Edge computing system optimization method and system based on user equipment and multiple unmanned aerial vehicles
Through multi-dimensional collaboration assisted mobile edge computing system, relay drones collect and forward data, optimize calculation frequency and communication scheduling, solving the problems of high energy consumption and poor system stability of a single drone, and minimizing system energy consumption and Improved stability.
Patent Information
- Application Number
- CN202411570711.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-06
- Publication Date
- 2024-12-03
AI Technical Summary
In the existing drone-assisted mobile edge computing system, a single drone needs to consume a lot of energy for data reception and calculation, resulting in poor fault tolerance and weak system stability, and high deployment and maintenance costs of edge servers.
Design a multi-dimensional collaborative assisted mobile edge computing system, and optimize the calculation frequency by forming a system based on ground user equipment and multi-UAVs, using relay drones to collect data and forward it to the computing drone for processing. , communication scheduling and data offloading strategies to reduce drone flight range and energy consumption and improve system stability.
It minimizes system energy consumption, improves system stability and fault tolerance, extends the working cycle of drones, and reduces the deployment and maintenance costs of edge servers.
Smart Images

Figure CN119071829A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the field of mobile edge computing technology, and in particular, relates to an edge computing system optimization method and system based on user equipment and multiple drones. Background Art
[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.
[0003] Mobile Edge Computing (MEC) is a distributed computing model that aims to push computing and data storage capabilities to the edge of the network, thereby processing data and providing services more efficiently. Mobile edge computing is considered to be a promising technology for providing computing-intensive services to user devices (UDs); by offloading the data to be processed to the MEC server, the real-time computing capabilities of user devices can be effectively enhanced, the computing latency and energy consumption of user devices can be reduced, and the life cycle of user devices can be further extended.
[0004] However, the coverage of mobile edge computing systems is usually limited by the location of edge servers, which may cause mobile edge computing services to be unavailable in certain geographic areas and the deployment and maintenance costs of edge servers to be too high. In recent years, drones equipped with MEC servers have been widely used, considering their advantages such as low cost and high spatiotemporal flexibility. Compared with traditional MEC servers, the deployment of drones equipped with MEC servers can expand the service range, get closer to user devices, and provide more reliable communication and computing services.
[0005] Most existing research on drone-assisted MEC systems is based on the assumption that user devices offload computing data directly to drones for processing. This direct data offloading method is not conducive to energy-constrained drones, because when the deployment area of user devices is widely distributed, drones need to consume a lot of flight energy to achieve reliable communication with all user devices. Therefore, some researchers have studied the design of drone-assisted MEC systems that relay data between ground user devices. However, if only a single drone is considered to provide MEC services, a single drone consumes energy not only to receive data transmitted by user devices, but also to calculate the data of user devices, which will lead to problems such as poor fault tolerance and weak system stability. Summary of the invention
[0006] In order to address the deficiencies of the prior art, the present invention provides an edge computing system optimization method, system, electronic device, computer-readable storage medium and computer program product based on user equipment and multiple drones, which can reduce the range that drones need to fly in order to perform ground unloading tasks, thereby correspondingly reducing the flight energy consumption of the drones, and can also enhance the stability of the system.
[0007] To achieve the above objectives, one or more embodiments of the present invention provide the following technical solutions: In a first aspect, the present invention provides an edge computing system optimization method based on user equipment and multiple drones, comprising: Establish a multi-dimensional collaborative assisted mobile edge computing system based on ground user equipment and drones; Constructing the objective function and constraint conditions of the energy consumption optimization of the multi-dimensional collaborative assisted mobile edge computing system, and solving the objective function, including: setting an iteration termination condition, giving the trajectory of the relay drone, decomposing the problem of solving the objective function into a system scheduling sub-problem and a trajectory optimization sub-problem and solving them iteratively; Under the condition of a fixed relay UAV trajectory, the Lagrangian dual method is used to solve the system scheduling subproblem until the iteration termination condition is met, and the optimal computing frequency, communication scheduling and data unloading strategy are obtained; based on the optimized computing frequency, communication scheduling and data unloading strategy, the continuous convex approximation method is used to solve the trajectory optimization subproblem until the iteration termination condition is met, and the optimal relay UAV trajectory is obtained.
[0008] In a second aspect, the present invention provides an edge computing system optimization system based on user equipment and multiple drones, comprising: The edge computing system building module is configured to: build a multi-dimensional collaborative auxiliary mobile edge computing system based on ground user equipment and drones; The edge computing system optimization problem solving module is configured to: construct the objective function and constraint conditions of the energy consumption optimization of the multi-dimensional collaborative assisted mobile edge computing system, and solve the objective function, including: setting an iteration termination condition, giving a relay drone trajectory, decomposing the objective function solution problem into a system scheduling sub-problem and a trajectory optimization sub-problem and performing iterative solutions; Under the condition of a fixed relay UAV trajectory, the Lagrangian dual method is used to solve the system scheduling subproblem until the iteration termination condition is met, and the optimal computing frequency, communication scheduling and data unloading strategy are obtained; based on the optimized computing frequency, communication scheduling and data unloading strategy, the continuous convex approximation method is used to solve the trajectory optimization subproblem until the iteration termination condition is met, and the optimal relay UAV trajectory is obtained.
[0009] In a third aspect, the present invention provides an electronic device comprising a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the above-mentioned edge computing system optimization method based on user equipment and multiple drones.
[0010] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program / instruction stored thereon, which, when executed by a processor, implements the steps of the above-mentioned edge computing system optimization method based on user equipment and multiple drones.
[0011] In a fifth aspect, the present invention provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of the above-mentioned edge computing system optimization method based on user equipment and multiple drones.
[0012] One or more of the above technical solutions have the following beneficial effects: (1) The present invention provides a new multi-dimensional collaborative assisted mobile edge computing system framework, which uses relay drones to collect computing data from ground user devices and forwards it to computing drones for processing. Compared with a single drone that consumes energy both to receive data transmitted by user devices and to calculate the data of user devices, the present invention designs a method for two drones to divide the work and cooperate, thereby improving the stability and fault tolerance of the system.
[0013] (2) The present invention decomposes the energy consumption optimization problem of the multi-dimensional collaborative assisted mobile edge computing system into two sub-problems: system scheduling and trajectory optimization. The two sub-problems are iteratively solved to obtain the optimized relay UAV trajectory, computing frequency, communication scheduling and data unloading strategy. The total energy consumption of the system is minimized by optimizing the relay UAV trajectory, computing frequency, communication scheduling and data unloading strategy.
[0014] (3) In the multi-dimensional collaborative assisted mobile edge computing system provided by the present invention, the remote user device does not need to communicate directly with the drone, but only needs to unload the data to be processed to the relay drone with the assistance of the relay user device. Since the relay user device is close to the relay drone, this can not only improve the communication stability between the relay drone and the user device, but also reduce the flight range required for the relay drone to collect user device data, which in turn reduces the flight energy consumption of the relay drone and prolongs the working cycle of the relay drone.
[0015] Advantages of additional aspects of the present invention will be given in part in the following description, and in part will become obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS
[0016] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.
[0017] Figure 1 This is a multi-dimensional collaboration framework between user equipment and drones according to the first embodiment.
[0018] Figure 2 This is a four-stage parallel communication scheme of the first embodiment.
[0019] Figure 3 This is a flowchart of iterative solution of the objective function of the first embodiment.
[0020] Figure 4 This is a comparison diagram of energy consumption of the first embodiment. DETAILED DESCRIPTION
[0021] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used in the present invention have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.
[0022] It should be noted that the terms used herein are only for describing specific embodiments, and are not intended to limit exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that the terms "include" and "have" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.
[0023] In the absence of conflict, the embodiments of the present invention and the features of the embodiments may be combined with each other.
[0024] Embodiment 1 This embodiment discloses an edge computing system optimization method based on user equipment and multiple drones, including: Step 1: Establish a multi-dimensional collaborative assisted mobile edge computing system based on ground user equipment and drones; Step 2: Construct the objective function and constraints for energy consumption optimization of multi-dimensional collaborative assisted mobile edge computing system; Step 3: Solve the objective function to obtain the optimal relay drone trajectory, calculation frequency, communication scheduling, and data offloading strategy; Step 1 includes, for example Figure 1As shown, the present invention proposes a MEC system framework based on the multi-dimensional collaboration between user devices and drones, which consists of two heterogeneous drones, M relay user devices (within the coverage of the drones) and M remote user devices (outside the coverage of the drones), wherein each ground user device and drone are equipped with a single antenna.
[0025] In this framework, one drone (i.e., the relay drone) acts as a relay server flying in the air to collect data from relay user devices, and another drone (i.e., the computing drone) acts as a MEC server hovering in the air to process the data collected by the relay drone. At the same time, the ground control base is deployed on the ground and connected to the computing drone through a tethered antenna to provide energy to the computing drone; The specific data offloading process in the framework is as follows: First, the remote user device offloads the task data to the relay user device. Second, the relay user device relays the task data received from the remote user device and its own task data to the relay drone. Finally, the relay drone forwards the received data to the computing drone for processing.
[0026] In the user device, Represents a collection of relay user devices, and Represents a set of remote user equipment assisted by a relay user equipment (one relay user equipment assists one remote user equipment); Assume that the computing drone is hovering at a fixed height of H meters in the air, and the relay drone is flying at a fixed height of H meters in the air. All user devices are deployed on the ground at a height of 0 meters; By using a three-dimensional Cartesian coordinate system, the positions of the relay user equipment and the remote user equipment can be expressed as and , and calculate the drone fixed at The operation cycle T is equally divided into (N+3) time slots, and the length of each time slot can be expressed as .
[0027] Step 2 builds a multi-dimensional collaborative assisted mobile edge computing system model, which includes: Step 201: construct a channel model; like Figure 2 As shown in the figure, in order to effectively utilize channel resources, a four-stage parallel communication scheme is designed. In the first stage, in the nth (n=(1, 2, ...N-1, N)) time slot, the remote user equipment Offload task data to relay user devices In the second stage, in the (n+1)th time slot, the time slot is divided into two sub-time slots of equal length. In the first sub-time slot, the relay user equipment From the remote user device The data is relayed to the relay drone; in the second sub-time slot, the relay user equipment Relay its own data to the relay drone and use Indicates the length of the sub-time slot. In the third stage, the relay drone will receive the data from the remote user device in the (n+2)th time slot. Data and relay user equipment The data is forwarded to the computing drone. In the fourth stage, the computing drone processes these task data in the (n+3)th time slot; Therefore, the relay UAV starts flying from the (n+1) time slot, and the position of the relay UAV can be expressed as ,in The starting and ending positions of the relay drone can be expressed as and Since the trajectory of the relay drone is initialized as a circle; in this case, the horizontal position of the relay drone should be the same at the initial point and the final point, which can be expressed as In addition, let and They are respectively represented as the speed and acceleration of the relay drone in the (n+1)th time slot, and need to satisfy and .
[0028] The relationship between the position, velocity, and acceleration of the relay drone is as follows:
[0029] In the formula, Indicates speed, represents acceleration, Indicates the length of the time slot; It is assumed that the communication link between the relay user equipment and the relay UAV and the communication link between the relay UAV and the computing UAV are line-of-sight (LoS); line-of-sight LoS refers to the transmission mode when there are no obstacles between the transmitter and the receiver.
[0030] Relay User Equipment The channel gain with the relay drone in the (n+1)th time slot can be expressed as:
[0031] In the formula, Indicates that the user equipment is relayed in the (n+1)th time slot The distance to the relay drone; The channel gain between the relay drone and the computing drone in the (n+2)th time slot can be expressed as:
[0032] In the formula, represents the distance between the relay drone and the computing drone in the (n+2)th time slot, Indicates the relay user equipment at the reference distance d=1m The channel gain between the relay drone and the computing drone (also the channel gain between the relay drone and the computing drone).
[0033] Remote User Equipment Relay User Equipment The channel gain can be expressed as:
[0034] In the formula, Indicates remote user equipment Relay User Equipment The distance is the path loss coefficient, Indicates the remote user equipment at the reference distance d=1m Relay User Equipment The channel gain between .
[0035] Step 202: construct an unloading and calculation model; First, a binary variable is defined Indicates remote user equipment and relay user equipment The communication status in the nth time slot. At the same time, the binary variables are also defined and Relay user equipment The communication status with the relay drone in the (n+1) time slot and the communication status between the relay drone and the computing drone in the (n+2) time slot.
[0036] if or or If it is 0, it is a disconnected state. Binary variable , and The relationship can be expressed as:
[0037] Based on the savory formula, Remote User Equipment Offload task data to relay user devices The transmission rate is:
[0038] In the formula, From the remote user device To relay user equipment The channel bandwidth, Remote user equipment The noise power, is the nth time slot from the remote user equipment To relay user equipment Data transmission power; Relay user equipment in the (n+1)th time slot Remote user equipment The transmission rate of the data and its own data relayed to the relay drone is:
[0039] In the formula, From the relay user equipment The channel bandwidth to the relay drone, It is a relay user device The noise power, and Respectively represent the relay user equipment in the (n+1)th time slot Your own data and remote user devices The transmission power of the data relay to the relay drone; The data transmission rate of the relay drone in the (n+2)th time slot is:
[0040] In the formula, is the channel bandwidth from the relay drone to the computing drone, is the noise power of the relay drone, represents the data transmission power from the relay UAV to the computing UAV in the (n+2)th time slot.
[0041] make The total amount of data offloaded for each remote user device, Indicates remote user equipment The amount of data unloaded in the nth time slot must satisfy the following conditions:
[0042] In addition, the definition The total amount of data unloaded for each relay user device, Relay user equipment The amount of data unloaded in the (n+1)th time slot must satisfy the following conditions:
[0043] make Indicates the relay user equipment in the (n+1)th time slot Relay remote user equipment The amount of data, +2] indicates that the relay drone unloads the remote user equipment in the (n+2)th time slot and relay user equipment The amount of data, +3] means calculating the amount of data to be processed by the drone in the (n+3)th time slot; then the following relationship can be obtained:
[0044] In the above case, the remote user equipment The amount of unloaded data and relay user equipment The amount of data unloaded by the relay drone should meet the following requirements:
[0045] Once the computing drone is assigned computing data, it can start to perform computing operations. Assume that the amount of data that the computing drone can calculate in the (n+3)th time slot is . Then, it should satisfy the following constraints:
[0046] Where C is the computing resources required for the UAV to process 1 bit of data.
[0047] Step 203: construct an energy consumption model of the UAV-assisted mobile edge computing system; The energy consumption of the UAV-assisted mobile edge computing system considered in this invention includes three parts: unloading energy consumption, hovering flight energy consumption and computing energy consumption. Since the ground control base directly supplies energy through the tethered antenna, the hovering energy consumption of the computing UAV is not considered.
[0048] Remote User Equipment Offload data to the relay user equipment in the nth time slot The energy consumption is:
[0049] The (n+1)th time slot relays the user equipment Offloading remote user equipment to relay drones The energy consumption of data and self-data are:
[0050] The energy consumption of the relay drone unloading data to the computing drone in the (n+2)th time slot is:
[0051] The energy consumption of calculating the data of the drone in the (n+3)th time slot is:
[0052] In the formula, is the effective capacitance coefficient of the computing drone affected by the chip structure, To calculate the frequency of the drone in the (n+3)th time slot; The present invention adopts a fixed-wing UAV model, and the flight energy consumption of the relay UAV can be expressed as:
[0053] In the formula, and are the parameters related to the hardware structure of the UAV (including weight, wing area, and air density), and g is the acceleration due to gravity.
[0054] Therefore, the total energy consumption can be expressed as:
[0055] Step 204, based on the above model, the UAV-assisted mobile edge computing system model is formulated as follows; make represents the trajectory of the relay drone, represents the communication schedule, Indicates the calculation frequency of the drone, Indicates the amount of unloaded data allocation, represents the unloading power allocation, so the system optimization model can be constructed as:
[0056]
[0057]
[0058] Wherein, constraint C1 represents the remote user equipment in the nth time slot. and relay user equipment The optimal amount of data to be offloaded between the remote user devices should exceed and relay user equipment The actual amount of data unloaded between Constraint C2 indicates that a remote user equipment can communicate with at most one relay user equipment in each time slot; Constraints C3 and C4 indicate that the relay user equipment is in the (n+1)th time slot. The optimal offload data between the relay drone should exceed the relay user equipment The actual amount of data offloaded to and from the relay drone; Constraint C5 indicates that the optimal amount of data offloaded between the relay UAV and the computing UAV in the (n+2)th time slot should exceed the actual amount of data offloaded between the relay UAV and the computing UAV; Constraint C6 ensures that remote user equipment In the nth time slot, the relay user equipment Offload data, relay user equipment Relay the remote user device to the relay drone in the (n+1)th time slot The relay drone forwards the remote user device to the computing drone in the (n+2)th time slot. and relay user equipment data; Constraint C7 represents the relationship between the amount of computational data and the computational frequency of the computational drone; Constraint C8 indicates remote user equipment The task data must be offloaded to the relay user equipment in the nth time slot , relaying user equipment in the (n+1)th time slot To receive the remote user equipment The data and its own data are relayed to the relay drone, and then forwarded from the relay drone to the computing drone in the (n+2)th time slot, and the computing drone performs the computing operation in the (n+3)th time slot; Constraints C9 and C10 indicate that the total amount of data offloaded by each remote user equipment and the total amount of data offloaded by each relay user equipment should be less than their respective upper limits; Constraint C11 indicates remote user equipment , relay user equipment The transmission power of the relay drone is subject to their respective upper limits. restrictions; Constraint C12 describes that within a flight cycle, the initial position and velocity of the relay UAV are equal to the final position and velocity of the relay UAV; Constraints C13-C14 represent the relationship between the position, velocity, and acceleration of the relay UAV; Constraints C15-C16 indicate that the speed and acceleration of the relay drone are subject to their maximum values and restrictions.
[0059] In step 3, the objective function of the model is solved, including: Since the optimization problem P1 of the above model is non-convex, the present invention designs an iterative data offloading scheduling algorithm to solve the non-convex optimization problem P1. The solution steps are as follows: Figure 3 As shown; Step 301: First, set the iteration termination threshold and , given the relay UAV trajectory Q; the initial non-convex optimization problem P1 is decomposed into two sub-problems of system scheduling and trajectory optimization using the block coordinate descent method; Step 302: Then, under the condition of fixing the trajectory of the relay drone, the Lagrangian dual method is used to solve the system scheduling subproblem until the iteration termination condition is met, and the optimal computing frequency, communication scheduling and data unloading strategy are obtained; the data unloading strategy includes unloading power and unloading data volume; Step 303: Based on the optimized computing frequency, communication scheduling and data offloading strategy, a continuous convex approximation method is used to solve the trajectory optimization subproblem until the iteration termination condition is met to obtain the optimal relay UAV trajectory.
[0060] Step 302: In the first sub-problem, by fixing the trajectory of the relay drone, the communication scheduling, unloading power, unloading data volume and computing frequency are optimized until the sub-problem meets the convergence condition. , stop iteration; get the optimal data offloading strategy, computing frequency and communication scheduling; specifically include: Step 3021, in the case of a fixed UAV trajectory, formulate the system scheduling sub-problem, including: In order to achieve the maximum solution space, The difference between should be as small as possible; similarly, and The accumulation of as well as The difference between them should be as small as possible; therefore, constraint C8 can be transformed into C8*, and C8* is:
[0061] because = = , so the constraints C1, C2, C3, C4 and Can be transformed into , , , and Therefore, the optimization problem P1 can be transformed into P2:
[0062]
[0063] Given the relay drone trajectory Q, the corresponding system scheduling subproblem can be expressed as:
[0064] Step 3022: The system scheduling subproblem P2.1 is a convex problem, and the present invention uses the Lagrange dual method to solve it; specifically, it includes: (1) Given and constrained C , C , C , the Lagrangian dual variables corresponding to C5 and C7 are The value of; therefore, the Lagrangian dual problem of P2.1 can be expressed as:
[0065] in, It can be expressed as:
[0066] (2) Calculate the solution of the system scheduling subproblem based on the Lagrange dual variables and the Karush-Kuhn-Tucker conditions;
[0067]
[0068] The proof of the above solution process is as follows: The Lagrange dual function of problem P2.1 is defined as:
[0069] because about , are convex, so they are middle , Take the derivative and calculate the Lagrangian function In Taking the derivative, we can get:
[0070] Separate order , , , , , we can get the optimal , , , and ; The best , , , Bring in , get the optimal The value of The value pair has a great impact, so through To get the best .
[0071] The above relevant certifications have been completed.
[0072] Since the Lagrangian function about is linear, and = , so the optimal Can be solved by To get:
[0073] Assume that the amount of unloaded data in each time slot can reach the maximum channel capacity. Therefore, we can get:
[0074] In addition, the best It can be obtained based on the following constraints:
[0075] Optimal It can be expressed as:
[0076] Then, the corresponding dual problem can be solved as follows:
[0077] (3) Use subgradient method to update Lagrangian dual variables; Specifically: To solve the dual problem, the subgradient method is used to solve the dual variables , , , and Perform iterative optimization, dual variables , , , and Updated in each iteration to:
[0078] Where t is the number of iterations of the subgradient method, , , Indicates the step length.
[0079] (4) Iteratively solve the system scheduling subproblem until the solution to the system scheduling subproblem satisfies the iteration termination condition; specifically: Repeat steps (2) to (3) until the objective function of the first subproblem meets the convergence condition. , terminate the iteration and obtain the optimal communication scheduling, computing frequency and offloading strategy.
[0080] Step 303: In the second sub-problem, based on the optimized offloading strategy, computing frequency and communication scheduling, the trajectory of the relay UAV is optimized until the convergence condition is met. , get the optimal relay drone trajectory Q; specifically including: Step 3031, the second sub-problem can be expressed as:
[0081]
[0082] Step 3032: Based on the optimized calculation frequency F, communication scheduling , power allocation P and offload data allocation , the continuous convex approximation technique is used to solve the second sub-problem and realize the trajectory optimization of the relay drone. Specifically: (1) Due to the objective function of subproblem P2.2 and constraint C , and C5 are both non-convex, so subproblem P2.2 is a non-convex problem. In order to solve the non-convexity of the objective function, a slack variable is introduced arrive In the above equation, we get its upper bound:
[0083] And the following constraints need to be met:
[0084] In the formula, Relative to and are all convex. Therefore, by Replace with , the objective function of problem P2.2 becomes convex. But is non-convex, so here we use the first-order Taylor expansion to process it. For a given local point , which can be converted into the following form:
[0085] Among them, the equation is Then, It can be approximated as:
[0086] The above inequality is convex because Relative to is linear. In addition, for the non-convex constraint C , C and C5, by using the continuous convex approximation technique, an approximate solution to problem P2.2 can be obtained.
[0087] Specifically: Considering the constraint C and C The left side expression of each and about is non-convex, and about is convex; Therefore, for a randomly given local point , The lower bound function can be expressed as: ; The lower bound function can be expressed as Similarly, for any given local point , the lower bound function of the left part of constraint C5 can be expressed as [n+2]. The corresponding lower bound function is as follows:
[0088] In the formula, , .
[0089] (2) Substitute the derived lower bound function and the approximated convex expression into problem P2.2 to obtain the new optimization problem P2.3:
[0090] The above problem P2.3 is a convex problem. The present invention uses the convex optimization tool CVX to solve the trajectory of the UAV.
[0091] (3) Repeat step 303 to optimize the trajectory of the relay UAV until the objective function of the second sub-problem meets the convergence condition, stop convergence, and obtain the optimal trajectory of the relay UAV.
[0092] Step 304: Input the optimal unloading power, unloading data volume, calculation frequency, communication scheduling obtained in step 302 and the UAV trajectory obtained in step 303 into the objective function P1 of the original optimization problem to obtain ; Step 305, repeat the operations from step 302 to step 304 until The relative reduction is less than , get the minimum equipment energy consumption and achieve energy consumption optimization. Figure 4 As shown, the convergence of energy consumption of the solution of this embodiment and the solution of user equipment without relay under different iteration times is shown. It can be seen from the figure that the solution of this embodiment has better performance advantage and good convergence.
[0093] Embodiment 2 This embodiment discloses an edge computing system optimization system based on user equipment and multiple drones, including: The edge computing system building module is configured to: build a multi-dimensional collaborative auxiliary mobile edge computing system based on ground user equipment and drones; The edge computing system optimization problem solving module is configured to: construct the objective function and constraint conditions of the energy consumption optimization of the multi-dimensional collaborative assisted mobile edge computing system, and solve the objective function, including: setting an iteration termination condition, giving a relay drone trajectory, decomposing the objective function solution problem into a system scheduling sub-problem and a trajectory optimization sub-problem and performing iterative solutions; Under the condition of a fixed relay UAV trajectory, the Lagrangian dual method is used to solve the system scheduling subproblem until the iteration termination condition is met, and the optimal computing frequency, communication scheduling and data unloading strategy are obtained; based on the optimized computing frequency, communication scheduling and data unloading strategy, the continuous convex approximation method is used to solve the trajectory optimization subproblem until the iteration termination condition is met, and the optimal relay UAV trajectory is obtained.
[0094] It should be noted that the above modules correspond to the steps in Embodiment 1, and the examples and application scenarios implemented by the above modules and the corresponding steps are the same, but are not limited to the contents disclosed in the above Embodiment 1. It should be noted that the above modules, as part of the system, can be executed in a computer system such as a set of computer executable instructions.
[0095] Embodiment 3 Embodiment 3 of the present invention provides an electronic device, including a memory and a processor, and computer instructions stored in the memory and running on the processor. When the computer instructions are executed by the processor, the steps of the above-mentioned edge computing system optimization method based on user equipment and multiple drones are completed.
[0096] Embodiment 4 Embodiment 4 of the present invention provides a computer-readable storage medium for storing computer instructions. When the computer instructions are executed by a processor, the steps of the above-mentioned edge computing system optimization method based on user equipment and multiple drones are completed.
[0097] Embodiment 5 Embodiment 5 of the present invention provides a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of the above-mentioned edge computing system optimization method based on user equipment and multiple drones.
[0098] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0099] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0100] These computer program instructions can also be loaded onto a computer or other programmable data processing device, and a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable device provide for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.
[0101] The description of each embodiment in the above embodiments has different emphases. For parts not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.
[0102] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.
Claims
1. A method for optimizing an edge computing system based on user equipment and multiple drones, characterized in that: include: Establish a multi-dimensional collaborative assisted mobile edge computing system based on ground user equipment and drones; Constructing the objective function and constraint conditions of the energy consumption optimization of the multi-dimensional collaborative assisted mobile edge computing system, and solving the objective function, including: setting an iteration termination condition, giving the trajectory of the relay drone, decomposing the problem of solving the objective function into a system scheduling sub-problem and a trajectory optimization sub-problem and solving them iteratively; Under the condition of a fixed relay UAV trajectory, the Lagrangian dual method is used to solve the system scheduling subproblem until the iteration termination condition is met, and the optimal computing frequency, communication scheduling and data unloading strategy are obtained; based on the optimized computing frequency, communication scheduling and data unloading strategy, the continuous convex approximation method is used to solve the trajectory optimization subproblem until the iteration termination condition is met, and the optimal relay UAV trajectory is obtained.
2. The edge computing system optimization method based on user equipment and multiple drones according to claim 1, characterized in that: The ground user equipment includes a remote user equipment and a relay user equipment, and the drone includes a relay drone and a computing drone; The remote user device unloads the task data to the relay user device, and the relay user device relays the task data received from the remote user device and its own task data to the relay drone, and the relay drone forwards the received data to the computing drone for processing.
3. The edge computing system optimization method based on user equipment and multiple drones according to claim 1, characterized in that: The objective function of solving the system scheduling subproblem using the Lagrangian dual method includes: Initialize the Lagrange dual variables corresponding to the constraints; Calculate the solution of the system scheduling subproblem based on the Lagrange dual variables and the Carlo-Kuhn-Tucker conditions; The subgradient method is used to update the Lagrangian dual variables, and the system scheduling subproblem is solved iteratively until the solution of the system scheduling subproblem meets the iteration termination condition.
4. The edge computing system optimization method based on user equipment and multiple drones according to claim 1, characterized in that: The continuous convex approximation method is used to solve the trajectory optimization sub-problem, including: Introduce slack variables into the objective function of the trajectory optimization subproblem to obtain the upper bound function; use the first-order Taylor expansion to process and randomly assign local points to obtain the corresponding lower bound function; The obtained lower bound function and the approximate convex expression are substituted into the objective function of the original trajectory optimization subproblem, and the trajectory of the UAV is solved using the convex optimization tool.
5. The edge computing system optimization method based on user equipment and multiple drones according to claim 1, characterized in that: The energy consumption of the multi-dimensional collaborative assisted mobile edge computing system includes: unloading energy consumption, hovering flight energy consumption and computing energy consumption.
6. The edge computing system optimization method based on user equipment and multiple drones according to claim 1, characterized in that: The data offloading strategy includes: data offloading amount and data offloading power.
7. An edge computing system optimization system based on user equipment and multiple drones, characterized in that: include: The edge computing system building module is configured to: build a multi-dimensional collaborative auxiliary mobile edge computing system based on ground user equipment and drones; The edge computing system optimization problem solving module is configured to: construct the objective function and constraint conditions of the energy consumption optimization of the multi-dimensional collaborative assisted mobile edge computing system, and solve the objective function, including: setting an iteration termination condition, giving a relay drone trajectory, decomposing the objective function solution problem into a system scheduling sub-problem and a trajectory optimization sub-problem and performing iterative solutions; Under the condition of a fixed relay UAV trajectory, the Lagrangian dual method is used to solve the system scheduling subproblem until the iteration termination condition is met, and the optimal computing frequency, communication scheduling and data unloading strategy are obtained; based on the optimized computing frequency, communication scheduling and data unloading strategy, the continuous convex approximation method is used to solve the trajectory optimization subproblem until the iteration termination condition is met, and the optimal relay UAV trajectory is obtained.
8. An electronic device comprising a memory, a processor and a computer program stored in the memory, characterized in that: The processor executes the computer program to implement the steps of an edge computing system optimization method based on user equipment and multiple drones as described in any one of claims 1-6.
9. A computer-readable storage medium having a computer program / instruction stored thereon, characterized in that: When the computer program / instructions are executed by a processor, the steps of an edge computing system optimization method based on user equipment and multiple drones as described in any one of claims 1 to 6 are implemented.
10. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the steps of an edge computing system optimization method based on user equipment and multiple drones as described in any one of claims 1 to 6 are implemented.
Citation Information
Patent Citations
Unmanned aerial vehicle assisted mobile edge computing system optimization method based on user cooperation
CN114500533A
Unmanned aerial vehicle cluster grouping assisted air-ground cooperative computing network resource allocation method
CN116761217A
Unmanned aerial vehicle assisted MEC energy consumption optimization method and system based on node clustering
CN117135690A
Maximum computational bit algorithm in drone-assisted wireless energy transfer mobile edge computing system
CN118781863A
Multi-carrier resource allocation method based on wireless-powered backscatter communication network
US20220407592A1
