A UAV-assisted vehicle edge computing collaborative task offloading method

Abstract

The present invention relates to a method for unloading collaborative tasks of vehicle edge computing assisted by a drone. First, the vehicles are clustered. For the task vehicles in the cluster, the task completion methods include: performing local calculations on some subtasks; utilizing vehicle-to-vehicle direct links to transfer subtasks to other vehicles in the cluster with computing capabilities for processing; unloading subtasks to drones via cluster heads for processing; first unloading to drones via cluster heads, then utilizing drones for relay forwarding, and finally unloading subtasks to base station servers for processing; unloading subtasks directly to edge computing servers via cluster head vehicles. Compared with the prior art, the present invention has the following significant advantages: task vehicles make deployment allocation decisions based on the delay, energy consumption, confidentiality, and penalty items of completing tasks, and select one or more schemes for collaborative task unloading calculations, thereby saving energy, reducing delays, and ensuring safety.

Description

Technical Field

[0001] The present invention relates to the field of vehicle edge computing, and in particular to a method for unloading collaborative tasks in vehicle edge computing assisted by a drone. Background Art

[0002] As an enabling technology for intelligent transportation systems, vehicle edge computing supports resource coordination and sharing between vehicles. The increase in mobile applications has triggered an explosive growth in the demand for high computing power. To support these applications, the idea of ​​computational offloading has emerged, where vehicles can transfer tasks to another node for processing. Mobile edge computing technology is a powerful paradigm for computing task offloading because the computing server is located near the vehicle, such as the roadside unit where the MEC server is located. Using cellular vehicle-to-everything technology, this is one of the key features of the fifth generation network. C-V2X is designed on two air interfaces; the Uu cellular interface is a traditional uplink / downlink interface that supports vehicle-to-network communication, and a sidelink interface for direct communication between nodes. This interface supports vehicle-to-vehicle communication and vehicle-to-infrastructure communication.

[0003] Unmanned aerial vehicles (UAVs), currently experiencing explosive growth and significant regulatory deregulation, have been widely used in civilian applications such as traffic monitoring, public safety, search and rescue, disaster relief, and recovery reconnaissance. UAVs not only provide extended coverage over a wide geographic area but also have unique characteristics such as rapid deployment, easy programming, and high scalability. Various payloads, such as IoT sensors, micro base stations, and embedded computing modules, can be installed on UAVs to achieve different sensing, communication, and computing tasks. In particular, if properly deployed and operated, UAVs can provide reliable and cost-effective wireless communication solutions for a large number of practical scenarios. UAVs can act as wireless relay stations or aerial base stations to improve connectivity and extend the coverage of ground wireless devices, because the height of UAVs enables wireless devices to effectively establish line-of-sight communication links, thereby mitigating potential signal blocking and shadowing. However, existing UAV technology has not been applied to vehicle edge computing, so a UAV-assisted vehicle edge computing collaborative task offloading method is now proposed. Summary of the invention

[0004] The purpose of the present invention is to overcome the shortcomings of the prior art and provide a UAV-assisted vehicle edge computing collaborative task offloading method, which is characterized in that the method specifically includes:

[0005] Cluster the vehicles in the target area;

[0006] For vehicle i that needs to perform computing tasks, its task Z i After being divided into several subtasks, vehicle i and other vehicles, drones, and base stations with idle computing resources in its cluster jointly offload the computing.

[0007] Furthermore, the vehicles are clustered using a greedy iterative algorithm through vehicle-to-vehicle communication, and the cluster head is selected according to the set requirements.

[0008] Furthermore, the joint offloading calculation aims to minimize the weighted sum of total delay and total energy consumption, and constructs the following optimization problem to solve the optimal offloading solution, specifically:

[0009]

[0010]

[0011]

[0012]

[0013]

[0014]

[0015]

[0016]

[0017] Among them, θ i is the penalty term, λ is the delay cost coefficient, μ is the communication and computing energy cost coefficient, T i To complete the task Z i The delay, To complete the task Z i Energy consumption; α i is the proportion of tasks that are calculated locally by vehicle i, β i,j is the ratio of tasks assigned by vehicle i to vehicle j with idle computing resources in its cluster, γ i is the proportion of tasks assigned to the UAV by the cluster head for vehicle i, χ i is the ratio of tasks assigned to the base station by vehicle i through the cluster head and the drone relay, ε i is the proportion of tasks of vehicle i that are directly assigned to the base station for calculation via the cluster head; Completed mission Z i The maximum delay, is the computing resource required by vehicle i to allocate to vehicle j for computing, and F j is the idle computing resource of vehicle j, f i UAV is the computing resource required by the task to be computed locally by vehicle i, and F UAV It is the maximum computing resource that can be allocated to the drone. is the bandwidth required for the A2G downlink transmission of the drone for the task assigned by vehicle i to the drone for relay forwarding, B UAV is the bandwidth of the drone, K i Completed mission Z i The confidentiality rate, Completed mission Z i The required minimum confidentiality rate; N is the set of vehicles that perform computing tasks, and M is the set of other vehicles with idle computing resources in the cluster where vehicle i is located.

[0018] Further,

[0019] in, represents the delay in completing the task computed locally by vehicle i, S i Represents the task Z in bits i The data size, C i is the computational resources required to execute each task, f i loc is the computing resource allocated by vehicle i to complete the task computed locally; represents the unloading delay of the task assigned by vehicle i to vehicle j for calculation, represents the data transmission rate between vehicle i and vehicle j, B i,j is the bandwidth allocated by vehicle i to vehicle j, P i,j is the transmission power between vehicle i and vehicle j, is the transmission channel between vehicle i and vehicle j, I i,j is the interference between vehicle i and vehicle j, σ 2 represents the noise power, is the computing resources allocated by vehicle j to complete the task assigned to vehicle i; is the transmission rate when the task is transmitted to the UAV via the cluster head in the cluster where vehicle i is located, is the transmission rate when the UAV transmits the task to the base station, f i MEC is the computing resource allocated by the base station to complete the task assigned to vehicle i; represents the unloading delay of the task assigned to the base station by vehicle i for calculation, is the transmission rate when vehicle i transmits tasks directly to the base station via the cluster head.

[0020] Further, in, represents the energy consumption of vehicle i, κ i represents the effective capacitance of the CPU architecture of vehicle i; represents the energy consumption of vehicle j, κ jis the effective capacitance of the CPU architecture of vehicle j with the VEC server; represents the energy consumption of the drone, is the energy consumption of the UAV to complete the task assigned by vehicle i, κ UAV is the effective capacitance of the CPU architecture of the drone, f i UAV is the computing resource allocated by the UAV to complete the task assigned by vehicle i; is the G2A uplink transmission energy consumption when transmitting tasks to the UAV via the cluster head in the cluster where vehicle i is located, P i,UAV is the transmission power when vehicle i transmits tasks to the UAV via the cluster head, is the G2A uplink transmission delay when transmitting tasks to the UAV via the cluster head in the cluster where vehicle i is located; is the A2G downlink transmission energy consumption when the UAV transmits tasks to the base station, P UAV,MEC is the transmission power between the drone and the base station; is the energy consumption of the base station, κ MEC is the effective capacitance of the base station's CPU architecture, f i MEC is the computing resource allocated by the base station to complete the task assigned to vehicle i, P i,MEC is the transmission power when vehicle i transmits tasks directly to the base station through the cluster head.

[0021] Further, in, is the confidentiality rate of vehicle j completing the task assigned by vehicle i, R i,e is the rate at which vehicle i transmits data to the eavesdropping vehicle; is the confidentiality rate of A2G downlink transmission when the UAV transmits tasks to the base station. is the confidentiality rate of G2A uplink transmission when the task is transmitted to the UAV via the cluster head in the cluster where vehicle i is located, is the confidentiality rate of the base station completing the task assigned to vehicle i, R e, MEC is the rate at which the eavesdropping vehicle transmits data to the base station.

[0022] Further, k i is the sum of the number of other vehicles, drones, and base stations used to complete computing task i, N i It is to carry out mission Z i The sum of the number of other vehicles, drones, and base stations with idle computing resources for joint offloading calculations.

[0023] Furthermore, after converting the optimization problem into a convex optimization problem through the successive convex approximation SCA algorithm, the convex optimization problem is solved to obtain the allocation ratio of each subtask, the computing resources and bandwidth required to complete each subtask, which is the optimal offloading solution.

[0024] Another object of the present invention is to provide a drone-assisted vehicle edge computing collaborative task offloading device, comprising:

[0025] Memory for storing computer programs;

[0026] A controller is used to implement the vehicle edge computing collaborative task offloading method as described above when executing the computer program.

[0027] Another object of the present invention is to provide a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of vehicle edge computing collaborative task unloading as described above are implemented.

[0028] Compared with the prior art, the present invention has the following significant advantages:

[0029] 1. This method adds vehicle-to-vehicle unloading and drone unloading relay solutions to the edge computing solution, saving energy, reducing latency, and ensuring security. It is suitable for a variety of scenarios, especially when the base station fails or the communication link is blocked;

[0030] 2. This method divides a single task into multiple subtasks, effectively reducing latency and energy consumption, and performing task collaborative offloading;

[0031] 3. This method takes into account the indicators of delay and energy consumption when making decision allocations, and is subject to confidentiality constraints, taking into account the security of the system;

[0032] 4. In order to prevent the use of multiple nodes or devices for small-scale data tasks, this method introduces a special penalty item for the offloading plan formulated for each vehicle node to avoid wasting resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 It is a network architecture diagram of the UAV-assisted vehicle edge computing collaborative task offloading method of the present invention.

[0034] Figure 2 It is a flow chart of the vehicle edge computing collaborative task unloading method assisted by the drone of the present invention, which takes into account delay energy consumption and confidentiality. DETAILED DESCRIPTION

[0035] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0036] Take an urban highway as a scenario, which has multiple lanes, such as Figure 1 As shown in the figure, there are four types of vehicles. Vehicle A is a mission vehicle with tasks to be calculated or unloaded; Vehicle B is a vehicle with a VEC server, which can unload calculations through vehicle-to-vehicle communication; Vehicle C is an eavesdropping vehicle that eavesdrops on information; and Vehicle D is an ordinary vehicle. A fixed-wing drone is set up above the highway to follow the flight. The drone has a relay and unloading module, which are independent of each other and do not affect each other. There is a large base station far away from the highway with sufficient computing resources for unloading calculations.

[0037] When performing vehicle edge computing collaborative task offloading, the vehicles are first clustered through vehicle-to-vehicle communication using a greedy iterative algorithm, and a cluster head is selected in each cluster. The cluster head represents the vehicles in the cluster to communicate with the drone and the base station. Then, for the vehicle that performs the computing task, its task is divided into several subtasks, and the vehicle and other vehicles, drones, and base stations with idle computing resources in its cluster perform joint offloading calculations. Among them, the vehicle uses the vehicle-to-vehicle direct link to transmit the subtask to other vehicles with idle computing resources in its cluster for calculation, uses the cluster head to directly transmit the subtask to the drone and base station for calculation, and uses the cluster head and the drone's relay module to indirectly transmit the subtask to the base station for calculation. It should be noted here that the base station needs to process two parts of subtasks, including drone relay forwarding and direct allocation through the cluster head.

[0038] When selecting cluster heads, some important metric sets are introduced, such as link lifetime, speed, and average distance between communicating vehicles. The vehicle with the smallest weighted value becomes the cluster head.

[0039] like Figure 2 As shown, a flowchart of a method for unloading a vehicle edge computing collaborative task assisted by a drone is provided in an embodiment of the present invention, and the method comprises the following steps:

[0040] Step 201: Cluster the vehicles through vehicle-to-vehicle communication using a greedy iterative algorithm, and then introduce some important metric sets for cluster head selection, such as link lifetime, speed, and average distance between communicating vehicles. The vehicle with the smallest weighted value becomes the cluster head.

[0041] Step 202: The computational task that vehicle i needs to perform is expressed as Where S i (in bits) represents task Z i The data size, C i (in CPU cycles per bit) is the amount of computing resources required to execute each bit, Completed mission Zi The maximum delay. represents the delay of local calculation of vehicle i, where α i is the task allocation ratio of vehicle i for local computing, f i loc is the computing resource allocated to the local server by vehicle i, and κ i represents the effective capacitance depending on the CPU architecture of vehicle i, and therefore the local computing energy consumption of vehicle i It is expressed as:

[0042]

[0043] Vehicle i transmits part of the task to the vehicle with idle computing power and VEC server in the cluster through PC5 interface. represents the data transmission rate between vehicle i and vehicle j when the vehicle i transfers the subtask to the vehicle j with idle computing capacity in the cluster, expressed as:

[0044]

[0045] Among them, B i,j is the bandwidth allocated by vehicle i to vehicle j, P i,j is the transmission power between vehicle i and vehicle j, I i,j is the interference caused by other vehicles, σ 2 Represents the noise power. represents the channel gain between vehicle i and vehicle j, considering the free space path loss model, expressed as

[0046]

[0047] Where a0 represents the received power at a reference distance of 1m when the transmission power is 1w, and d x,y represents the distance between vehicle i and vehicle j, ||·|| represents the Euclidean norm of the vector;

[0048] Therefore, the offloading delay of the task assigned by vehicle i to vehicle j for computation is It consists of task transmission time and task execution time, which can be expressed as:

[0049]

[0050] where β i,j is the ratio of tasks assigned from vehicle i to vehicle j, is the computing resource required by vehicle i to allocate to vehicle j for the task of computing.

[0051] Similarly, the transmission and computing energy consumption between vehicles It is expressed as:

[0052]

[0053] Where P i,j is the power transferred between vehicle i and vehicle j, κ j is the effective capacitance of the CPU architecture of vehicle j with the VEC server, and the confidentiality rate K ij Defined as the data transmission rate from vehicle i to vehicle j The data transmission rate R of vehicle i to the eavesdropping vehicle i,e The difference is:

[0054]

[0055] Vehicle i transfers part of the task to the UAV through the cluster head for offloading calculation. The communication between the cluster head and the UAV adopts the IEEE802.11p protocol with carrier sense multiple access protocol with conflict avoidance for data transmission.

[0056] Latency of offloading computation or relaying via drones Delayed by G2A uplink transmission UAV calculation delay A2G downlink transmission delay Calculate the delay with the base station MEC server Composition, expressed as

[0057]

[0058] Among them, γ i is the proportion of tasks assigned by vehicle i to the drone MEC server for offloading computation, χ i is the proportion of tasks assigned by vehicle i to the drone MEC server for relay forwarding (forwarded to the base station MEC server for offloading calculation), is the transmission rate from the cluster head to the UAV in the computing task of vehicle i, is the transmission rate of vehicle i’s computing task transmitted from the cluster head and the UAV to the MEC server, f i UAV is the computing resource allocated by the UAV to complete the computing task assigned to vehicle i, and f i MEC It is the computing resource allocated by the base station for the computing task assigned to vehicle i.

[0059] Drone energy consumption By calculating energy consumption G2A uplink transmission energy consumption and A2G downlink transmission energy consumption Additive composition, where:

[0060]

[0061]

[0062]

[0063] Among them, P i,UAV is the transmission power sent by vehicle i to the UAV through the cluster head, P UAV,MEC is the transmission power sent by the drone to the base station.

[0064] The confidentiality rates of G2A uplink transmission and A2G downlink transmission are:

[0065]

[0066]

[0067] Among them, R e,UAV To eavesdrop on the data rate transmitted from the vehicle to the drone.

[0068] Vehicle i forwards part of the task to the base station MEC server for offloading calculation (part of it is directly transmitted to the drone by the cluster head, and the other part is first transmitted to the drone via the cluster head and then forwarded to the base station MEC server via the drone relay). The corresponding energy consumption and confidentiality rate K i,MEC It is expressed as:

[0069]

[0070]

[0071]

[0072] Among them, f i MEC It is the computing resource allocated by the base station MEC server to complete the vehicle's computing tasks. is the transmission rate from vehicle i to the base station MEC server, R e,MEC is the transmission data rate from the eavesdropping vehicle to the MEC server, ε i is the proportion of tasks assigned by vehicle i to the base station MEC server for offloading calculation, P i,MEC is the transmission power sent by vehicle i directly to the base station through the cluster head.

[0073] Step 203: Complete Task Z i Required delay T i , Energy consumption and confidentiality rate C i Expressed as:

[0074]

[0075]

[0076]

[0077] To prevent the use of multiple nodes or devices (nodes here refer to vehicles, and devices generally refer to drones and base stations) for small data tasks, a special penalty term θ is introduced for the offloading method formulated for each vehicle, which is defined as:

[0078]

[0079] where k i Is to complete the calculation task Z i The sum of the number of nodes and devices used when N i is for computing task Z i The sum of the number of other vehicles, drones, and base stations with idle computing resources when formulating a joint offloading plan. This penalty item can make the tasks with small computing data volume and not urgent time be calculated locally as much as possible or use a small number of other nodes / devices to calculate, avoiding resource waste.

[0080] In summary, the problem becomes a joint optimization of communication resource allocation, the goal is to minimize the completion of all tasks Z i The weighted sum of delay and energy consumption can be expressed as

[0081]

[0082]

[0083]

[0084]

[0085]

[0086]

[0087]

[0088]

[0089] Among them, θ i is the penalty term, λ is the delay cost coefficient, μ is the communication and computing energy cost coefficient, T i To complete the task Z i The delay, To complete the task Z i Energy consumption; αi is the proportion of tasks calculated locally by vehicle i, β i,j is the ratio of tasks assigned by vehicle i to vehicle j with idle computing resources in its cluster, γ i is the proportion of tasks assigned to the UAV by the cluster head for vehicle i, χ i is the ratio of tasks assigned to the base station by vehicle i through the cluster head and the drone relay, ε i is the proportion of tasks of vehicle i that are directly assigned to the base station for calculation via the cluster head; Completed mission Z i The maximum delay, is the computing resource required by vehicle i to allocate to vehicle j for computing, and F j is the idle computing resource of vehicle j, f i UAV is the computing resource required by the task to be computed locally by vehicle i, and F UAV It is the maximum computing resource that can be allocated to the drone. is the bandwidth required for the A2G downlink transmission of the drone for the task assigned by vehicle i to the drone for relay forwarding, B UAV is the bandwidth of the drone, K i Completed mission Z i The confidentiality rate, Completed mission Z i The minimum confidentiality rate required; N is the set of vehicles that perform computing tasks, and M is the set of other vehicles with idle computing resources in the cluster where vehicle i is located. Constraint C1 ensures that the task is fully assigned, constraint C2 is the task partition variable constraint, constraint C3 ensures that the completion delay does not exceed the maximum delay, constraints C4 and C5 ensure that the computing resources of vehicles and drones do not exceed the upper limit, constraint C6 ensures that the allocated bandwidth of vehicles and drones does not exceed the upper limit, and constraint C7 ensures that the confidentiality rate of each task is not lower than the lower limit.

[0090] The above optimization problem can be solved by an SCA (successive convex approximation)-based algorithm, which transforms the non-convex objective function and constraints into a suitable approximate convex one, solves the resulting optimization problem in an iterative manner, and obtains the optimal unloading solution.

[0091] The present invention also provides a drone-assisted vehicle edge computing collaborative task unloading device, which includes a memory and a controller; wherein a computer program is stored on the memory; when the controller executes the computer program, the above-mentioned vehicle edge computing collaborative task unloading method is implemented.

[0092] The present invention also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the above-mentioned vehicle edge computing collaborative task unloading method are implemented. The computer-readable 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.

[0093] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0094] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A UAV-assisted vehicle edge computing collaborative task offloading method, characterized in that: The method specifically includes: Cluster the vehicles in the target area; For vehicle i that needs to perform computing tasks, its task Z i After being divided into several subtasks, vehicle i and other vehicles, drones, and base stations with idle computing resources in its cluster jointly offload the computing; The joint offloading calculation aims to minimize the weighted sum of total delay and total energy consumption, and constructs the following optimization problem to solve the optimal offloading solution, which is specifically: Among them, θ i is the penalty term, λ is the delay cost coefficient, μ is the communication and computing energy cost coefficient, T i To complete the task Z i The delay, To complete the task Z i Energy consumption; α i is the proportion of tasks that are calculated locally by vehicle i, β i,j is the ratio of tasks assigned by vehicle i to vehicle j with idle computing resources in its cluster, γ i is the proportion of tasks assigned to the UAV by the cluster head for calculation by vehicle i, χi is the proportion of tasks assigned to the base station by vehicle i firstly through the cluster head and then through the UAV relay, and εi is the proportion of tasks assigned to the base station by vehicle i directly through the cluster head for calculation; Completed mission Z i The maximum delay, is the computing resource required by vehicle i to allocate to vehicle j for computing, and F j is the idle computing resource of vehicle j, is the computing resource required by the task to be computed locally by vehicle i, and F UAV It is the maximum computing resource that can be allocated to the drone. is the bandwidth required for the A2G downlink transmission of the drone for the task assigned by vehicle i to the drone for relay forwarding, B UAV is the bandwidth of the drone, K i Completed mission Z i The confidentiality rate, Completed mission Z i The required minimum confidentiality rate; N is the set of vehicles that perform computing tasks, and M is the set of other vehicles with idle computing resources in the cluster where vehicle i is located; in, represents the delay in completing the task computed locally by vehicle i, S i Represents the task Z in bits i The data size, C i is the computing resources required to perform each task, is the computing resource allocated by vehicle i to complete the task computed locally; represents the unloading delay of the task assigned by vehicle i to vehicle j for calculation, represents the data transmission rate between vehicle i and vehicle j, B i,j is the bandwidth allocated by vehicle i to vehicle j, P i,j is the transmission power between vehicle i and vehicle j, is the transmission channel between vehicle i and vehicle j, I i,j is the interference between vehicle i and vehicle j, σ 2 represents the noise power, is the computing resources allocated by vehicle j to complete the task assigned to vehicle i; is the transmission rate when the task is transmitted to the UAV via the cluster head in the cluster where vehicle i is located, is the transmission rate when the UAV transmits tasks to the base station. is the computing resource allocated by the base station to complete the task assigned to vehicle i; represents the unloading delay of the task assigned to the base station by vehicle i, is the transmission rate when vehicle i transmits tasks directly to the base station via the cluster head; in, represents the energy consumption of vehicle i, κ i represents the effective capacitance of the CPU architecture of vehicle i; represents the energy consumption of vehicle j, κ j is the effective capacitance of the CPU architecture of vehicle j with the VEC server; represents the energy consumption of the drone, is the energy consumption of the UAV to complete the task assigned by vehicle i, κ UAV It is the effective capacitance of the CPU architecture of the drone. is the computing resource allocated by the UAV to complete the task assigned by vehicle i; is the G2A uplink transmission energy consumption when transmitting tasks to the UAV via the cluster head in the cluster where vehicle i is located, P i,UAV is the transmission power when vehicle i transmits tasks to the UAV via the cluster head, is the G2A uplink transmission delay when transmitting tasks to the UAV via the cluster head in the cluster where vehicle i is located; is the A2G downlink transmission energy consumption when the UAV transmits tasks to the base station, P UAV,MEC is the transmission power between the drone and the base station; is the energy consumption of the base station, κ MEC is the effective capacitance of the base station's CPU architecture, f i MEC is the computing resource allocated by the base station to complete the task assigned to vehicle i, P i,MEC is the transmission power when vehicle i transmits tasks directly to the base station through the cluster head; in, is the confidentiality rate of vehicle j completing the task assigned by vehicle i, R i,e is the rate at which vehicle i transmits data to the eavesdropping vehicle; is the confidentiality rate of A2G downlink transmission when the UAV transmits tasks to the base station. is the confidentiality rate of G2A uplink transmission when the task is transmitted to the UAV via the cluster head in the cluster where vehicle i is located, is the confidentiality rate of the base station completing the task assigned to vehicle i, R e,MEC is the rate at which the eavesdropping vehicle transmits data to the base station; k i is the sum of the number of other vehicles, drones, and base stations used to complete computing task i, N i To complete the task Z i The sum of the number of other vehicles, drones, and base stations with idle computing resources when formulating a joint offloading plan.

2. The method for unloading collaborative tasks of vehicle edge computing assisted by a drone according to claim 1, characterized in that: The vehicles are clustered using a greedy iterative algorithm through vehicle-to-vehicle communication, and the cluster head is selected according to the set requirements.

3. The method for unloading collaborative tasks of vehicle edge computing assisted by a drone according to claim 1, characterized in that: After converting the optimization problem into a convex optimization problem through the successive convex approximation SCA algorithm, the convex optimization problem is solved to obtain the allocation ratio of each subtask, the computing resources and bandwidth required to complete each subtask, which is the optimal offloading solution.

4. A drone-assisted vehicle edge computing collaborative task offloading device, characterized in that: include: Memory for storing computer programs; A controller, used to implement the vehicle edge computing collaborative task offloading method as described in any one of claims 1 to 3 when executing the computer program.

5. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of unloading the vehicle edge computing collaborative task as described in any one of claims 1 to 3.

Images