Multi-unmanned aerial vehicle resource allocation method and device for edge calculation and nonlinear energy collection, electronic equipment and storage medium

Abstract

The invention discloses a multi-unmanned aerial vehicle resource allocation method and device for edge calculation and nonlinear energy collection, electronic equipment and a storage medium, and belongs to the technical field of unmanned aerial vehicles, and the method comprises the steps: constructing a calculation task model and a nonlinear energy collection model; building an unmanned aerial vehicle resource allocation model by taking unmanned aerial vehicle energy collection time, transmission power and calculation task unloading proportion as decision variables, taking total transmission power and unmanned aerial vehicle energy sufficiency as constraints and taking calculation total time and total transmission energy consumption weighted minimization as a target function; and solving the model by adopting a simulated annealing algorithm to obtain an optimal resource allocation scheme. In a scene in which the unmanned aerial vehicle assists the mobile edge calculation, energy collection of the unmanned aerial vehicle is considered, unified optimization is carried out, and further guidance can be provided for application of the unmanned aerial vehicle in the mobile edge calculation; an unmanned aerial vehicle nonlinear energy collection model which is more practical is established, and the accuracy and the application effect of the optimization model can be improved.

Description

Technical Field

[0001] The present invention relates to the field of unmanned aerial vehicle technology, and in particular to a multi-unmanned aerial vehicle resource allocation method, device, electronic device and storage medium for edge computing and nonlinear energy collection. Background Art

[0002] For mobile edge computing systems, drones can rescue data calculations with higher priority through some intelligent path routing and scheduling, thereby utilizing the incremental system resources brought by the use of drones. In other words, the energy consumption and computing power of drones themselves limit their application in mobile edge computing systems. At this time, some edge nodes can be set up in advance. These nodes can not only undertake part of the computing tasks, that is, drones can offload part of the computing tasks to edge nodes, but also provide wireless charging for drones, which can greatly alleviate the above two limitations. The integration of drones into mobile edge computing systems has attracted widespread attention from researchers in academia and industry.

[0003] However, due to the limited flight time and energy capacity of drones, it is necessary to combine edge computing with energy harvesting. Linearity / nonlinearity mainly refers to the relationship between the output current of a drone and its received power. In reality, the relationship between the output current of a drone and its received power is not linear, but previous work only considered a simplified linear energy harvesting model, which does not conform to the actual situation. Summary of the invention

[0004] The present invention provides a multi-UAV resource allocation method, device, electronic device and storage medium for edge computing and nonlinear energy collection to solve the problems mentioned in the background technology.

[0005] The present invention is achieved through the following technical solutions.

[0006] In a first aspect, the present invention provides a multi-UAV resource allocation method for edge computing and nonlinear energy harvesting, comprising: Build a computing task model and solve the total computing time of each drone; Construct a nonlinear energy harvesting model and solve it to obtain the harvesting power of each drone; The UAV resource allocation model is constructed with the UAV energy collection time, transmission power and computing task offloading ratio as decision variables, the total transmission power and UAV energy sufficiency as constraints, and the weighted minimization of the total computing time and the total transmission energy consumption as the objective function. The UAV energy sufficiency constraint is related to the collection power and the total computing time, and the objective function is related to the total computing time. The simulated annealing algorithm is used to solve the UAV resource allocation model and obtain the optimal resource allocation plan for multiple UAVs.

[0007] Furthermore, the computational task model is: Each drone is numbered to form a drone serial number set: , In the formula, express N A set of drone serial numbers; No. i The computing task scale of a UAV is ; No. i The computing power of a drone is ; No. i The proportion of computing tasks offloaded from drones to edge nodes is ; The computing power of edge nodes is ; The number of CPU cycles to execute a unit instruction is c ; No. i The local computation time of each drone is: ; No. i The unloading calculation time of each drone is: ; No. i The total computation time for each drone is: .

[0008] Furthermore, the nonlinear energy harvesting model is: The ideality factor of a diode is n ; The transmission power at the edge node is P ; The load resistance is ; The thermal voltage at the diode is ; The reverse bias saturation current is ; The antenna impedance is ; The channel power gain is ; The geographical location of the edge node is ; No. i The geographical location of the drone is ; Assign to i The transmission power of a UAV is ; The truncation order is ; No. i The Euclidean distance between a drone and an edge node: , In the formula, Indicates i The Euclidean distance between each drone and the edge node; No. i The received power of each drone is: ; No. i The approximate relationship between the output current of a drone and its received power is: , In the formula, Indicates i The output current of each drone, express of j Power, , , ,in, , , It is expressed in the Taylor expansion of the output current of the drone and its received power of j The coefficient before the power, represents the rectifier characteristic constant, represents the antenna impedance, express of j Power, The waveform factor represents the unit power. is the distribution of the input signal, and Represents its 2nd power and i Power, and They represent the power of 2 and i The mathematical expectation of a power; No. i The relationship between the collected power of a UAV and its output current is: , In the formula, Indicates i The collection power of a drone.

[0009] Furthermore, the UAV resource allocation model is: Constraints on the value of the decision variables for the drone energy collection time: , In the formula, Indicates i Energy collection time for each drone; Total transmission power constraint: , In the formula, Indicates that the i The transmission power of each drone; The value constraints of the decision variables for the proportion of unloading UAV computing tasks are: , In the formula, Indicates i The percentage of computing tasks offloaded from drones to edge nodes; Drone energy sufficiency constraints: , In the formula, , , and Respectively represent i The original energy, computing energy per unit time, hovering energy per unit time and return energy of each drone, Indicates i The collection power of a drone, Indicates i The local computation time of each drone is Indicates i The total computation time of each drone; Calculate the total time: ; Total transmission energy consumption: ; The objective function is to minimize the weighted total calculation time and total transmission energy consumption: , In the formula, Represents the weighted value of total calculation time and total transmission energy consumption, and is the weight coefficient.

[0010] Furthermore, the simulated annealing algorithm is used to solve the UAV resource allocation model: Step 1: Randomly generate a solution as the initial solution; Step 2: Given Temperature U A sufficiently large initial value U 0. Iterations L , M ; Step 3: For 1, ..., L Do steps 4 to 7; Step 4: Generate a new solution z' using the two-transformation method; Step 5: Calculate the difference between the objective function after transformation and before transformation ; Step 6: If Then accept z' as the new current solution, otherwise with probability Accept z' as the new current solution; Step 7: If continuous M If none of the new solutions are accepted, the current solution is output as the optimal solution and the process ends; Step 8: U Reduce and go to step 3.

[0011] In a second aspect, the present invention provides a device for implementing the above-mentioned multi-UAV resource allocation method for edge computing and nonlinear energy collection, the device comprising: The computing task model building module is used to build the computing task model and solve the total computing time of each UAV; The nonlinear energy harvesting model building module is used to build a nonlinear energy harvesting model and solve the harvesting power of each drone. The UAV resource allocation model construction module is used to construct the UAV resource allocation model with the UAV energy collection time, transmission power and computing task offloading ratio as decision variables, the total transmission power and UAV energy sufficiency as constraints, and the weighted minimization of the total computing time and the total transmission energy consumption as the objective function. The UAV energy sufficiency constraint is related to the collection power and the total computing time, and the objective function is related to the total computing time. The UAV resource allocation model solving module is used to solve the UAV resource allocation model using the simulated annealing algorithm to obtain the optimal resource allocation plan for multiple UAVs.

[0012] In a third aspect, the present invention provides an electronic device comprising a processor and a memory, wherein the memory is used to store program code, and the processor is used to execute the above-mentioned multi-UAV resource allocation method for edge computing and nonlinear energy collection according to instructions in the program code.

[0013] In a fourth aspect, the present invention provides a computer-readable storage medium for storing program code, wherein the program code is used to execute the above-mentioned multi-UAV resource allocation method for edge computing and nonlinear energy collection.

[0014] The present invention proposes a multi-UAV resource allocation method, device, electronic device and storage medium for edge computing and nonlinear energy collection. The method aims at a scenario in which an edge node performs auxiliary edge computing and provides wireless power supply for multiple UAVs. The method uses a computing task model and a nonlinear energy collection model, takes UAV energy collection time, transmission power and computing task offloading ratio as decision variables, takes total transmission power and UAV energy sufficiency as constraints, and takes weighted minimization of total computing time and total transmission energy consumption as the objective function to construct a UAV resource allocation model.

[0015] The beneficial effects of the present invention are: In the scenario of drone-assisted mobile edge computing, the present invention takes into account the energy collection of drones and performs unified optimization, which can provide further guidance for the application of drones in mobile edge computing; a more practical drone nonlinear energy collection model is established, which can improve the accuracy and application effect of the optimization model.

[0016] Linearity / non-linearity mainly refers to the relationship between the output current of the drone and its received power. In reality, the output current of the drone and its received power are not in a linear relationship. The present invention more accurately analyzes the relationship between the output current, received power and its own physical properties (such as the ideal factor of the diode, the thermal voltage at the diode, the reverse bias saturation current, etc.) of the drone, rather than simply assuming that the output current and received power are linear. The use of a more realistic nonlinear energy collection model can more accurately guide the optimization of transmission power allocation. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 This is a flow chart of the multi-UAV resource allocation method described in Example 1; Figure 2 This is a schematic diagram of a scenario of the multi-UAV resource allocation method described in Example 1; Figure 3 A schematic diagram of a task scenario of the multi-UAV resource allocation method according to Example 1; Figure 4 Schematic diagram of the optimal solution for the energy collection time, transmission power and computing task offloading ratio of the drone in Example 1; Figure 5 This is a schematic diagram of the structure of the multi-UAV resource allocation device described in Example 2. DETAILED DESCRIPTION

[0018] The following further describes the structures involved in the present invention or the technical terms used in these descriptions. These descriptions are only used as examples to illustrate how the methods of the present invention are implemented, and cannot constitute any limitation to the present invention. Embodiment 1

[0019] Figure 1 and Figure 2 The flowchart and scenario diagram of a multi-UAV resource allocation method for edge computing and nonlinear energy harvesting are respectively shown. This embodiment now introduces a multi-UAV resource allocation method for edge computing and nonlinear energy harvesting, including:

[0020] S110, constructing a computing task model, and solving the total computing time of each UAV.

[0021] A computing task model is constructed based on the computing task scale of the drone, the computing power of the drone, the computing power of the edge nodes, the proportion of computing tasks offloaded from the drone to the edge nodes, and the number of CPU cycles to execute a unit instruction.

[0022] The computing task of a drone refers to the fact that in many application scenarios of drones, the drone needs to process and analyze the data it collects in real time. For example, a drone needs to analyze the images it captures to identify whether a device has defects. Image analysis is essentially a computing process, so it can be called a drone computing task.

[0023] The size of a computing task refers to the amount of data that needs to be processed in a computing task.

[0024] Computing power refers to the main frequency, that is, the frequency of the CPU's internal crystal oscillator, in MHz.

[0025] The number of CPU cycles to execute a unit instruction, or CPI, refers to the average number of clock cycles required to execute an instruction.

[0026] The specific computing task model is: Each drone is numbered to form a drone serial number set: , In the formula, express N In this embodiment, N =3, the drone is Figure 3 As shown in the circle in the figure, the serial number of the drone is marked in the circle.

[0027] Above N Among the drones, i The computing task scale of a UAV is In this embodiment, l 1= l 2=50, l 3=100, Figure 3 The circle on the left side indicates the scale of the drone’s computing tasks.

[0028] Above N Among the drones, i The computing power of a drone is In this embodiment, f 1=100, f 2=120, f 3=150, Figure 3 The circle on the right side of the image shows the computing power of the drone.

[0029] Above N Among the drones, i The proportion of computing tasks offloaded from drones to edge nodes is .

[0030] The computing power of edge nodes is In this embodiment, , the edge nodes are Figure 3 As shown in the square in the figure, the computing power of the edge node is indicated below the square, the dotted arrow pointing from the circle to the square represents computing offloading, and the solid arrow pointing from the square to the circle represents energy harvesting.

[0031] The number of CPU cycles to execute a unit instruction is c In this embodiment, c =3.

[0032] No. i The local computation time of each drone is: , In the formula, Indicates i The local computing time of each drone. The local computing time is a function of the computing task offloading ratio, and the computing task offloading ratio is a decision variable. The drone offloads part of the computing tasks to the edge node, and the remaining computing tasks are calculated by the drone itself. The local computing time of the drone refers to the time required for the drone to calculate itself.

[0033] No. i The unloading calculation time of each drone is: , In the formula, Indicates i The offloading computation time of each drone. The offloading computation time is a function of the offloading ratio of computing tasks, and the offloading ratio of computing tasks is a decision variable. The drone offloads part of the computing tasks to the edge node, and the remaining computing tasks are calculated by the drone itself. The offloading computation time of the drone refers to the time required for the edge node to calculate.

[0034] No.i The total computation time for each drone is: , In the formula, Indicates i The total computation time of each drone.

[0035] S120. Construct a nonlinear energy collection model and solve the collection power of each UAV.

[0036] A nonlinear energy harvesting model is constructed based on the physical characteristics of the drone, the geographic location of the drone, the geographic location of the edge nodes, the transmission power at the edge nodes, and the transmission power of the drone.

[0037] The physical characteristics of the drone refer to the ideality factor of the diode, the load resistance, the thermal voltage at the diode, the reverse bias saturation current, the antenna impedance, and the channel power gain.

[0038] The nonlinear energy harvesting model is specifically: The ideality factor of a diode is n In this embodiment, n =1.5.

[0039] The transmission power at the edge node is P In this embodiment, P =20, Figure 3 The square in indicates the transmission power at the edge node.

[0040] The load resistance is In this embodiment, .

[0041] The thermal voltage at the diode is In this embodiment, .

[0042] The reverse bias saturation current is In this embodiment, .

[0043] The antenna impedance is In this embodiment, .

[0044] The channel power gain is In this embodiment, .

[0045] The geographical location of the edge node is In this embodiment, .

[0046] No. i The geographical location of the drone is In this embodiment, , , .

[0047] Assign to i The transmission power of a UAV is .

[0048] The truncation order is In this embodiment .

[0049] No. i The Euclidean distance between a drone and an edge node: , In the formula, Indicates i The Euclidean distance between each drone and the edge node, Indicates i The geographic location of the drone.

[0050] No. i The received power of each drone is: , In the formula, Indicates i The received power of a UAV.

[0051] No. i The approximate relationship between the output current of a drone and its received power is: , In the formula, Indicates i The output current of each drone, express of j Power, , , ,in, , , It is expressed in the Taylor expansion of the output current of the drone and its received power of j The coefficient before the power, represents the rectifier characteristic constant, represents the antenna impedance, express of j Power, The waveform factor represents the unit power. is the distribution of the input signal, and Represents its 2nd power and i Power, and They represent the power of 2 and i In this embodiment, is a Gaussian distribution with mean 1 and variance 0.1.

[0052] No. i The relationship between the collected power of a UAV and its output current is: , In the formula, Indicates i The collection power of a drone.

[0053] S130. Construct a UAV resource allocation model with UAV energy collection time, transmission power and computing task offloading ratio as decision variables, total transmission power and UAV energy sufficiency as constraints, and weighted minimization of total computing time and total transmission energy consumption as the objective function, wherein the UAV energy sufficiency constraint is related to the collection power and total computing time, and the objective function is related to the total computing time.

[0054] The transmission power in the decision variable refers to the transmission power allocated by the edge node to each drone. The total transmission power of the edge node is fixed, so in an edge computing system where multiple drones need edge nodes to transmit energy for them, a decision on the allocation of transmission power must be made. The total transmission power constraint means that the sum of the transmission power allocated by the edge node to each drone is equal to its total transmission power.

[0055] The specific construction of the drone resource allocation model is as follows: Constraints on the value of the decision variables for the drone energy collection time: , In the formula, Indicates i Energy harvesting time for each drone.

[0056] Total transmission power constraint: , In the formula, Indicates that the i The transmission power of a UAV.

[0057] The value constraints of the decision variables for the proportion of unloading UAV computing tasks are: , In the formula, Indicates i The percentage of computing tasks offloaded from drones to edge nodes.

[0058] Drone energy sufficiency constraints: , In the formula, , , and Respectively represent i The original energy, computing energy per unit time, hovering energy per unit time and return energy of each drone, Indicates i The collection power of a drone, Indicates i The local computation time of each drone is Indicates i The total computation time of each drone. In this embodiment, , , , .

[0059] Calculate the total time: , In the formula, t Indicates the total calculation time, Indicates i The total computation time of each drone is Indicates i Energy harvesting time for each drone.

[0060] Total transmission energy consumption: , In the formula, represents the total transmission energy consumption, Indicates i Energy collection time of each drone, Indicates that the i The transmission power of a UAV.

[0061] The objective function is to minimize the weighted total calculation time and total transmission energy consumption: , In the formula, Represents the weighted value of total calculation time and total transmission energy consumption, represents the total transmission energy consumption, and is the weight coefficient. In this embodiment, , .

[0062] So far, we have obtained a UAV resource allocation model with UAV energy collection time, transmission power and computing task offloading ratio as decision variables, total transmission power and UAV energy sufficiency as constraints, and weighted minimization of total computing time and total transmission energy consumption as the objective function. , local computing time , calculate the total time The objective function is related to the total calculation time. t Regarding, the total time t The total time taken to calculate related.

[0063] S140. Use simulated annealing algorithm to solve the UAV resource allocation model and obtain the optimal resource allocation plan for multiple UAVs.

[0064] It should be noted that the meaning of the same parameters involved in all the above formulas is consistent. For parameters whose meanings are not directly stated, they can be found in the formulas in the context.

[0065] The above-mentioned drone resource allocation model belongs to a nonlinear non-convex programming model, so this embodiment adopts a simulated annealing algorithm to solve it, and the specific steps are as follows: S1401. Randomly generate a solution as an initial solution.

[0066] S1402, given temperature U A sufficiently large initial value U 0. In this embodiment, U 0=100, number of iterations L =8, given M =8; among which, M is a parameter in the algorithm. When continuous M If none of the new solutions are accepted, the current solution is output as the optimal solution.

[0067] S1403, Yes L =1,…,8 do S1404 to S1407.

[0068] S1404. Generate a new solution z' using the two-transformation method.

[0069] S1405: Calculate the difference between the objective function after transformation and the objective function before transformation .

[0070] S1406, if Then accept z' as the new current solution, otherwise with probability Accept z' as the new current solution.

[0071] S1407, if continuous M=If none of the 8 new solutions are accepted, the current solution is output as the optimal solution and the process ends.

[0072] S1408, Temperature U Reduce and go to S1403.

[0073] In this embodiment, the weighted minimum value of the total transmission energy consumption and the total calculation time can be solved to be 12.37, and the following can be solved: Figure 4 The optimal solution for drone energy collection time, transmission power and computing task offloading ratio is shown. The number next to the circle indicates the energy collection time of the corresponding drone, the number on the solid arrow indicates the transmission power, and the number on the dotted arrow indicates the computing task offloading ratio. Embodiment 2

[0074] This embodiment introduces a multi-UAV resource allocation device for edge computing and nonlinear energy harvesting. Figure 5 The multi-UAV resource allocation device implements the multi-UAV resource allocation method described in Example 1. The multi-UAV resource allocation device can be implemented in the form of hardware and / or software, and the multi-UAV resource allocation device can be configured in an electronic device. The multi-UAV resource allocation device includes: a computing task model construction module 210, a nonlinear energy collection model construction module 220, a UAV resource allocation model construction module 230, and a UAV resource allocation model solving module 240.

[0075] The computing task model building module 210 is used to build a computing task model and solve the total computing time of each UAV.

[0076] The nonlinear energy collection model construction module 220 is used to construct a nonlinear energy collection model to solve and obtain the collection power of each UAV.

[0077] The UAV resource allocation model construction module 230 is used to construct a UAV resource allocation model with the UAV energy collection time, transmission power and computing task offloading ratio as decision variables, the total transmission power and UAV energy sufficiency as constraints, and the weighted minimization of the total calculation time and the total transmission energy consumption as the objective function, wherein the UAV energy sufficiency constraint is related to the collection power and the total calculation time, and the objective function is related to the total calculation time.

[0078] The UAV resource allocation model solving module 240 is used to solve the UAV resource allocation model using a simulated annealing algorithm to obtain an optimal resource allocation solution for multiple UAVs.

[0079] Through the multi-UAV resource allocation device, the multi-UAV resource allocation method for edge computing and nonlinear energy collection described in Example 1 can be implemented. Embodiment 3

[0080] This embodiment also introduces an electronic device, which includes a memory and a processor, the memory is used to store computer programs, the processor is used to execute computer programs, and the computer program, when executed by the processor, implements the steps of the multi-UAV resource allocation method provided in any of the above embodiments.

[0081] The processor may include one or more processing cores, such as a 3-core processor, an 8-core processor, etc. The processor may be implemented in at least one hardware form of digital signal processing (DSP), field-programmable gate array (FPGA), and programmable logic array (PLA). The processor may also include a main processor and a coprocessor, the main processor is a processor for processing data in the awake state, also known as a central processing unit (CPU); the coprocessor is a low-power processor for processing data in the standby state. The memory may include one or more storage media. The memory may also include a high-speed random access memory and a non-volatile memory, such as one or more disk storage devices and flash memory storage devices. In this embodiment, the memory is at least used to store the following computer program, wherein, after the computer program is loaded and executed by the processor, the relevant steps in the multi-UAV resource allocation method disclosed in any of the aforementioned embodiments can be implemented. In addition, the resources stored in the memory may also include an operating system and data, etc., and the storage method may be short-term storage or permanent storage. The operating system may be Windows or other types of operating systems. The data may include but is not limited to the data involved in the above method. Embodiment 4

[0082] This embodiment also introduces a computer-readable storage medium, on which program code is stored. When the program code is executed by a processor, the steps of the multi-UAV resource allocation method provided in any of the above embodiments can be implemented.

[0083] The storage medium may include: 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, and other media that can store program codes.

[0084] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only illustrative, for example, the division of the units is only a logical function division, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed.

[0085] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0086] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A multi-UAV resource allocation method for edge computing and nonlinear energy harvesting, characterized in that: include: Build a computing task model and solve the total computing time of each drone; Construct a nonlinear energy harvesting model and solve it to obtain the harvesting power of each drone; The UAV resource allocation model is constructed with the UAV energy collection time, transmission power and computing task offloading ratio as decision variables, the total transmission power and UAV energy sufficiency as constraints, and the weighted minimization of the total computing time and the total transmission energy consumption as the objective function. The UAV energy sufficiency constraint is related to the collection power and the total computing time, and the objective function is related to the total computing time. The simulated annealing algorithm is used to solve the UAV resource allocation model and obtain the optimal resource allocation plan for multiple UAVs.

2. The multi-UAV resource allocation method for edge computing and nonlinear energy harvesting according to claim 1 is characterized in that: The calculation task model is: Each drone is numbered to form a drone serial number set: ， In the formula, express N A set of drone serial numbers; No. i The computing task scale of a UAV is ; No. i The computing power of a drone is ; No. i The proportion of computing tasks offloaded from drones to edge nodes is ; The computing power of edge nodes is ; The number of CPU cycles to execute a unit instruction is c ; No. i The local computation time of each drone is: ; No. i The unloading calculation time of each drone is: ; No. i The total computation time for each drone is: .

3. The multi-UAV resource allocation method for edge computing and nonlinear energy harvesting according to claim 2 is characterized in that: The nonlinear energy harvesting model is: The ideality factor of a diode is n ; The transmission power at the edge node is P ; The load resistance is ; The thermal voltage at the diode is ; The reverse bias saturation current is ; The antenna impedance is ; The channel power gain is ; The geographical location of the edge node is ; No. i The geographical location of the drone is ; Assign to i The transmission power of a UAV is ; The truncation order is ; No. i The Euclidean distance between a drone and an edge node: ， In the formula, Indicates i The Euclidean distance between each drone and the edge node; No. i The received power of each drone is: ; No. i The approximate relationship between the output current of a drone and its received power is: ， In the formula, Indicates i The output current of each drone, express of j Power, , , ,in, , , It is expressed in the Taylor expansion of the output current of the drone and its received power of j The coefficient before the power, represents the rectifier characteristic constant, represents the antenna impedance, express of j Power, The waveform factor represents the unit power. is the distribution of the input signal, and Represents its 2nd power and i Power, and They represent the power of 2 and i The mathematical expectation of a power; No. i The relationship between the collected power of a UAV and its output current is: ， In the formula, Indicates i The collection power of a drone.

4. The multi-UAV resource allocation method for edge computing and nonlinear energy harvesting according to claim 3 is characterized in that: The UAV resource allocation model is: Constraints on the value of the decision variables for the drone energy collection time: ， In the formula, Indicates i Energy collection time for each drone; Total transmission power constraint: ， In the formula, Indicates that the i The transmission power of each drone; The value constraints of the decision variables for the proportion of unloading UAV computing tasks are: ， In the formula, Indicates i The percentage of computing tasks offloaded from drones to edge nodes; Drone energy sufficiency constraints: ， In the formula, , , and Respectively represent i The original energy, computing energy per unit time, hovering energy per unit time and return energy of each drone, Indicates i The collection power of a drone, Indicates i The local computation time of each drone is Indicates i The total computation time of each drone; Calculate the total time: ; Total transmission energy consumption: ; The objective function is to minimize the weighted total calculation time and total transmission energy consumption: ， In the formula, Represents the weighted value of total calculation time and total transmission energy consumption, and is the weight coefficient.

5. The multi-UAV resource allocation method for edge computing and nonlinear energy harvesting according to claim 1 is characterized in that: The simulated annealing algorithm is used to solve the UAV resource allocation model: Step 1: Randomly generate a solution as the initial solution; Step 2: Given Temperature U A sufficiently large initial value U 0. Iterations L , M ; Step 3: For 1, ..., L Do steps 4 to 7; Step 4: Generate a new solution z' using the two-transformation method; Step 5: Calculate the difference between the objective function after transformation and before transformation ; Step 6: If Then accept z' as the new current solution, otherwise with probability Accept z' as the new current solution; Step 7: If continuous M If none of the new solutions are accepted, the current solution is output as the optimal solution and the process ends; Step 8: U Reduce and go to step 3.

6. A device, characterized in that: The device is used to implement the multi-UAV resource allocation method for edge computing and nonlinear energy collection according to any one of claims 1 to 5, comprising: The computing task model building module is used to build the computing task model and solve the total computing time of each UAV; The nonlinear energy harvesting model building module is used to build a nonlinear energy harvesting model and solve the harvesting power of each drone. The UAV resource allocation model construction module is used to construct the UAV resource allocation model with the UAV energy collection time, transmission power and computing task offloading ratio as decision variables, the total transmission power and UAV energy sufficiency as constraints, and the weighted minimization of the total computing time and the total transmission energy consumption as the objective function. The UAV energy sufficiency constraint is related to the collection power and the total computing time, and the objective function is related to the total computing time. The UAV resource allocation model solving module is used to solve the UAV resource allocation model using the simulated annealing algorithm to obtain the optimal resource allocation plan for multiple UAVs.

7. An electronic device, characterized in that: It includes a processor and a memory, the memory is used to store program code, and the processor is used to execute the multi-UAV resource allocation method for edge computing and nonlinear energy collection according to any one of claims 1 to 5 according to instructions in the program code.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium is used to store program code, and the program code is used to execute the multi-UAV resource allocation method for edge computing and nonlinear energy collection as described in any one of claims 1-5.

Images