] To flexibly relocate edge computing servers based on mobile user needs, drones are used to provide computing and storage capabilities for mobile edge computing networks. However, due to limitations in battery capacity and hardware, drones often cannot maintain flight for extended periods of time. Therefore, existing multi-drone edge computing resource scheduling schemes prioritize minimizing energy consumption for drone nodes and assume that drones can serve all ground users.

[0004] However, in reality, the number of ground users is large, the tasks are complex, and the amount of computation to be done is large. However, the limited computing power of drones makes it difficult to provide edge computing services for all user terminals. Existing technical solutions do not address this application scenario by scheduling edge computing resources. Summary of the Invention

[0005] The present invention provides a multi-UAV edge computing resource scheduling method and device, which is used to solve the defect in the prior art that there is no resource scheduling solution for edge computing service scenarios where UAVs cannot fully cover user terminals, and realize resource scheduling that maximizes the edge computing service range with load balancing of UAVs.

[0006] The present invention provides a multi-UAV edge computing resource scheduling method, comprising:

[0007] Subtract the total computing power consumed by the edge computing tasks to be offloaded to each drone from all user terminals from the idle computing power of each drone to obtain the remaining computing power of each drone.

[0008] Construct an objective function based on the number of user terminals to offload edge computing tasks to each drone and the remaining computing power of each drone;

[0009] The objective function is solved to obtain the drone to which the edge computing task to be offloaded from each user terminal is to be obtained, so that the total number of user terminals corresponding to all drones is maximized and the maximum value of the remaining computing power of the drones is minimized.

[0010] According to a multi-UAV edge computing resource scheduling method provided by the present invention, before obtaining the remaining computing power of each UAV by subtracting the total computing power consumed by the edge computing tasks to be offloaded to each UAV by all user terminals from the idle computing power of each UAV, the method further includes:

[0011] Obtain the computational amount of the edge computing task to be offloaded from each user terminal to each drone based on the size and computational complexity of the edge computing task to be offloaded from each user terminal to each drone;

[0012] Obtain the completion time of the edge computing task to be offloaded from each user terminal to each drone based on the preset maximum completion delay and offloading time of the edge computing task to be offloaded from each user terminal to each drone;

[0013] The computing power consumed by the edge computing tasks to be offloaded from all user terminals to each drone is obtained by dividing the computing power of the edge computing tasks to be offloaded from all user terminals to each drone by the completion time and adding the two amounts.

[0014] According to a multi-UAV edge computing resource scheduling method provided by the present invention, before obtaining the completion time of the edge computing task to be offloaded from each user terminal to each UAV based on the preset maximum completion delay and offloading time of the edge computing task to be offloaded from each user terminal to each UAV, the method further includes:

[0015] Obtaining a transmission rate at which each user terminal sends the edge computing task to each drone based on the subchannel bandwidth of each user terminal, the transmission power for sending the edge computing task to each drone, the channel power gain between the user terminal and each drone, and the receiving noise power of each drone;

[0016] The size of the edge computing task to be offloaded by each user terminal to each drone is divided by the rate to obtain the offloading time for each user terminal to offload the edge computing task to each drone.

[0017] According to a multi-UAV edge computing resource scheduling method provided by the present invention, before obtaining the transmission rate at which each user terminal sends the edge computing task to each UAV based on the subchannel bandwidth of each user terminal, the transmission power for sending the edge computing task to each UAV, the channel power gain between each UAV and the receiving noise power of each UAV, the method further includes:

[0018] Calculate the distance between each user terminal and each drone based on the location of each user terminal and the location of each drone;

[0019] The channel power gain between each user terminal and each drone is calculated based on the height of each drone, the distance between each drone and each user terminal, and the preset reference power.

[0020] According to a multi-UAV edge computing resource scheduling method provided by the present invention, the remaining computing power of each UAV is greater than or equal to 0;

[0021] Each user terminal can offload the edge computing task to at most one drone;

[0022] The number of user terminals to offload edge computing tasks to each drone is less than or equal to the number of sub-channels per drone;

[0023] The horizontal coordinate of each UAV is greater than or equal to the minimum value of the horizontal coordinates of all user terminals, and less than or equal to the maximum value of the horizontal coordinates of all user terminals;

[0024] The vertical coordinate of each UAV is greater than or equal to the minimum value of the vertical coordinates of all user terminals, and less than or equal to the maximum value of the vertical coordinates of all user terminals.

[0025] According to a multi-UAV edge computing resource scheduling method provided by the present invention, solving the objective function to obtain the UAV to which the edge computing task is to be offloaded from each user terminal includes:

[0026] Introducing non-negative slack variables into the objective function;

[0027] Converting the objective function after introducing the non-negative slack variables into a standard form;

[0028] Solve the objective function converted into a standard form based on the branch and bound method to obtain the drone to which the edge computing task is to be offloaded from each user terminal;

[0029] If the objective function converted into the standard form does not converge and the number of iterations does not reach the preset value, the drone of each user terminal to be offloaded from the edge computing task solved in this solution is updated based on the genetic algorithm;

[0030] The objective function converted into the standard form is solved again according to the updated UAV until the objective function converted into the standard form converges or the number of iterations reaches the preset value.

[0031] The present invention also provides a multi-UAV edge computing resource scheduling device, comprising:

[0032] The calculation module is used to subtract the total computing power consumed by the edge computing tasks to be offloaded to each drone by all user terminals from the idle computing power of each drone to obtain the remaining computing power of each drone;

[0033] A construction module is used to construct an objective function based on the number of user terminals to offload edge computing tasks to each drone and the remaining computing power of each drone;

[0034] A solution module is used to solve the objective function and obtain the drone to which each user terminal offloads the edge computing task, so that the total number of user terminals corresponding to all drones is maximized and the maximum value of the remaining computing power of the drones is minimized.

[0035] The present invention also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and runnable on the processor. When the processor executes the program, the steps of any of the above-mentioned multi-UAV edge computing resource scheduling methods are implemented.

[0036] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the multi-UAV edge computing resource scheduling methods described above.

[0037] The present invention also provides a computer program product, comprising a computer program, which, when executed by a processor, implements the steps of any of the above-mentioned multi-UAV edge computing resource scheduling methods.

[0038] The multi-UAV edge computing resource scheduling method and device provided by the present invention construct an objective function based on the number of user terminals to offload edge computing tasks to each UAV and the surplus computing power of each UAV, and optimize the offloading decision of the user terminal through solving the objective function, so as to maximize the UAV edge computing service range while taking into account the load balancing of the UAV. It is suitable for real-world scenarios where there are a large number of ground users and complex tasks, the computing power of the UAV is limited, and it may be difficult to provide services to all users. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solutions in the present invention or the prior art, a brief introduction is given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0040] Figure 1 This is a flow chart of the multi-UAV edge computing resource scheduling method provided by the present invention;

[0041] Figure 2 It is a structural diagram of the UAV edge computing system model in the multi-UAV edge computing resource scheduling method provided by the present invention;

[0042] Figure 3 This is a schematic diagram of the solution process of the objective function in the multi-UAV edge computing resource scheduling method provided by the present invention;

[0043] Figure 4 Schematic diagram of the relationship between the drone service coverage, the maximum computing power of the drone, and the number of drones in the multi-drone edge computing resource scheduling method provided by the present invention;

[0044] Figure 5 Schematic diagram of the relationship between the UAV load variance and load weight in the multi-UAV edge computing resource scheduling method provided by the present invention;

[0045] Figure 6 This is a schematic diagram of the structure of the multi-UAV edge computing resource scheduling device provided by the present invention;

[0046] Figure 7 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0047] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0048] The following combination Figure 1 A method for scheduling edge computing resources of multiple drones according to the present invention is described, comprising: step 101, subtracting the total computing power consumed by edge computing tasks to be offloaded to each drone by all user terminals from the idle computing power of each drone to obtain the remaining computing power of each drone;

[0049] The UAV edge computing system model is as follows Figure 2 Assume that the UAV edge computing system includes M UAVs and K ground user terminals, let Θ = {1, 2, ..., M}, Δ = {1, 2, ..., K} represent the set of UAVs and the set of ground user terminals respectively.

[0050] The coordinates of the kth user terminal are ω k =(x k ,y k ,0), the coordinates of the mth UAV are q m =(q m,x ,q m,y ,H),m∈Θ. Among them, H is the flight altitude of the UAV.

[0051] The purpose of this embodiment is to select user terminals to allocate drones for edge computing. The idle computing power of a drone is the computing power that the drone can use for edge computing. Assume that the idle computing power of the mth drone is f m , the computing power consumed by the edge computing task to be offloaded from the kth user terminal to the mth drone is μ m,k a k Among them, a k represents the computing power consumed to complete the edge computing task of the kth user terminal, μ m，k =1 means that the kth user terminal offloads the edge computing task to the mth UAV, that is, the kth user terminal will be served, μ m,k =0 means that the kth user terminal does not offload the edge computing task to the mth UAV.

[0052] The total computing power consumed by the edge computing tasks to be offloaded from all user terminals to the mth drone is The remaining computing power after the mth drone resource scheduling is

[0053] Step 102: construct an objective function based on the number of user terminals to offload edge computing tasks to each drone and the remaining computing power of each drone;

[0054] The number of user terminals to offload edge computing tasks to each drone is according to and The objective function is constructed in two parts.

[0055] Optionally, the objective function is ω l >0 is the UAV load balancing weight, and θ>0 is the normalization factor, the purpose of which is to unify the two parts of the objective function to the same magnitude.

[0056] because Therefore, the right half of the objective function considers minimizing the maximum value of the remaining computing power among multiple drones, that is, the consideration of load balancing.

[0057] Step 103: Solve the objective function to obtain the drone to which each user terminal offloads the edge computing task, so that the total number of user terminals corresponding to all drones is maximized and the maximum value of the remaining computing power of the drones is minimized.

[0058] Solve the objective function and get μ m，k The value of , that is, whether the kth user terminal offloads the edge computing task to the mth drone, so that maximum, The maximum value in is the smallest.

[0059] The embodiment maximizes the service range of the UAV edge computing by constructing a target function according to the number of user terminals to be unloaded to each UAV and the residual computing power of each UAV, and solving the unloading decision of the user terminal by the target function, while considering the load balancing of the UAV.

[0060] On the basis of the above embodiment, before the residual computing power of each UAV is obtained by subtracting the total computing power consumed by the edge computing tasks of all user terminals to be unloaded to each UAV from the idle computing power of each UAV in the embodiment, the computing amount of the edge computing task of each user terminal to be unloaded to each UAV is obtained according to the size and computing complexity of the edge computing task of each user terminal to be unloaded to each UAV.

[0061] The edge computing task of the kth user terminal is expressed as a three-element array (l k ,c k , t k ), where l k represents the size of the edge computing task of the kth user terminal, c k represents the computing complexity of the edge computing task of the kth user terminal, i.e., the computing power required to process a unit of task amount, and t k represents the preset maximum completion time delay of the edge computing task of the kth user terminal. Optionally, without loss of generality, let t k = T, k e A. The computing amount of the edge computing task of the kth user terminal is l k c k .

[0062] The completion time of the edge computing task of each user terminal to be unloaded to each UAV is obtained according to the preset maximum completion time delay and the unloading time of the edge computing task of each user terminal to be unloaded to each UAV.

[0063] The completion time of the edge computing task of the mth user terminal is obtained by subtracting the unloading time τ m,k of the edge computing task unloaded by the mth user terminal to the kth UAV from the preset maximum completion time delay T of the edge computing task of the mth user terminal.

[0064] The total computing power consumed by the edge computing tasks of all user terminals to be unloaded to each UAV is obtained by adding the computing amount of the edge computing task of each user terminal to be unloaded to each UAV divided by the completion time.

[0065] The computing power consumed by the edge computing task of the kth user terminal unloaded to the mth UAV is l The total computing power consumed by all user terminals for offloading edge computing tasks to each UAV is

[0066] On the basis of the above-mentioned embodiments, in the present embodiment, before the completion time of each user terminal offloading the edge computing task to each UAV is obtained according to the preset maximum completion delay and offloading time of each user terminal offloading the edge computing task to each UAV, the present embodiment further comprises: obtaining the transmission rate of each user terminal transmitting the edge computing task to each UAV according to the subchannel bandwidth of each user terminal, the transmission power of transmitting the edge computing task to each UAV, the channel power gain between each user terminal and each UAV, and the receiving noise power of each UAV.

[0067] The user terminal uploads the edge computing task to the UAV in an FDMA (Frequency Division Multiple Access) manner, and the subchannel bandwidth of each user terminal is B w The transmission rate of the kth user terminal transmitting the edge computing task to the mth UAV is:

[0068]

[0069] wherein R m,k is the rate between the kth user terminal and the mth UAV, is the receiving noise power of the UAV, P m,k is the transmission power of the kth user terminal transmitting the edge computing task to the mth UAV, and h m,k is the channel power gain between the kth user terminal and the mth UAV.

[0070] The size of the edge computing task of each user terminal to be offloaded to each UAV is divided by the rate to obtain the offloading time of each user terminal offloading the edge computing task to each UAV.

[0071] The offloading time of the mth user terminal offloading the edge computing task to the kth UAV is

[0072] Based on the above embodiments, this embodiment further includes: calculating the distance between each user terminal and each drone based on the position of each user terminal and the position of each drone; and calculating the channel power gain between each user terminal and each drone based on the height of each drone, the distance between each drone and each user terminal, and the preset reference power.

[0073] The channel power gain between the mth UAV and the kth user terminal is:

[0074]

[0075] Wherein, β0 is the preset reference power.

[0076] Based on the above embodiments, the remaining computing power of each drone is greater than or equal to 0;

[0077]

[0078] Each user terminal can offload the edge computing task to at most one drone;

[0079]

[0080] The number of user terminals to offload edge computing tasks to each drone is less than or equal to the number of sub-channels N of each drone;

[0081]

[0082] The horizontal coordinate of each UAV is greater than or equal to the minimum value of the horizontal coordinates of all user terminals, and less than or equal to the maximum value of the horizontal coordinates of all user terminals;

[0083] x min ≤q m,x ≤x max

[0084] The vertical coordinate of each UAV is greater than or equal to the minimum value of the vertical coordinates of all user terminals, and less than or equal to the maximum value of the vertical coordinates of all user terminals.

[0085] y min ≤q m,y ≤y max ,m∈Θ

[0086] The optimization problem of maximizing the coverage of drone swarm service is formulated as follows:

[0087]

[0088]

[0089]

[0090] C3:x min ≤q m，x ≤x max ,y min ≤q m，y ≤y max , m∈Θ

[0091] Among them, constraint C1 means that the idle computing power of each drone can complete the edge computing tasks offloaded to it by each user terminal within its specified time. Constraint C2 is μ m，k The value restriction of , constraint C3 is the flight range constraint of the UAV.

[0092] Based on the above embodiments, Figure 3 As shown, the objective function is solved in this embodiment to obtain the drone to which the edge computing task to be offloaded from each user terminal includes: introducing a non-negative slack variable into the objective function;

[0093] Since the objective function is a non-convex problem, a non-negative slack variable β is introduced and the position of the drone is fixed, resulting in the following optimization problem:

[0094]

[0095] stC1, C2,

[0096]

[0097] Converting the objective function after introducing the non-negative slack variables into a standard form;

[0098] The objective function after introducing non-negative slack variables is a mixed integer linear programming problem, which is transformed into the following standard form:

[0099]

[0100] stAx≤b

[0101] x(i)∈{0, 1}, x(MK+1)≥0, i∈[1, MK]

[0102] Among them, the quantities are:

[0103]

[0104] B K×(MK+1) The generated pseudo code is as follows:

[0105]

[0106] C 1×(MK+1) The generated pseudo code is as follows:

[0107]

[0108] D M×(MK+1) The generated pseudo code is as follows:

[0109]

[0110]

[0111] E M×(MK+1) The generated pseudo code is as follows:

[0112]

[0113] F M×(MK+1) The generated pseudo code is as follows:

[0114]

[0115] Solve the objective function converted into a standard form based on the branch and bound method to obtain the drone to which the edge computing task is to be offloaded from each user terminal;

[0116] After being converted into a standard mixed integer linear programming problem, the branch and bound method can be used to solve it. The genetic algorithm is then used to update the drone's position, and the iterations are repeated until convergence or the maximum number of iterations is reached.

[0117] If the objective function converted into the standard form does not converge and the number of iterations does not reach the preset value, the drone to which the edge computing task to be offloaded from each user terminal in this solution is updated based on the genetic algorithm;

[0118] The objective function converted into the standard form is solved again according to the updated UAV until the objective function converted into the standard form converges or the number of iterations reaches the preset value.

[0119] The multi-UAV edge computing resource scheduling method provided in this embodiment is simulated, and the CPU (Central Processing Unit) operating frequency of the UAV is used as the computing power measurement indicator. The main simulation parameters include x min =0,x max =50,y min =0,y max =50,fmax With p m,k 0.5GHz and 1W respectively. N=20,K=20,B w 100kHz, 10mW, l k is [0,10 6 ] Random value within bits, c k is 1000, f m =f max , θ is 1 / f max ,ω l =0.5.

[0120] Figure 4 The relationship curve between the drone service coverage and the maximum computing power of the drone and the number of drones is shown. It can be seen that the drone service coverage increases with the increase of its computing power and the number of drones. Figure 5 The relationship curve between the UAV load variance and the load weight is shown. It can be seen that as the load weight increases, the UAV load will gradually tend to a balanced state. Figure 4 and Figure 5 The effectiveness of the solution in this embodiment is proved.

[0121] The multi-UAV edge computing resource scheduling device provided by the present invention is described below. The multi-UAV edge computing resource scheduling device described below and the multi-UAV edge computing resource scheduling method described above can be referenced to each other.

[0122] like Figure 6 As shown, the device includes a calculation module 601, a construction module 602 and a solution module 603, wherein:

[0123] The calculation module 601 is used to subtract the total computing power consumed by the edge computing tasks to be offloaded to each drone by all user terminals from the idle computing power of each drone to obtain the remaining computing power of each drone;

[0124] The construction module 602 is used to construct an objective function according to the number of user terminals to offload edge computing tasks to each drone and the remaining computing power of each drone;

[0125] The solution module 603 is used to solve the objective function and obtain the drone to which each user terminal is to offload the edge computing task, so that the total number of user terminals corresponding to all drones is maximized and the maximum value of the remaining computing power of the drone is minimized.

[0126] This embodiment constructs an objective function based on the number of user terminals to offload edge computing tasks to each drone and the remaining computing power of each drone. The objective function is solved to optimize the offloading decision of the user terminal, thereby maximizing the drone edge computing service range while taking into account the load balancing of the drone. It is suitable for real-world scenarios where there are a large number of ground users and complex tasks, the drone computing power is limited, and it may be difficult to provide services to all users.

[0127] Figure 7 An example of a physical structure diagram of an electronic device is shown below. Figure 7 As shown, the electronic device may include: a processor 710, a communication interface 720, a memory 730, and a communication bus 740, wherein the processor 710, the communication interface 720, and the memory 730 communicate with each other through the communication bus 740. The processor 710 can call the logic instructions in the memory 730 to execute a multi-UAV edge computing resource scheduling method, which includes: subtracting the total computing power consumed by the edge computing tasks to be offloaded to each UAV by all user terminals from the idle computing power of each UAV to obtain the remaining computing power of each UAV; constructing an objective function based on the number of user terminals to offload edge computing tasks to each UAV and the remaining computing power of each UAV; solving the objective function to obtain the UAV to which each user terminal is to offload edge computing tasks, so that the total number of user terminals corresponding to all UAVs is maximized and the maximum value of the remaining computing power of the UAV is minimized.

[0128] In addition, the logic instructions in the above-mentioned memory 730 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0129] On the other hand, the present invention also provides a computer program product, which includes a computer program, which can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the multi-UAV edge computing resource scheduling method provided by the above methods, which includes: subtracting the total computing power consumed by all user terminals to offload edge computing tasks to each UAV from the idle computing power of each UAV to obtain the remaining computing power of each UAV; constructing an objective function based on the number of user terminals to offload edge computing tasks to each UAV and the remaining computing power of each UAV; solving the objective function to obtain the UAV to which each user terminal to offload edge computing tasks is to be offloaded, so that the total number of user terminals corresponding to all UAVs is maximized and the maximum value of the remaining computing power of the UAV is minimized.

[0130] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the multi-UAV edge computing resource scheduling method provided by the above-mentioned methods, the method comprising: subtracting the total computing power consumed by the edge computing tasks to be offloaded to each UAV by all user terminals from the idle computing power of each UAV to obtain the surplus computing power of each UAV; constructing an objective function based on the number of user terminals to offload edge computing tasks to each UAV and the surplus computing power of each UAV; solving the objective function to obtain the UAV to which each user terminal to offload edge computing tasks is to be offloaded, so that the total number of user terminals corresponding to all UAVs is maximized and the maximum value of the surplus computing power of the UAV is minimized.

[0131] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0132] Those skilled in the art can clearly understand the technical solutions of the various embodiments from the above description of the embodiments, and the various embodiments can be implemented by means of software with the necessary general hardware platforms, and of course, can also be implemented by hardware. Based on such understanding, the above technical solutions, essentially or in other words, the part of the prior art that makes a contribution, can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a ROM / RAM, a magnetic disk, an optical disk, and the like, and includes a number of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0133] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for some technical features therein; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A multi-UAV edge computing resource scheduling method, characterized in that: include: Subtract the total computing power consumed by the edge computing tasks to be offloaded to each drone from all user terminals from the idle computing power of each drone to obtain the remaining computing power of each drone. Construct an objective function based on the number of user terminals to offload edge computing tasks to each drone and the remaining computing power of each drone; Solve the objective function to obtain the drone to which each user terminal offloads the edge computing task, so that the total number of user terminals corresponding to all drones is maximized and the maximum value of the remaining computing power of the drones is minimized; The objective function is: in, Represents the remaining computing power after the mth UAV resource scheduling; represents the number of user terminals to offload edge computing tasks to each drone; f m represents the idle computing power of the mth drone; a k represents the computing power consumed to complete the edge computing task of the kth user terminal, μ m，k =1 means that the kth user terminal offloads the edge computing task to the mth drone, μ m，k =0 means that the kth user terminal does not offload the edge computing task to the mth UAV; ω l >0 is the UAV load balancing weight, θ>0 is the normalization factor; M represents the number of UAVs, K represents the number of ground user terminals; q represents the coordinates of the UAV, and μ represents whether the user terminal offloads the edge computing task to the UAV.

2. The multi-UAV edge computing resource scheduling method according to claim 1, characterized in that: Before obtaining the remaining computing power of each drone by subtracting the total computing power consumed by the edge computing tasks to be offloaded to each drone by all user terminals from the idle computing power of each drone, the following steps are also included: Obtain the computational amount of the edge computing task to be offloaded from each user terminal to each drone based on the size and computational complexity of the edge computing task to be offloaded from each user terminal to each drone; Obtain the completion time of the edge computing task to be offloaded from each user terminal to each drone based on the preset maximum completion delay and offloading time of the edge computing task to be offloaded from each user terminal to each drone; The computing power consumed by the edge computing tasks to be offloaded from all user terminals to each drone is obtained by dividing the computing power of the edge computing tasks to be offloaded from all user terminals to each drone by the completion time and adding them up.

3. The multi-UAV edge computing resource scheduling method according to claim 2, characterized in that: Before obtaining the completion time of the edge computing task to be offloaded from each user terminal to each drone based on the preset maximum completion delay and offloading time of the edge computing task to be offloaded from each user terminal to each drone, the method further includes: Obtaining a transmission rate at which each user terminal sends the edge computing task to each drone based on the subchannel bandwidth of each user terminal, the transmission power for sending the edge computing task to each drone, the channel power gain between the user terminal and each drone, and the receiving noise power of each drone; The size of the edge computing task to be offloaded by each user terminal to each drone is divided by the rate to obtain the offloading time for each user terminal to offload the edge computing task to each drone.

4. The multi-UAV edge computing resource scheduling method according to claim 3 is characterized in that: Before obtaining the transmission rate at which each user terminal sends the edge computing task to each drone based on the subchannel bandwidth of each user terminal, the transmission power of sending the edge computing task to each drone, the channel power gain between the user terminal and each drone, and the receiving noise power of each drone, the method further includes: Calculate the distance between each user terminal and each drone based on the location of each user terminal and the location of each drone; The channel power gain between each user terminal and each drone is calculated based on the height of each drone, the distance between each drone and each user terminal, and the preset reference power.

5. The multi-UAV edge computing resource scheduling method according to any one of claims 1 to 4, characterized in that: The remaining computing power of each drone is greater than or equal to 0; Each user terminal can offload the edge computing task to at most one drone; The number of user terminals to offload edge computing tasks to each drone is less than or equal to the number of sub-channels per drone; The horizontal coordinate of each UAV is greater than or equal to the minimum value of the horizontal coordinates of all user terminals, and less than or equal to the maximum value of the horizontal coordinates of all user terminals; The vertical coordinate of each UAV is greater than or equal to the minimum value of the vertical coordinates of all user terminals, and less than or equal to the maximum value of the vertical coordinates of all user terminals.

6. The multi-UAV edge computing resource scheduling method according to any one of claims 1 to 4, characterized in that: Solving the objective function to obtain the drone to which the edge computing task is to be offloaded from each user terminal includes: Introducing non-negative slack variables into the objective function; Converting the objective function after introducing the non-negative slack variables into a standard form; Solve the objective function converted into a standard form based on the branch and bound method to obtain the drone to which the edge computing task is to be offloaded from each user terminal; If the objective function converted into the standard form does not converge and the number of iterations does not reach the preset value, the drone of each user terminal to be offloaded from the edge computing task solved in this solution is updated based on the genetic algorithm; The objective function converted into the standard form is solved again according to the updated UAV until the objective function converted into the standard form converges or the number of iterations reaches the preset value.

7. A multi-UAV edge computing resource scheduling device, characterized in that: include: The calculation module is used to subtract the total computing power consumed by the edge computing tasks to be offloaded to each drone by all user terminals from the idle computing power of each drone to obtain the remaining computing power of each drone; A construction module is used to construct an objective function based on the number of user terminals to offload edge computing tasks to each drone and the remaining computing power of each drone; A solution module, configured to solve the objective function and obtain a drone to which each user terminal is to offload edge computing tasks, so that the total number of user terminals corresponding to all drones is maximized and the maximum value of the remaining computing power of the drones is minimized; The objective function is: in, Represents the remaining computing power after the mth UAV resource scheduling; represents the number of user terminals to offload edge computing tasks to each drone; f m represents the idle computing power of the mth drone; a k represents the computing power consumed to complete the edge computing task of the kth user terminal, μ m，k =1 means that the kth user terminal offloads the edge computing task to the mth drone, μ m，k =0 means that the kth user terminal does not offload the edge computing task to the mth UAV; ω l >0 is the UAV load balancing weight, θ>0 is the normalization factor; M represents the number of UAVs, K represents the number of ground user terminals; q represents the coordinates of the UAV, and μ represents whether the user terminal offloads the edge computing task to the UAV.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the steps of the multi-UAV edge computing resource scheduling method as described in any one of claims 1 to 6 are implemented.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the multi-UAV edge computing resource scheduling method as described in any one of claims 1 to 6 are implemented.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the multi-UAV edge computing resource scheduling method as described in any one of claims 1 to 6 are implemented.

Images